| Optional Modules | Capabilities |
| Control | Generationoflinearisedturbinemodels. |
| Seismic | Earthquake simulation. |
| Hardwaretest | Test a real turbine controller by interfacing it to a Bladed simulation running in real time. |
| Offshore support structure | Multiple member tower and support structure. |
| AdvancedTransmission | Allows user-defined gearbox/transmission dynamics to be |
| Interface AdvancedPitchactuator | integrated intoBladedsimulationsthroughaDLLinterface. |
| Interface | Allowsuser-definedpitchactuatordynamicstobeintegrated into Bladed simulations through a DLL interface. |
| Electrical dynamics | Electrical dynamics models of generators and power |
| converters;networkvoltageandfrequencytransients. |
| WindFarmerlink | Site-specificloadcalculationsusingWindFarmeroutput. |
# 1.3 The Bladed Educational version
Bladed Educational is available solely for academic and teaching purposes. Only the Base Module is available, and there are a number of further restrictions, as follows:
# Turbine model:
The number of blade radial stations is limited to 10 maximum;
The number of tower stations is limited to 5 maximum;
The pitch actuator model is fixed to a first order lag response with 0.3s time constant. No other pitch actuator models are allowed.
The external controller facility is not available.
The use of encrypted data is not supported.
# Environment:
Simulation periods are limited to 60s maximum.
The "seed" used to initialise the random number generators used to synthesise wind turbulence and wave time series is fixed.
Only the longitudinal component of wind turbulence can be used in simulations. (The other components can be synthesised, but will be ignored in simulations.)
Although the Seismic module is not available, a single seismic calculation is available for demonstration purposes.
# Calculations and post-processing:
Calculations cannot be run in Batch mode.
Model linearisation is not available, except as a demo calculation.
The following post-processing calculations are not available, except as demo calculations: ultimate loads ultimate load cases linear model
The Extreme Load Extrapolation calculation is not available.
# 1.4 The Bladed Demonstration version
A demonstration version of Bladed is available.
A demo project, demo_a.prj, is supplied with Bladed and will be found in the installation folder. This contains simplified details of a representative but artificial 3-bladed 2MW offshore variable speed pitch regulated turbine. This can be used to run most of the available calculations.
The demonstration version is restricted as follows:
Operation Modal analysis calculation Turbulent wind generation
Restrictions Turbine details may not be changed Turbulence details may not be changed
Steady calculations and simulations Turbine details may not be changed; other calculation details may be changed
Post processing calculations Calculation details and input data may not be changed. The Extreme Load Extrapolation calculation is not available.
Batch system Calculations cannot be run in Batch mode.
Saving project files Not allowed
Writing reports Not allowed, but a sample report may be viewed.
Encrypted data Not supported
Graphics: viewing calculation Unrestricted
results
When running a calculation, a message box warns that the appropriate demo data will be used for the calculation.
Other project files are also supplied for demonstration purposes. Although these can be viewed, they cannot be used for running calculations.
# 1.5 Support
Bladed is supplied with a one-year maintenance and support agreement, which can be renewed for further periods. This support includes a ‘hot-line’ help service by telephone, fax or e-mail:
Telephone: $+44$ (0)117 972 9900
User portal: https://renewableenergysoftwareportal.dnvgl.com/
E-mail bladed@dnvgl.com
It is DNV GL policy to work with clients to respond to their needs so that the software can be constantly improved. As with any software of this complexity, a total absence of bugs cannot be guaranteed, and any reports of bugs, along with comments or suggestions on any aspect of the software, are welcomed. The maintenance and support agreement includes free provision of any revisions or upgrades of the software during the period of the agreement.
Modifications to the software to meet the individual needs of specific clients can be made by arrangement. Such work will be charged for at commercial rates.
# 1.6 Documentation
Much of the information contained in this manual is also available in the on-line help facility.
There is also a Bladed Theory Manual which explains in more detail the theory behind the calculation methods used.
Section 2 of this User Manual gives an overview of how Bladed can be used. The later sections then provide more detail on each function. Sections $\supseteq$ to $\AA^{\underline{{5}}}$ explain in detail how to set up a model of a turbine. Section $\underline{{6}}$ covers the specification of the wind field, and the sea state for offshore turbines. Section 7 then explains how to set up and run wind turbine calculations. The post-processing, graphics and reporting facilities are described in Sections 8, 9 and 10 respectively.
# 1.7 Acknowledgements
Bladed was developed with assistance from the Commission of the European Communities under the JOULE II programme, project no. JOU2-CT92-0198.
# 2 USING BLADED
This chapter gives an overview of the ways in which Bladed can be used. It covers the following topics:
General description and layout of the user interface
• Entering data
• Using project files
• Performing calculations
• Viewing results Compiling reports
More detailed descriptions follow in later chapters.
# 2.1 General description and layout of the user interface
On starting Bladed, the main Toolbar appears together with the Calculations screen. The Toolbar consists of a set of graphical icons and a number of pull-down menus. The Calculations screen allows the user to select, define and execute a particular calculation.
There are various ways of opening further screens which allow the user to define the characteristics of the various parts of the turbine, as well as the characteristics of the wind and various parameters which control the execution of calculations. There is also a graphics facility for viewing results.
# 2.1.1 Main toolbar - pull-down menus
The pull-down menus may be used as follows:
File: use this menu to create, open and save project files, and to import modules from other project files. A project file (.prj) contains wind turbine information and/or parameters defining calculations. Using the file type selector on the File Open dialogue box, it is also possible to open a project backup file (.prx), or a file containing all the details relating to a calculation previously carried out $(.\Phi\mathsf{p j})$ . This is useful for re-running the calculation with or without modifications.
Use the Import facility to import individual modules from other project or calculation details files into the currently active project.
The project file header information may be entered or edited using Project Info. Use Protect project to enter a security password to prevent the file being modified by unauthorised users. The Encrypt facility allows any part of the turbine model to be encrypted with a password, so that the data can still be used for calculations but is completely invisible to the user. If necessary, different parts of the turbine can be encrypted by different users, each with a separate password. To make data visible again, use Decrypt and enter the correct password.
Specify: this menu allows the user to move directly to a particular screen for specifying any part of the turbine, or any calculation or load case. These screens are also accessible via the Toolbar Icons described below.
Calculation: this menu allows particular calculations to be carried out. See also the Calculation icon on the main toolbar described below. It also allows unwanted calculation results to be deleted.
Batch: this is used to control batch processing (see 7.2.7) of multiple runs. It also gives access to the WindFarmer Link screen, for which a WindFarmer Link licence is required.
Reports: This menu also offers the possibility to write a project report or a calculation report, to append graphs to a report, and to edit or print existing reports. It also gives a choice of report format, which may be ASCII or Microsoft Word.
Tools: Copy results allows calculation results to be copied from one location to another. In doing so, the results can be converted between binary and ascii formats if required. There is also a facility to Delete results. Compare Projects allows two project and/or calculation details files to be compared, and can generate a detailed report of the differences. Create Header Files launches a tool to create header files for multiple ASCII data files so that Bladed post processing and data view can be used.
There is also a facility for configuring the current printer, and to specify user preferences for certain option settings (see 2.8).
Windows: switches between any of the currently open windows, and gives access to the turbine summary information window and to the 3D graphical display (which can be animated using the doublearrow icon).
Help: activates the on-line Help facility, which contains detailed information on Bladed and how to use it. It also gives on-line access to the Bladed User and Theory manuals, and to a facility for upgrading the dongle or security device by entering an appropriate password.
# 2.1.2 Toolbar icons
Beneath the pull-down menus on the toolbar are the Toolbar Icons. Clicking on any one of these opens up the corresponding screen, as follows:
• Blades (see 3) to define the blade properties. Aerofoil sections (see 3.8) to access a database of aerofoil section data.
。 Rotor (see 4.1) to define the properties of the rotor and overall turbine configuration. Tower (see 4.5) to define the tower properties. Power Train (see 4.7) to define the drive train, generator, energy losses and electrical network.
• Nacelle (see 4.13) to define the nacelle geometry and mass.
• Control (see 5) to define both power production and supervisory control systems.
• The wind input (see 6.1) to define the wind speed and direction including spatial and temporal variations. Sea state to define the waves (see 6.12) currents (see 6.13) and tide height (see 6.14) for offshore turbines. Flexibility Modeller (see 7.1) to specify the modal analysis of blade and tower vibrations. Calculations (see 7.2) to select, specify and execute any particular calculation. Data View (see 9) to view graphs or generate tables of results. Analyse (see 8) to specify post-processing of results.
# 2.1.3 The calculation window
The Calculations window lists all the available calculations, any one of which may be selected by clicking on it with the mouse. Next to each calculation is an indicator light. If this is green, the calculation may be performed. A red light indicates that no data is available to perform the selected calculation, while a yellow light indicates that some of the required data has been defined but not all.
Below the calculations is a small window showing all the data modules which need to be defined in order for the selected calculation to be performed. Double-clicking on one of these data modules immediately opens up the relevant screen in which the data for that module is defined. It is therefore a simple matter, having selected the calculation to be performed, to work through all required data modules which are still undefined and assign the relevant data.
# 2.1.4 Sequence of operations
Before any turbine calculations can be done, it is necessary to specify the principal characteristics of the blades and rotor. Open the relevant screens by one of the means described above, i.e. from the Specify pull-down menu, from the relevant Toolbar icon, or from the Calculations screen with one of the turbine calculations selected.
In order to be able to define the aerodynamic characteristics of the blades, it is necessary to enter or import the relevant aerofoil datasets into the aerofoils database if it is not already present, using the Aerofoil icon on the toolbar, or from the Specify pull-down menu.
It is also necessary to define the following modules:
Physical constants (although standard values will have been defined by default)
Aerodynamics control
Calculation definition depending on which calculation is selected (further parameters specific to the
selected calculation)
These are all on the Calculation Parameters screen, accessible from the Calculations screen or from the Specify pull-down menu on the toolbar.
Having defined this fundamental data, it should then be possible to perform the following steady state turbine calculations at a fixed rotational speed and pitch angle defined on the Calculation definition screen specific to each calculation:
· Aerodynamic information (to examine the aerodynamics at each blade station, such as lift and drag, inflow, tip loss etc. at a specified wind speed)
• Performance coefficients (dimensionless power, torque and thrust coefficients as a function of tip speed ratio)
• Power curve (at specified fixed rotor speed and pitch angle).
These calculations are quick and the results can instantly be examined graphically using the Data View icon on the toolbar. From the power curve, the annual energy yield can also be calculated. This is one of the post-processing calculations, and again the results can be seen using Data View.
The Outputs button on the Calculations screen allows the user to specify in some detail which loads and other outputs are desired. Once this is done, the Steady-state operational and parked loads calculations can also be carried out (at specified fixed rotor speed and pitch angle), and Data View used to see the results.
At the preliminary design stage, various turbine parameters can be adjusted and all these calculations can be repeated rapidly until satisfactory results are obtained.
As further data is defined, more complex calculations can be carried out. For example, by defining suitable combinations of drive train, generator and control characteristics, it becomes possible to calculate the steady power curve and steady operational loads with the rotor speed and/or pitch angle varying with wind speed as appropriate.
After defining the numbers of vibrational modes on the Flexibility Modeller screen, along with the necessary mass and stiffness information as indicated on the Calculations screen, it becomes possible to perform a Modal analysis calculation. Blade and tower bending can then be taken into account in the calculations.
Simulations can also be performed once the Simulation control and Time varying windseastate modules are defined. As always, the minimum data required is always indicated on the Calculations screen, making it easy to ensure that the necessary data is defined for each calculation to be done.
Note also the Show options button at the bottom of the Calculations screen. This provides a rapid means of switching particular features off or on for a particular calculation. For example, provided the energy losses module has been defined, the energy losses can be switched off or on very easily using this feature, for all calculations where energy losses are actually relevant.
# 2.2 Entering data
In each window, data is entered by typing in the required information in the fields provided. Where there is a choice of alternative options, selection buttons or pull-down selectors are provided. In some cases check boxes are provided for enabling or disabling particular items (the item is selected when a tick appears in the box).
Where numerical values are entered, except in the case of dimensionless numbers, the units in which the data is to be specified are shown. Double-clicking on the entry brings up a units conversion box, allowing the user to enter values in a choice of different units.
Once data in a window has been edited, it is necessary to assign the changes by clicking the OK button. Alternatively, any changes made since the data was last assigned may be reversed by clicking the Cancel button.
# 2.3 Using project files
A project file stores all the currently defined information relating to the turbine, the wind field information, and calculation parameters. Click OK on any open windows to ensure that the data is assigned before saving to a project file. Project files have a.prj extension.
When a calculation is performed, all the information which is of relevance to that calculation is stored in a special calculation project file which is stored with the calculation results. This file will have a. $\boldsymbol{\mathfrak{S}}$ pj extension, and can be loaded just like a standard project file.
A project report may be generated from the Reports pull-down menu on the main toolbar.
# 2.4 Performing calculations
Calculations may be initiated either from the Calculations pull-down menu on the toolbar, or from the Calculations window by clicking the Run Now button. For some of the steady state calculations and all the simulations, the Calculations window also gives access to facilities for switching various calculation options on or off, and for specifying the calculation outputs required. Post-processing calculations may also be initiated using the Execute button in the post-processing window obtained by clicking the Analyse icon on the toolbar.
Long calculations may be stacked up using the Run in Batch button in the Calculations window. The whole batch of calculations may then be started so that they run one after the other, for example overnight. This is done using the Batch menu item on the toolbar.
Except for some auxiliary calculations, the user will be asked to specify where the calculation outputs are to be sent. A dialogue box allows the drive and directory to be selected, and a run name must also be specified. This is used to identify all the output files produced by that calculation. A new output directory may be created by editing the directory name.
While the calculation is running, a window appears which displays a progress bar showing how far the calculation has progressed, together with any warnings which may arise during the calculation. An ‘Abort’ button allows the user to stop a calculation which is in progress. Any part-written output files which have already been produced should be available. This means that a long simulation can be stopped without losing the results which have so far been produced.
# 2.5 Viewing results
After a calculation has been completed, the results can be viewed by clicking the Data View icon on the toolbar, and selecting the desired data for each channel. Up to six channels may be plotted on one graph. The outputs from any calculation just completed will normally be selected by default. Otherwise, the dialogue box allows any drive and directory to be selected. If the results of several calculations are available within the directory, a pull-down selector allows the user to choose the run name which was specified when the calculation was initiated.
The output files available from that calculation are shown in the Data Group window. Selecting one of these displays a contents list for the file from which the user chooses the variable of interest. Click OK to assign the data to the selected graph channel.
Often there is just one independent variable, for example Time in the case of dynamic simulations. Sometimes there is a choice of independent variables: for example a blade bending moment from a simulation could be plotted either against time or radius. Double-click to change the choice of independent variable.
If two independent variables are defined, a three-dimensional graph will result. Three-dimensional plots are most useful for small datasets.
The View Messages button displays any warning or error messages generated during the run. In a few cases, some additional information is available by clicking the Further Info button. Other buttons allow whole runs or individual data groups to be deleted.
# 2.6 Compiling reports
If Microsoft Word is installed, it is possible to generate neatly formatted project and calculation reports in Word format. It is also possible to insert calculation results into these reports, either as tables or as graphs. See also User Preferences (see 2.8).
If Word is not available, change the report format to ASCII using the Reports pull-down menu. This will cause reports to be generated in tab-delimited ASCII format, suitable for reading into other packages if required. This option does not allow graphs to be inserted into the report.
Reports are generated from the Reports pull-down menu on the main toolbar. There are two types of reports:
Project reports: these may contain any currently assigned modules defining the turbine itself, as well as calculation parameters and wind field details if required.
Calculation reports: these may be generated for any calculation which has been carried out, or which is awaiting execution in the Batch queue. A dialogue box allows the desired calculation to be selected Calculation reports may include any turbine details which are relevant to the calculation, as well as details of the calculation itself and the wind field used if applicable. They may be generated as stand-alone reports, or appended to another report such as a project report.
When reports are generated, they may be appended to existing report files if desired. Thus it is possible to create a project report, and subsequently append calculation reports to it whenever calculations are done, and also append calculation results either as graphs or tables.
# 2.7 Data Encryption
Any part of the turbine data may be encrypted, using a password entered by the user. Encrypted data becomes invisible to the user, but can still be used to run calculations. It can only be decrypted by entering the correct password. The encrypt data, either click the Encrypt button on the appropriate data entry screen, or use the File … Encrypt pull-down menu on the toolbar and tick the modules you wish to encrypt. You will first be prompted for an encryption group name, which identifies the encrypted data. Each encryption group has its own password, so it is possible for different users, perhaps different component suppliers, to encrypt their own data. When a new encryption group is created, the user is prompted to enter a password, and then to verify it. Passwords are case-sensitive. If an existing group is selected, the user must be able to enter the correct password in order to add more data to the group.
Bladed then offers the user the possibility to specify to disable certain simulation outputs, in case this output data is thought to give too much away. For example a blade manufacturer may choose to encrypt the blade data, and might decide to prevent subsequent calculation results from including the blade loads or deflections, or to allow blade loads only the root station for example, or deflections only at the tip. Of course it is important not to restrict the calculation outputs so far that the calculations are no longer useful. Encryption of the blade data automatically causes the rotor mode shapes to be encrypted, and encryption of tower data causes the tower mode shapes to be encrypted. Modal frequencies remain visible to the user.
To decrypt data, either click the Decrypt button on the appropriate data entry screen, or use the File … Decrypt pull-down menu on the toolbar and choose which encryption group to decrypt. The correct password then needs to be entered.
# 2.8 User preferences
Select Tools and Preferences from the main toolbar menu, and select desired options as follows:
# 2.8.1 Standard tab
Make backups of project files: to make a backup file (.prx) whenever a project file (.prj) is opened. Warn when starting calculation: to display a warning when a shelled calculation is about to start. Use binary format for calculation results: binary output is faster and results in smaller files. Close batch client on exit: On exit of Bladed: Close Batch Viewer, and unless this machine is the manager of the batch, also close the Batch Framework, resulting in aborting running jobs and removing this machine as a runner of the batch.
Report format: Select ASCII or WORD.
Insert graphs as links: to append graphs to WORD reports as links to external metafiles. If this option is not selected, graphs will be fully imported into the WORD document.
Dongle search order combo: specifies whether to use a local dongle, network dongle or favour a local dongle over a network dongle (in the latter case if a local dongle is not found, Bladed will search for a network dongle). If you have a network dongle you can change the way in which the network is searched by clicking the Net dongle settings tab.
# 2.8.2 Net dongle settings tab
Connect to any available host: Allows the licence manager on the local machine to perform a broadcast search over the entire network for available network hosts. This setting is not recommended as it significantly increases the amount of network traffic and may result in the incorrect licence being reserved if multiple network dongles are found.
Connect to specified host: Restricts connection to a network dongle hosted on the named IP. This is the recommended option.
Advanced Configuration: Allows advanced settings to be entered for HASP4 and Sentinel HASP network licences using the vendor’s configuration tools. However, the default settings are usually fine.
If Bladed detects that multiple hosts have been defined in Sentinel Admin Control Center (ACC), the option to retain existing settings becomes available. Configuration of multiple hosts should be performed using ACC.
If the option to connect to a specified host is selected with multiple hosts already defined, Bladed will prompt the user whether to overwrite the existing configuration.
# 2.9 Context-sensitive help
For context-sensitive help, press the F1 key at any time.
# 2.10 Dongles
Full Bladed licences are activated using a security device or dongle. Only demo use and Educational licences do not require a dongle. Two types of dongle are used:
# Stand-alone dongle
This is plugged into the USB port of the computer on which Bladed is installed and run. The Sentinel HASP device driver must be installed on the computer. This dongle contains a single licence for the Bladed base module and any optional modules which have been purchased.
# Network dongle
This is plugged into the USB port of a computer on the network, for example a network server (this computer is referred to here as the ‘server’). Both the Sentinel HASP run-time environment and the HASP4 Licence Manager must be installed and running on the server, and are available to install from within the Bladed installer. The HASP4 Licence Manager enables Bladed versions pre-v.3.85 to be run using the same network dongle, and is also required for running Bladed v3.85 and later. The Sentinel HASP Licence Manager is automatically installed as part of the Sentinel HASP run-time environment installation.
Note: When installing the HASP4 Licence Manager, choose the option to install as a service.
The network dongle may contain more than one licence for the Bladed base module and for any optional modules which have been purchased. The number of licences determines the maximum number of users who can simultaneously use each module. The licences are accessible to users on other computers on the network, here referred to as ‘client’ computers. The client computers must have Bladed and the Sentinel HASP run-time environment installed, and do not need to install the HASP4 Licence Manager. To connect to the server only, the option “Use network dongle” should be selected from the combo at the foot of the Standard tab of the Preferences dialog, available from the Tools … Preferences menu on the Bladed toolbar (see 2.8.1). The server IP address should be entered in the Net dongle settings tab (see 2.8.2).
# Local use of network dongle
The network dongle may also be used as a stand-alone dongle on the licence server. Licences are checked out in the same way as a remote network dongle. The option “Use local dongle” should be selected in the Standard tab of the Preferences dialog (see 2.8.1).
Bladed versions pre-v.3.85 must not be started on the server if remote clients wish to use network licences as this would disable the network dongle.
Note: Before removing the network dongle from the server, all licences should be freed and the HASP4 service stopped as described below.
# Remote updating of dongles
Remote update codes may be issued to you from time to time by Garrad Hassan. To apply these, select Help... Security device upgrade from the Toolbar, paste the codes into the box provided, and click the Upgrade button.
For network dongles, it is important to free all HASP4 licences before applying remote update codes by closing all Bladed instances running on client computers and stopping the HASP4 Licence Manager (service name “HASP Loader”). The service should be restarted after the update has been applied. There is no need to stop the Sentinel HASP Licence Manager service when applying a remote update.
After successfully applying the dongle update, Bladed will generate a C2V file which contains all the licensing data on the dongle. Please attach this file to an email and send to bladed@dnvgl.com.
# Company code
From Bladed version 4.1 a unique company code is stored on the dongle. This allows third party suppliers of Bladed project files to authenticate a request for a project file against the dongle. To see the company code, select the Help About menu item.
# 2.11 Component download
The component downloader allows users to download Bladed project files containing components from third party suppliers with whom they have an agreement. To launch the component downloader select the Tools “Launch component downloader” menu item. The component downloader can also be accessed from some of the other screens. A list of providers can be seen in the drop down list. Select the provider required, enter your username and password as previously agreed with the provider and select Fetch components. This will then produce a list of available components. Type a download path into the download location box and select Download next to the required component.
# 3 DEFINING THE TURBINE BLADES
This section describes how to construct a model of the turbine blades, including geometrical, mass and stiffness properties as well as the aerodynamic characteristics of the aerofoil sections.
Click the Blades icon on the toolbar to open the Blade Properties screen.
Before any calculations can be carried out, the user must define the geometry (see 3.2) and aerodynamic characteristics (see 3.8) of the rotor blades on the Blade Geometry tab.
If any dynamic calculations are required, check the Mass checkbox on the Mass and Stiffness tab to allow the blade mass distribution (see 3.3) to be defined.
If blade flexibility (see 7.1) is to be modelled, also check the Stiffness checkbox to allow the blade stiffness distribution (see 3.5) to be entered. An axial degree of freedom, a torsional degree of freedom and shear stiffness can also be defined by checking the appropriate checkboxes.
Holding the mouse over any of the blade property labels will bring up additional tool tip text which describes what the property is in more detail.
The blade properties are defined at a number of stations (see $\underline{{3.1}}$ ) along the blade, although additional point masses (see 3.3.2) may also be defined anywhere on the blade by clicking the Add button on the Additional Mass tab.
The Graph menu controls the graphical display of blade properties. The planform graph and the cross section plot are both updated as soon as any new values are assigned, giving an instant indication if faulty values are entered. Controls are provided to change the format of stiffness graphs, and to print the graph or save it to a metafile, which can subsequently be incorporated into a report.
Click the Add button on the Blade Geometry tab to add a new blade station. Stations are automatically sorted by radial position. To highlight a station, click on the station number, or doubleclick on a graph point. Click Delete to remove a highlighted station.
To enter or edit data, highlight the data item by clicking on it, or by moving to it using the arrow keys and then pressing the Return key. Press Escape to restore the previous value.
Select a block of cells by dragging the mouse. Whole rows may be selected by clicking or dragging on the row descriptions. The selected cells may be copied to the clipboard, or the contents of the clipboard may be pasted in. In this way, data from a spreadsheet or a tab-delimited ASCII file may be directly pasted in. Click Undo to reverse a paste operation.
Note that the Moving/Fixed and Foil Section entries may not be pasted in.
Click on Moving/Fixed to specify which parts of the blade are pitch or aileron controlled. Click on Foil Section to select an aerofoil section (see 3.8) or to define a new section.
If a sharp discontinuity (see 3.1) in blade characteristics is required, split a blade station into an inboard and an outboard station at the same radial position. This is especially useful to mark the discontinuity between a fixed and a pitchable or an aileron-controlled part of the blade. Use the Split and Join buttons to split or re-join highlighted blade stations. The first and last stations may not be split.
On the Blade Information tab, a name may be entered for the blade. The entire blade may also be made pitchable or fixed, depending on whether the blade is mounted to the hub by means of a pitch bearing. Futhermore it is possible to insert a wedge outboard of this pitch bearing by specifying:
Encrypt Blade: The blade geometry, mass and stiffness data may be encrypted, allowing other users to run calculations without being able to see the details of the blade. See 2.7 for further information on encryption.
The aerofoil data used on the blade may also be encrypted in the same way. The aerofoil sections to encrypt are selected from the list on the Blade Information tab.
Aerofoil conflicts: The list of aerofoils on the Blade Information tab will indicate if any of the (nonencrypted) aerofoil datasets used on the blade conflict with aerofoils in the database. In this case, click Resolve conflict to see the differences in detail and to decide whether to replace or rename conflicting aerofoil datasets.
# 3.1 Choosing blade stations
Blade stations are the points along the blade at which blade geometry, aerofoil, mass and stiffness data are defined. They are also the points at which blade loads will be calculated, if desired - see Defining outputs (7.2.4).
The blade stations must include the root and tip stations. Blade station positions are measured from the blade root, which is the point at which the blade is attached to the blade root section (see 4.2). Thus the first blade station must be given a radial position of zero, and the position of the last station defines the length of the blade. This will be less than the rotor radius if a blade root section is defined, or if the blade is not straight.
A reasonably even spacing between blade stations is recommended. Try to define a blade station close to or preferably at the same position as any point mass which may be required - see Blade mass distribution (3.3).
# Discontinuities
The blade may have sharp discontinuities in the mass distribution, the stiffness distribution, and the aerofoil section. There will also be a discontinuity where a blade changes from fixed to pitchable, or at the start or end of any aileron or airbrake section.
To specify such a discontinuity, highlight the blade station at the radial position where the discontinuity lies, and click the Split button to split the blade station into an inboard and an outboard station at the same radial position. The data may then be specified differently on either side of the discontinuity.
A split station may be re-joined if necessary to remove the discontinuity, by clicking the Join button.
# 3.2 Blade geometry
The blade geometry is defined at each blade station (see 3.1) clicking on the data item to be defined or changed. The following data is required at each station:
Distance: This can be entered as a distance along the blade or as a distance along the blade root Zaxis. Select the appropriate option at the base of the screen.
Distance along blade: the distance from the blade root to the current blade station, along the blade neutral axis, which does not have to be a straight line. It must be zero for the first station. If distance is entered along the blade root Z-axis, this value is calculated based on the distance along the blade root Z-axis and the neutral axis. Distance along blade root Z-axis: the distance of the blade station along the nominal pitch axis (with no pre-sweep or pre-cone). If distance is entered along the blade, this value is calculated based on the distance along the blade and the neutral axis. Chord: the distance from the leading edge to the trailing edge, i.e. along the chord line. Aerodynamic Twist: the local angle of the chord line. More positive values of twist and set angle push the leading edge further upwind. (See diagram below) • Thickness: the thickness of the blade as a percentage of the chord at that station. Neutral axis $(\pmb{\times})$ : the distance from the blade root Z-axis to the neutral axis in the x direction. This would be non-zero if for example the blade was pre-bent. (See diagram below)
Neutral axis (y): the distance from the blade root Z-axis to the neutral axis in the y direction. (See diagram below) Neutral axis, local $(\pmb{x}^{\prime})$ : the perpendicular distance from the chord line to the neutral axis in local coordinates, as a percentage of the chord. (See diagram below) Neutral axis, local (y’): the distance along the chord line from the leading edge to the neutral axis in local coordinates, as a percentage of the chord. (See diagram below) • Foil section: an index number defining the aerofoil section (see 3.8) at that station Moving/Fixed: differentiates between a fixed part of the blade and a part which is movable to achieve aerodynamic regulation or braking, either by bodily changing the pitch of that part of the blade, or by deploying an aileron, flap or other aerodynamic control (see 4.1) surfaces.
Blade icing can also be specified, as discussed in section (3.3.4).
Figure 3-1: Diagram showing the Chord, the position of the Neutral axis relative to the blade root Z-axis and the leading edge, and the aerodynamic twist (seen from above). Note that this figure shows a Neutral axis with negative x and y co-ordinates with respect to the blade root co-ordinate system.
# 3.3 Blade mass distribution
For dynamic calculations it is necessary to define the mass of the rotor blades. Click the Mass checkbox off if static calculations are desired and the mass data has not been defined.
# 3.3.1 Distributed mass
For each blade station (see 3.1) enter:
• Centre of mass $(\times^{\prime})$ : the perpendicular distance from the chord line to the centre of mass in local coordintes, as a percentage of the chord. (See diagram below) Centre of mass (y’): the distance along the chord line from the leading edge to the centre of mass in local coodinates, as a percentage of the chord. (See diagram below)
Mass per unit length: the gradient of the blade mass distribution at each station, with respect to the actual distance along the blade. • Polar inertia per unit length: the gradient of the blade polar mass moment of inertia at each station about the local $\scriptstyle{Z_{\mathsf{m}}}$ -axis.. Radii of gyration ratio: the radius of gyration of mass about $\mathsf{v}_{\mathsf{m}}$ divided by the radius of gyration of mass about $\mathsf{x_{m}}$ . This defaults to the relative profile thickness but can be over-written by un-checking the “Use default radii of gyration ratio” checkbox. Mass axis orientation: the orientation of the principle axis of inertia. This defaults to the orientation of aerodynamic twist, but can be over-written by un-checking the “Use default mass axis orientation” checkbox. (See diagram below)
Figure 3-2: Diagram showing the position of the Centre of mass relative to the leading edge, and the mass axis orientation
Click the Additional Mass tab to specify any point masses (see 3.3.2) and/or vibration dampers (see 3.3.3) on the blades.
# 3.3.2 Point masses
If additional point masses are required at specific locations along the blade, click the Additional Mass tab and enter the point masses in the table. Click the Add button to add each point mass, and then enter the data required by clicking on the appropriate entry. Note that masses are automatically sorted by radial position. To remove a mass, highlight it by clicking on its number, and click Delete.
For each mass, the following data is required:
Mass: The Mass required.
Distance along blade: The position of the mass along the blade, measured from the blade root. Chordwise position $(\times^{\prime})$ : the position of the mass perpendicular to the chord as a percentage of the chord at that point. (See diagram below) Chordwise position (y’): The chordwise position of the mass, measured backwards from the leading edge as a percentage of the chord at that point. (See diagram below)
Figure 3-3: Diagram showing the position of point masses relative to the leading edge
Additional pitching inertia should also be specified on this screen.
# 3.3.3 Vibration dampers
Particularly when operating in stall, the aerodynamic damping of the edgewise vibrational modes of the blades may be very low or even negative, causing large vibrations and corresponding oscillatory loads. To prevent this, some blades are fitted with vibration dampers. These consist of mass-spring-damper system tuned to vibrate at the resonant frequency. Enter the mass, frequency and damping factor (fraction of critical damping), as well as the damper position and the direction of vibration relative to the local chord line. For the direction, the sign convention is the same as for blade twist (see 3.2).
# 3.3.4 Blade icing
Bladed represents blade ice accretion by adding mass to the blade at the aerofoil leading edge. This additional mass will modify the blade mass totals and mass distribution properties such as polar inertia and centre of mass.
There are two options for the calculation of ice mass on the turbine blades. The first calculation option is according to the specification of Germanischer Lloyd (GL 2010) [9], the second is according to the draft IEC 61400-1 edition 4 standard.
The methodology used for the GL 2010 approach is as follows. The mass distribution increases linearly from zero at the rotor axis to the value $\mu_{e}$ at half the radius and then remains constant up the blade tip. The value $\mu_{e}$ is calculated as follows:
$$
\mu_{e}=\,\rho c_{m i n}(c_{m a x}+c_{m i n})(0.00675+\exp(-0.32R))
$$
where $R$ is the rotor radius [m], $c_{m a x}$ is the maximum chord [m], $c_{m i n}$ is tip chord $[\mathsf{m}]$ and $\rho$ is the ice density with default value 700 $[\mathsf{k g}/\mathsf{m}^{3}]$ . The chord length at the blade tip is an input in Bladed. The maximum chord value is computed using from the blade geometry (see 3.2) information input by the user.
The methodology used for the IEC Ed 4 is to apply a mass distribution that increases linearly from zero at the rotor axis to the maximum value at the blade tip. The ice mass distribution is calculated using:
$$
M(r)=\rho C_{85}r
$$
where, $M(r)$ is the mass distribution on the leading edge of the rotor blade $[{\mathsf{k g}}/{\mathsf{m}}]$ , $\rho$ is a constant parameter with default value of $0.125\ [\mathsf{k g}/\mathsf{m}3],$ , $C_{85}$ is the chord length at $85\%$ rotor radius, and $r$ is the radial position measured from the rotor axis [m]. The parameter $C_{85}$ is calculated by Bladed based on the geometry information input by the user.
For both ice models, the radial distance from the rotor axis is assumed to be a nominal radius i.e. ignoring the effect of rotor cone, blade prebend and blade root mounting angle. The GL 2010 rotor radius $R$ is always half of the “nominal rotor diameter” as shown in the “Turbine and Rotor” screen in the Bladed GUI. Each blade station radial position $r$ can be calculated using the blade root length added to the blade station “distance along blade root Z axis” at each blade station.
Blade icing can be defined on the “Blade Information” tab of the blade screen.
Note that, unlike in version 4.8 and earlier, the change in mass due to blade icing is no-longer reflected in the Turbine Information screen (when clicking on the Mass Totals button). Instead, the mass distribution and blade mass with ice is provided in the verification $({\Phi}\backslash{\sf E})$ file that is output when simulations are run.
Ice accretion can also modify the aerodynamic properties of aerofoils. No modification is applied to the aerofoil data that has been input into Bladed. If design standards require these modifications to be made then it is recommended that the user applies these corrections manually to the aerofoil input data.
# 3.4 Blade mounting orientation
Blade mounting sweep angle: A positive sweep angle means that the blade tip is swept back from the pitch axis, away from the direction of rotation.
Blade mounting cone angle: This has the effect of adding a contribution to the pre-bend deflection which increases linearly with distance along the blade. The blade root Z-axis is coned upwind from the pitch axis.
Note: when no (zero) mounting sweep or cone angles are specified, the “blade root Z-axis” is the same as the pitch axis. If there are non-zero mounting sweep and cone angles specified, then there is an orientation difference between the pitch axis and blade root Z-axis. All subsequent blade properties are then defined in relation to the blade root Z-axis rather than the pitch axis.
# 3.5 Blade stiffness distribution
For calculations involving blade flexing or vibrational dynamics (see 7.1) it is necessary to define the stiffness distribution of the rotor blades. Click the Stiffness checkbox off if calculations not involving blade flexing are desired and the stiffness data has not been defined.
The stiffness must be defined about the principal axis of inertia at each blade station (see 3.1). The stiffness is the product of Young’s Modulus for the material and the second moment of area for the $\mathsf{x_{p}}$ or $\forall\mathsf{p}$ directions as appropriate. The principal axis orientation is defined as an input, and defaults to the aerodynamic twist. In this case it is assumed to be parallel and perpendicular to the chord line. If the principal axis orientation is different from the aerodynamic twist, click the Use default principal axis orientation to off. (see diagram below)
To enter or edit data, click on the appropriate entry, or move to it with the arrow keys and press Return.
# 3.6 Additional degrees of freedom
To enable an axial degree of freedom, torsional degree of freedom or shear stiffness to be defined, click the appropriate check boxes. The following data then needs to be defined:
Torsional stiffness: Required if a torsional degree of freedom is included. Defined with respect to the shear centre.
Axial stiffness: Required if an axial degree of freedom is included.
Shear centre $(\times^{\prime})$ : Required if a torsional degree of freedom or shear stiffness is included. This is the perpendicular distance from chord line to shear centre in local coordinates as a percentage of the chord. (See diagram below)
Shear centre (y’): Required if a torsional degree of freedom or shear stiffness is included. This is the distance along the chord from the leading edge to the shear centre in local coodinates, as a percentage of the chord. (See diagram below)
Shear stiffness along $\pmb{x_{p}}$ : Required if shear stiffness is included. Defined in the principal axis direction as specified by the principal axis orientation. (See diagram below)
Shear stiffness along $\pmb{\upgamma_{\mathsf{p}}}$ : Required if shear stiffness is included. Defined in the principal axis direction as specified by the principal axis orientation. (See diagram below)
To enter or edit data, click on the appropriate entry, or move to it with the arrow keys and press Return.
Figure 3-4: Diagram showing the orientation of the principal axes and the location of the shear centre relative to the leading edge
# 3.7 Conversion from older versions of Bladed
If you open a Bladed project file from a version prior to version 4.0, Bladed will convert the blade into the new format according to the following rules:
The neutral axis is assumed to follow the centre of gravity for this conversion
| REFNUM | up to32 characters | A unique identifier for this dataset |
| * | | Any commentlineshere,starting withanasterisk(*), |
| * | | willbeincorporated in thedatabase |
| * * | | |
| XA | % | Pitchmomentcentre,backfromleadingedge |
| THICK | % | Thickness/chordratio |
| REYN | | Reynoldsnumber |
| DEPANG | degrees | Ailerondeploymentangle |
| NALPHA | | Numberofanglesofattacklistedinthetable |
| NVALS | 2or3 | Number of coefficients:2 if just lift and drag |
| | coefficients are available, 3 if pitch moment |
| | coefficients are also present |
The table of coefficients follows, consisting of NALPHA lines of data each containing:
Angle of attack (degrees), Lift coefficient, Drag coefficient $[,$ moment coefficient if NVALS $^{=3}$ ]
with the values on the line separated by commas, spaces or tabs. Finally, there should be a line consisting of the termination string:
# ENDSECTION
# Example ASCII aerofoil dataset
A very simple example for an ASCII aerofoil database file, just to illustrate the format required, is given here:
REFNUM SET1 \* Example dataset \* Use these lines for a description XA 25.0 THICK 18.0 REYN 3.E6 DEPANG 0.0 NALPHA 12 NVALS 3 -180 0 0 0 -20 -1 0.3 0.037 -12 -0.88 0.03 -0.067 -8 -0.48 0.0107 -0.078 -4 -0.02 0.0073 -0.089 0 0.45 0.0061 -0.104 4 0.891 0.0069 -0.121 8 1.323 0.016 -0.119 12 1.728 0.0218 -0.116 16 2.119 0.07 -0.111 20 2.478 0.225 -0.112 180.0 0.0 0.0 0.0 ENDSECTION
# 3.11 Defining normal aerofoil sections
A normal aerofoil section should be defined for every blade station which is either: • Fixed (see Blade geometry (3.2)) or • Moving (see Blade geometry (3.2)) and pitchable (see Control surfaces (4.1)).
Normal aerofoil sections allow interpolation on up to three Reynolds numbers and up to two thickness/chord ratios - see Aerofoil sections (3.8). A very flexible interpolation scheme is used, so for example it is not necessary for the individual datasets to have matching angles of attack.
A number of different aerofoil sections may be defined, for use at different blade stations.
On the Blade properties window, click on the foil section to be defined. A drop-down list allows an already-defined section to be selected, or select Define... to open the Define Aerofoil Sections window. This allows the characteristics of already-defined sections to be viewed or edited, or new ones created by clicking New.
To view or edit an existing aerofoil section: In the Define Aerofoil Sections window, select the section number required.
To set up a new aerofoil section: In the Define Aerofoil Sections window, press New to start a new foil section.
# To edit an aerofoil section:
Up to six aerofoil Dataset (see 3.9) names may be entered in the boxes provided. Click on any white or red box to open the Aerofoil Dataset Selection window. This presents a list of all suitable datasets from the database (they must be in ascending order of Reynolds number and thickness). If desired, the selection criteria may be modified to further restrict the list displayed. Select a dataset from the list and click OK.
If fewer than six boxes are required, the boxes used must form a rectangular pattern so that the interpolation scheme is fully defined. The top left hand box must always be used.
# To remove an aerofoil section:
There is no need to delete an aerofoil section if it is no longer needed at any blade station. The sections which are still in use will be renumbered when the Blade Properties window is closed, starting from 1. The aerofoil datasets themselves of course remain in the database.
# 3.12 Defining aileron sections
An aileron section should be defined for every blade station which is defined as Moving (see Blade geometry (3.2)) if ailerons are specified for the aerodynamic control surfaces (see 4.1). Flaps and airbrakes can be treated as if they are ailerons.
Aileron sections allow interpolation between a set of aerofoil datasets (see 3.9) defined for different aileron deployment angles.
A number of different aileron sections may be defined if required, for use at different blade stations.
On the Blade properties window, click on the foil section to be defined. A drop-down list allows an already-defined section to be selected, or select Define... to open the Define Aerofoil Sections window. This allows the characteristics of already-defined sections to be viewed or edited, or new ones created by clicking New.
To view or edit an existing aerofoil section: In the Define Aerofoil Sections window, select the section number required.
To set up a new aerofoil section: In the Define Aerofoil Sections window, press New to start a new foil section.
# To edit an aesection:
Click Add or Insert to increase the number of aerofoil datasets (see 3.9) to interpolate between. Click Delete to remove one. Double-click on any entry to open the Aerofoil Dataset Selection window. This presents a list of all suitable datasets from the database (they must be in ascending order of deployment angle). If desired, the selection criteria may be modified to further restrict the list displayed. Select a dataset from the list and click OK.
No entries may be left blank.
# To remove an aerofoil section:
There is no need to delete an aerofoil section if it is no longer needed at any blade station. The sections which are still in use will be renumbered when the Blade Properties window is closed, starting from 1. The aerofoil datasets themselves of course remain in the database.
# 4 DEFINING THE REST OF THE TURBINE
Having defined the turbine blades, this section describes how to build up a model of the complete turbine structure including the rotor and hub, power train, tower and nacelle.
# 4.1 Defining the rotor
Click the Rotor icon on the toolbar to define the basic characteristics of the turbine rotor and hub. The following data must be defined:
Rotor diameter Number of blades Hub height: from the ground to the centre of the rotor (i.e. the intersection of the blade and shaft axes). Tower height: from the ground or sea surface to the yaw bearing (only needed if the tower itself has not been defined). See also: Tower (4.5). Hub vertical offset: the vertical distance from the yaw bearing to the hub. Blade set angle: the angle at which the blade is mounted onto the hub. More positive values of set angle push the leading edge further upstream. It is usually convenient to define the set angle as zero, but particularly for stall regulated machines, the set angle provides a simple way of rotating the whole blade (both pitching and non-pitching sections) without having to re-define the twist distribution. Cone angle: the angle between the blade axis and the rotor plane. Tilt angle: the angle between the shaft and the horizontal (normally positive). • Overhang: the horizontal distance between the rotor centre and the tower centreline. Lateral offset: the horizontal offset between the shaft and tower axes. Rotational sense: the turbine may rotate clockwise or anti-clockwise when viewed from upstream. Position: the rotor may be upwind or downwind of the tower in normal operation. Speed type: the turbine may be defined as a fixed or a variable speed turbine. See also Generator (4.10), Control (5). Control surfaces: Specifies whether the blade is fixed, pitchable (includes full or partial span pitch) or aileron controlled (includes any type of flap or airbrake). See also Blade Geometry (3.2), Control (5), Start-up (5.13), Normal stop (5.14), Emergency stop (5.15). Transmission: select Gearbox or Direct Drive. For a direct drive system, either set the gearbox ratio (see 4.8) to 1, or select the Direct Drive option here. If the Direct Drive option is selected, the generator mass can be separately identified on the Hub screen (see 4.2), otherwise it must be included in the Nacelle mass (see 4.13). Cut-in wind speed: the steady wind speed at which the turbine is brought on or off line in low wind speeds. Cut-out wind speed: the steady wind speed at which the turbine is brought on or off line in high wind speeds.
Clicking the Hub tab allows the hub configuration to be defined, together with generator mass information if the Direct Drive option is selected. This is described in Section 4.2.
# 4.2 Blade and rotor coordinate summary
In this section, the relationship between blade and rotor properties and coordinate systems are displayed, using views from the side and in front of the turbine.
These views are shown for a clockwise upwind turbine.
Figure 4-1: Hub and blade coordinate system summary (view from the side of the turbine)
Figure 4-2: Hub and blade coordinate system summary (view from in front of the turbine)
# 4.3 Defining the hub
From the Rotor (see 4.1) screen, click the Hub tab to define the hub characteristics.
Enter the Spinner diameter. This is the diameter of any spinner or nose-cone, within which the blades themselves experience no aerodynamic forces.
# Blade root section
The blade root section connects the blade root to the shaft. It is assumed to be cylindrical in section. It is defined by:
Length: this defines the actual radius of the first blade station (see ${\underline{{3}}}.$ ). Set this to zero if there is no root section. The rotor diameter (see 4.1) must be twice the sum of the root section length and the blade length, defined by the radial position of the last blade station.
Diameter: the diameter of the cylindrical tube which forms the blade root section.
Drag Coefficient: the drag coefficient to be used for any part of the blade root section which is outside the spinner. Zero lift coefficient is assumed.
If the spinner entirely covers the root section, the diameter and drag coefficient values will not be used.
For one and two bladed rotors, click the check box to specify a teetered hub if required. See also Options (7.2.3). For teetered hubs, enter a Delta-3 angle if pitch-teeter coupling is required. A positive delta-3 angle acts to stabilise teeter by increasing the angle of attack when the blade teeters into the wind. A Special setting is available in case other forms of pitch-teeter coupling are required, by special arrangement with Garrad Hassan.
# Teeter restraint
If a teetered hub is selected, a teeter restraint system may be specified by clicking the Teeter restraint button (see 4.4).
# Mass information
If the Direct Drive option is selected, enter the moment of inertia of the generator rotor here.
For dynamic calculations, check the Mass checkbox to allow the necessary masses and inertias to be specified as follows:
• Hub mass: the mass of the hub, including the spinner and any blade root section. Hub mass centre: the distance from the intersection of the shaft and blade axes to the centre of mass of the hub, in a direction measured away from the tower. Moments of inertia: the moment of inertia of the hub mass about the shaft axis must be defined. The inertia about an axis perpendicular to the shaft may also be entered with its origin about the hub centre of mass.
For direct drive systems, enter also the generator mass information as follows:
Generator mass Generator mass centre: the distance between the centre of mass of the generator and the centre of mass of the hub. Moments of inertia: the moment of inertia of the complete generator may be defined, about the shaft axis and a perpendicular axis. In this case, ensure that the mass information entered on the Nacelle (see 4.13) screen does not include these contributions.
For one-bladed rotors only, define the counterweight as follows:
• Mass: the mass of the counterweight.
• Inertia about shaft: the moment of inertia of the counterweight about the shaft axis.
# Note
The Turbine Information window shows the total masses and inertias of all turbine components whose mass characteristics are defined. Click Mass Totals, or use the Windows pull-down menu on the main toolbar to open this window.
# 4.4 Teeter restraint
A spring and damper model of a teeter restraint system is provided. If a teetered rotor has been specified, click the Teeter restraint button on the Hub screen, select Standard model, and enter the following parameters:
· Free teeter angle: The teeter restraint only start to act when this teeter angle is exceeded. Spring preload: The torque required to start moving the teeter restraint. Spring stiffness: The rotational stiffness of the teeter restraint. Teeter damping: The rotational damping of the teeter restraint.
If the teeter hinge has a delta-3 angle, the teeter restraint is on the flapwise motion resolved through the delta-3.
Click Special if a client-specific teeter restraint model has been provided by arrangement with Garrad Hassan, and enter the appropriate parameters.
See also: Calculation options (7.2.3).
# 4.5 Defining the tower
Click the Tower button on the toolbar to define the tower characteristics. Two types of tower model are available. These are the tubular axisymetric (or monopile) and for users with an Offshore Support Structure module, the multi-member model.
# 4.5.1 Monopile Tower
The tower is modelled by defining its characteristics at a number of stations from the tower base to the tower top. Tower properties must be defined for at least two tower stations (the tower base and tower top), although if tower vibrations (see 7.1) are to be modelled, a minimum of 5 stations is recommended in order to achieve a reasonable degree of accuracy.
Check the Tower geometry check-box to define the tower dimensions. These are used to calculate the tower shadow (see 6.3) and windage loads.
Click the Add button to add a new tower station. Stations are automatically sorted by height. To highlight a station, click on the station number or a graph point. Click Delete to remove a highlighted station.
To enter or edit data, highlight the data item by clicking on it, or by moving to it using the arrow keys and then pressing the Return key. Press Escape to restore the previous value.
Highlight a block of cells by dragging the mouse. These cells may be copied to the clipboard, or the contents of the clipboard may be pasted in. In this way, data from a spreadsheet or a tab-delimited ASCII file may be directly pasted in. Click Undo to reverse a paste operation.
Enter the tower station height and tower diameter for each station. In the case of an onshore (see 4.5.5) turbine the station heights are defined relative to ground level, whilst for an offshore (see 4.5.5) turbine, station heights are defined relative to the mean water level.
If the first tower station is above the ground or sea bed level, the tower is assumed to be mounted on a rigid pedestal, as indicated on the Tower Geometry diagram. If the height of the first tower station is below the ground or sea bed level, the tower foundation is assumed to be buried as indicated, and no external forces are assumed to act on the buried portion of the tower.
For all cases, the height of the top tower station must correspond to the tower height defined on the Rotor (see 4.1) screen.
The Show button controls the graphical display of tower properties. The graph is updated as soon as any new values are assigned, giving an instant indication if faulty values are entered. Controls are provided to print the graph or save it to a Metafile, which can subsequently be incorporated into a report.
Check the Mass check-box to allow the mass per unit length to be entered at each tower station. This is necessary for correct calculation of the tower base loads, and also for modelling tower vibrations (see 7.1), in which case it is also necessary to check the Stiffness check-box and supply bending stiffness values at each station. These are defined as the product of the Young’s modulus and the second moment of area. Check the Shear flexibility check-box to include effects of shear flexibility and to allow entering the shear stiffness at each station. These are defined as the product of the Shear modulus and the shear area that equals half of the cross-section area for a circular cross-section. Check the Torsional degree of freedom check-box to allow the torsional stiffness and polar moment of inertia to be entered at each station.
Alternatively, for a tower of circular cross-section, the mass and stiffness distributions can be calculated automatically by entering the wall thickness at each station. Type the name of a material and enter its density and Young’s modulus in the boxes provided. If the torsional degree of freedom or the shear flexibility is included, the Shear modulus should also be entered. A number of different materials may be entered if desired. Then select the appropriate material for each station.
If a discontinuity is to be defined, for example a step change in wall thickness or other property, simply enter two stations at exactly the same height to define the discontinuity. Do not enter very closely spaced stations.
# 4.5.2 Multi-member tower
Users with an Offshore Support Structure module licence can define the tower and support structure as a series of nodes, joined by a series of members.
Click the Add Node button to add new nodes. The height, and x and y positions of the node should be specified. The $\times$ -axis initially points south, but can be rotated by entering a non-zero value in the $\mathbf{x}.$ -axis clockwise from south box. Point masses can be defined at any of the nodes. The node at which the nacelle joins the tower must be selected from the Nacelle node drop-down menu. The selected node must be the highest node on the structure.
Clicking Add Member will allow members to be defined to join the nodes. The grid will display two rows per member, one for each end of the member. Most parameters can be different at each end, but some must be the same, for example the material properties and whether or not the member is flooded. Selecting the Allow tapered wall thickness option will enable the wall thickness to vary along the member. Clicking on the materials column will bring up a drop down list of all the materials defined. Select the appropriate material for that member. If a diameter, wall thickness and material have been defined, the various mass, stiffness and inertia properties will be calculated automatically. These can be overwritten if desired. The diameter is assumed to be the outside diameter for the purposes of mass, stiffness, inertia and drag calculations.
Marine growth (fouling): For an offshore turbine, enter a value for the thickness in millimetres of marine growth (fouling) at each end of a member, if required. The thickness is assumed to vary linearly along the member. The increase in member diameter due to the marine growth affects both the drag on the tower member and its mass. Note that the additional mass due to marine growth is not included in the tower mass reported in the Turbine Information window.
When the multi-member tower option is selected, each member is defined with its own local coordinate system. The local $\times$ -axis is defined along the member axis, in the direction from End 1 to End 2. Thus $\mathsf{F x}$ is the member axial force and $\mathsf{M}\times$ the torsional bending moment. Clicking on the Member axes button will bring up the direction cosines for the member z-axis. These will have defaulted to an orientation in line with the convention for other offshore engineering codes. The default orientation in Bladed is to set the local member y-axis in the horizontal plane, with the local member z-axis making up a right-handed coordinate system. For a vertical member the default orientation is for the local y-axis to be in the global $\times$ -direction, with the local z-axis in the global y-direction. Thus for a vertical member Mz describes the principal overturning moment and My the side-side overturning moment. These orientations can be changed by the user if required.
Clicking on the Display Structure button will bring up a 3D drawing of the support structure. It is possible to rotate, pan and zoom using the appropriate toolbar options. Clicking on the green $\mathsf{w}_{\mathsf{N}^{\prime\prime}}$ toolbar option will label the nodes in the display, and the $\mathfrak{"}\mathsf{M}^{\prime\prime}$ will label the members. If members are selected in the Member grid whilst the plot is open, the selected members will be highlighted in red and the local coordinate systems displayed.
# 4.5.3 Geometric Stiffening
The elastic stiffness of the tower is constant. However, in reality the stiffness of a structure will vary depending on the axial and shear loads. The change of stiffness of the structure depending on these loads is the geometric stiffening. This effect is important for blades, and is included by default, but is not always important for support structures. Enabling “Use geometric stiffening” will give more accurate results but may result in slower simulations.
# 4.5.4 Flanges and point masses
Flanges should be modelled as point masses, not by an increase in wall thickness between closelyspaced stations or nodes. For the monopole tower model, Click the Point masses button to define the
height and additional mass of each flange or other point mass. For the multi-member tower model, point masses can be added at any node by entering the value directly into the nodes grid.
# 4.5.5 Vibration dampers
Click the Vibration damper button to define a mechanical vibration damper, characterised by its mass $(M)$ , the resonant frequency $(\omega)$ to which it is tuned, and the damping factor ( ${\mathit{\check{\zeta}}},$ fraction of critical damping). Enter also the damper position: the number of the node or station at which it is attached. A damper is usually Omnidirectional, but a Unidirectional damper may also be defined: this would be mounted in the nacelle, and the Damper angle specifies its direction of motion relative to the fore-aft direction.
The damper is modelled by the Bladed multibody dynamics as a mass-spring-damper (M-K-C) system, with properties defined as follows:
$$
\begin{array}{l}{M=M}\\ {K=\omega^{2}M}\\ {C=2\zeta\sqrt{K M}}\end{array}
$$
The stiffness and damping values can be specified as constants or using a look-up table. The look-up table allows a non-linear vibration damper system to be modelled. Please see the Project Info User Guide for more information [7] on how to specify the look-up tables.
In the “Tower damper” output group, Bladed outputs the displacement and velocity of the tower damper relative to the tower node to which the damper is attached. Note that as the tower node rotates due to modal deflection or rigid body rotation, the coordinate system in which the displacements and velocities are reported will also rotate.
# 4.5.6 Support structure superelement
For offshore turbines, the jacket support structure is sometimes modelled as a superelement. The superelement can be included as a component in the multibody framework in a similar way to the other flexible components. Full details of Bladed’s superelement functionality can be found in DNV GL document “Superelement Support Structure - User Guide for Bladed” document number 110052-UKBRT-37
# 4.5.7 Environment
Click either Land or Sea in the Environment panel to define whether the turbine is sited onshore or offshore. The Tower Geometry diagram will then show either the ground level or the sea bed and mean water level, as appropriate. For an offshore turbine, enter the mean sea depth. To change the mean water depth relative to the base of the tower, for the monopile tower model change the Depth of first tower station as appropriate, and for the multi-member model, change the Z-axis origin above mean water level. Note that for individual calculations, the actual sea depth can be changed by specifying the tide height (see 6.14).
Enter the aerodynamic drag coefficient to enable aerodynamic forces and damping to be calculated. For an offshore turbine, enter also the hydrodynamic drag and inertia coefficients, to enable wave and current forces and hydrodynamic drag to be calculated, or specify them individually for each tower station or member. For a description of wave and current loading, see the Theory Manual.
For an offshore multi-member turbine only, specify also the density of any marine growth. If any members have non-zero marine growth thickness specified, this density is used to calculate the additional mass due to marine growth.
# Note
The Turbine Information window shows the total masses and inertias of all turbine components whose mass characteristics are defined. Click the Mass totals button or use the Windows pull-down menu on the main toolbar to open this window. Marine growth is excluded from these mass totals.
# 4.5.8 Hydrostatics of Isolated Elements
Hydrostatic forces are calculated by considering the pressure forces on structural elements in isolation. Support structure members can be defined as “sealed” or “unsealed”. “Sealed” elements can be defined as “flooded” or “unflooded”. These terms are defined in this section, illustrated in Figure 4-3, and summarised in Table 4-1.
# 4.5.8.1 Sealed vs. Unsealed members
The “sealed” or “unsealed” options represent whether the ends of the members are sealed with a plate.
This results in different pressure forces being applied to the elements.
Pressure forces on “sealed” elements:
Element end pressure forces act on the whole element end cross-sectional area calculated using member outer radius at each end.
Distributed forces on the element sides act only on the outside surface of the member
For an element in isolation, the buoyancy force corresponds to the mass of water displaced by metal and the space inside the members.
Pressure forces on “unsealed” elements:
• Element end pressure forces act on the element wall cross-sectional area at each end, calculated using the element radius and thickness. Distributed forces on the element sides act on the inside and outside surface of the member For an element in isolation, the buoyancy force corresponds to the mass of water displaced by the metal only.
“Unsealed” elements are assumed to be full of water up to the water line. In this case, the inertia of water inside the element is accounted for by including hydrodynamic added mass, which acts only in a direction perpendicular to the element axis.
# 4.5.8.2 Flooded vs. Unflooded members
“Sealed” elements can be specified as “flooded” or “unflooded”. This determines whether the sealed element contains water. For flooded members, the mass of enclosed water is added directly to the element structural mass.
The hydrostatic pressure forces on “flooded” and “unflooded” elements are identical.
Figure 4-3: Sealed and unsealed structural elements
| BladedHydrodynamicInputParameters |
| Parameter Name | Description |
| |
| Body Interaction Order | Describestheorderinwhichtheinter-bodycross-termsappearintheadded o l s s o s i s where N is the total number of wave activated bodies in the system. Each entry |
| Body Hydrodynamic Origin(ifoptionto manually specify has | should appear on a new line. (x,y,z,roll,pitch,yaw) vector describing the position and orientation of the body |
| been selected) Global viscous drag | parameter. Optional 6x1 vector defining six-degree-of-freedom viscous drag characteristics for the body. The drag coefficients form a 6x6 matrix; all except thesixdiagonalelementsareassumedzero,anditisthesesixthatareentered |
| loads will be applied at the same support-structure node as the BEM ones. |
| Excitation Data Check box.Tick to indicate that the direction ordinate for the excitation force Cartesian Direction |
| Convention | data is provided in global Cartesian angles (measured clockwise from positive X).Un-tick to specify nauticalangles(measuredclockwise fromNorth). |
| North Direction | Cartesian angle of excitation force direction ordinate North axis, measured anti-clockwisefrom theglobalX direction. |
| ExcitationForce Data Frequencies | Excitation force data frequency ordinate data provided as a list of scalar numbers. |
| Excitation Force Data Directions | Excitation force data direction ordinate data provided as a list of scalar numbers in radians if Cartesian angles are being used and degrees if nautical bearings are being used. |
| Excitation Force Data Amplitudes | Excitationforceamplitude andphase data.Listsofmatrices,one entryforeach Excitation Force Directions value. Each matrix has six rows (Surge, Sway, |
| ExcitationForceData Phases | Heave,Roll,Pitch,Yaw)anda number of columns equal to the number of Excitation Force Frequencies.. Radiation Data |
|
| Added Mass atInfinite Frequency | Addedmassandmomentsofinertiaatinfinitefrequency.Includesinter-body cross-terms so this data is a.6x6N Matrix for each body where N is the total number of wave activated bodies (total number of hydrodynamic components) in the model. If the impulse response function time ordinate is equally spaced, then the |
| Time Ordinate Resolution Time Ordinate | spacing may be specified here to speed up run-time interpolation. A value of lessthanOshouldbeenteredifthespacingisuneven. Impulse response function time ordinate, entered as a scalar list of time values. Bladed will interpolate the Impulse Response Function Matrices for this data. |
| Impulse Response Function Matrices. | terms in column-wise the order specified by 'Body Interaction Order'. Data is entered as list of 6x6N matrices, one list entry foreach Impulse Response Time value.Nis the totalnumberof waveactivatedbodies(hydrodynamic |
| Cut-Off Time | components) in the system.Each new matrix must start on a new line if added manuallyusingthe ListEditor. Length of time-history used in the evaluation of the radiation force convolution for the body in question. Short time will speed up the simulation but will result in lossofaccuracy.See5.4.1for moreinformation. |
| Independent RadiationCalculation | Tick to specify a minimum radiation convolution calculation timestep that is coarser than the minimum integrator timestep. This will speed up simulations (as the convolution calculation will be completed less frequently), but may result in loss of accuracy. If this option is not selected, the radiation force |
| Timestep Calculation Timestep | calculation will be completed at a timestep determined by the Bladed numerical integrator. Minimum timestep at which to calculate the radiation convolution force term - onlyenteredifIndependentRadiationCalculationTimestepisselected. |
| Reduce Radiation Convolution Resolution | isnotselected,theconvolutionwillbecompletedwitharesolutiondictatedby thenumericalintegratortimestep. |
| Convolution Resolution | Timeresolutiontoimposeontheradiationforceconvolutionbuffer-only enteredifReduceRadiationConvolutionResolutionisselected.IftheBladed numericalintegratortakesstepsizesthatarelargerthanthetimeresolution specifiedherethentheBladedintegratorstepsizewillbeusedandthe |
| convolutiontimeresolution. Hydrostatic Data |
| Mean Displacement Mass | |
| Centre of Buoyancy | (x,y,z)vectordescribingthepositionofthebodycentreofbuoyancyrelative tointhehydrodynamicsoutputcoordinatesystem(theorientationofthe hydrodynamic coordinate system matches the Bladed global coordinates system) |
| HydrostaticStiffness | 6x6 Hydrostatic stiffness matrix for Surge, Sway, Heave, Roll, Pitch, Yaw. Stiffness terms relate to changes in the hydrostatic force on the body as it moves about its Hydrodynamic CoordinateSystem Origin. |
# 4.6.5 The flow solver data import tool
A large amount of hydrodynamic data must be entered for each hydrodynamic body and so an automatic importing tool has been provided for users of the WAMIT or AQWA potential flow solvers. Once hydrodynamic bodies have been added for all appropriate sections of the tower in Bladed, click on the Import flow solver data… button to open the importing tool.
The flow solver data import process consists of two stages. New flow solver data must be processed in to the format required by Bladed. Following this, the data may be imported to the turbine model. Once flow solver data has been processed once, it may be imported multiple times as Bladed saves the processed data in the flow solver data output folder. A single hydrodynamic data file is saved for each body, with a \*.wdnh extension. The flow solver data import tool presents the option to skip directly to the import itself alongside the option to process data from either WAMIT or AQWA. During the import process, the individual body.wdnh files much be matched up with the hydrodynamic bodies applied to the tower nodes.
The parameters required by the processing stage of the flow solver data import tool are described below. The processing tool will write a set of Bladed formatted output files alongside the.wdnh files and so the effect of the processing on the hydrodynamic data may be easily analysed in the standard Data View tool.
| FlowSolverDataProcessingBodyInputParameters |
| Parameter Name | Description Parameter to allowdown-samplingoftheexcitationforcedata.Use avalueof 1 |
| Data Down- Sampling Factor | to keep the full flow solver output, 2 to use 1 frequency in 2, 3 to use 1 in 3 etc. This is can be useful because the flowsolversimulationfrequencyresolution tends to be very fine for time-domain simulations to ensure that the impulse response functioncalculationis numerically accurate.The result is that a large amount of excitationforcefrequenciesareprovided.Bladedinterpolateslinearlybetween still requires a high data resolution so a value of 1 is recommended unless project files is very limited. Thelengthofimpulseresponsefunctiontocalculate.A typicalvalueis120s as |
| ImpulseResponse Function Length | significantimpulse response functions(allthose withappreciablemagnitude) have decayed. Investigations may reveal that shorter times (as low as 30s) may |
| ImpulseResponse FunctionResolution | thisprocessingstepandpotentially cut shorterbeforerunninga simulation. Thetimeresolutionatwhichtheimpulseresponsefunctionisevaluated.Bladed willinterpolatefortimestepsbetweenthedatapoints.Afineresolutionallows ladedtointerpolatethedatamoreaccuratelyonsmallersimulationtimesteps 0.1s is typical although the user should check the smoothness of the resulting impulseresponsefunctiondata. |
| Radiation Damping Minimum Cut-off Period | The minimum wave period (highest frequency) to be considered for the impulse response function calculation. This parameter specifies a minimum period below which flow solver radiation damping data is discarded for impulse response function calculation. If the mesh describing the body geometry used in the flows then there will be a large amounts of noise in the radiation damping data at high frequencies. The user can (and should) choose to discard this noise by specifying a minimum wave period(maximum frequency)to process. If this means cutting off data frequencies where the radiation damping data, prior to noise disturbance, does not appear to have decayed to zero for all components, then a finer mesh (with a smaller panel size) should be used in the flow solver simulations. |
| Cartesian Direction Convention | Booleanparameterindicatingwhether thedataprocessedforBladedshoulduse CartesianorNauticalwavedirections.Cartesiananglesareassumedmeasured anti-clockwise from positive global X, whilst nautical directions are assumed to be which the pre-processor provides the excitation data to Bladed, not the format of the direction data supplied by the flow solver (this is typically fixed and handled automaticallyforWAMITandAQWA).ACartesiansystemisrecommendedunless the user has a specific reason to for viewing the excitation data in nautical |
| North Direction | coordinates. TheCartesiandirectionofNorthforuseinwhentheCartesianDirection Convention is not set. The North direction is entered as a direction measured anti- |
| WaterDensity | clockwisefrompositiveGlobalX. Waterdensitytousetodimensionalisetheflowsolverdata. |
# 4.6.6 Flow solver calculation requirements
Any commercial boundary element method potential flow solver may be used to provide Bladed inputs as long as it is capable of evaluating the following for each body:
Hydrodynamic added mass as a function of frequency. The value at infinite frequency (zero wave period) is also required, although a high frequency value may be used in its place (AQWA).
Radiation damping as a function of wave frequency if the flow solver processing tool used. Alternatively, the impulse response function must be calculated directly for a suitable length of time (until oscillations in all terms decay to negligible amplitude). Wave excitation force amplitude coefficient and phase. The product of the amplitude coefficient and the wave amplitude yields the force amplitude for a given frequency. • The hydrostatic water-plane moments of area expressed as a hydrostatic stiffness matrix.
In addition, the mean mass of water displaced by the body may be evaluated by the flow solver, or may be determined independently by the user.
All of these quantities must be evaluated for the body in a manner that accounts for the radiation from and diffraction around adjacent bodies in the device or array, or obstacles in the water, so a single flow solver run should be used for all the hydrodynamic bodies in a single Bladed simulation. The flow solver quantities should be evaluated for the bodies placed in the positions they would adopt in calm seas, or at a steady operating position (so for example the effect of mooring lines or negatively buoyant bodies pulling a body downwards on the overall draught of the turbine should be accounted for in the body positioning when the flow solver calculation is set up). These physical locations may then be entered in to Bladed (converted in to the Bladed global coordinate system) as the Hydrodynamic Coordinate System Origin for each body. The option to auto-calculate this parameter using the node locations provided in the tower geometry screen is enabled by default however however.
Where flow solver output data is provided in matrices, containing direct and inter-body (body interaction) cross-terms, the body order in which the cross-terms appear should be noted and entered in the ‘Body Interaction Order’ parameter in Bladed. Bodies should be listed in the format “Body at node $\mathsf{X^{\prime\prime}}$ , with each entry being specified on a new line.
The wave frequency range selected for use in the flow solver should encompass the complete range of wave regular wave frequencies to be run in Bladed and the frequencies of all the components used to model the sea states for any irregular wave simulations. Bladed will interpolate the hydrodynamic data for sea-state frequencies in between the computed values. To reduce computational time it is advantageous to exclude frequency components with close to zero spectral density from the input.SEA file at both the high and low frequency ends of the spectrum. The SEA file generator tool does this by default.
Flow solver output data should be analysed carefully before use in a Bladed model as it is possible that some of the radiation damping matrix terms will not have decayed to zero at either end of the initial frequency range selected. It is also possible that too coarse a frequency range will prevent significant peaks from being correctly resolved. The frequency range may need to be widened or the resolution improved even if it means extension beyond the frequency range where the device is known to be responsive or the range at which there is significant energy in the spectrum. Running at a finer resolution than the input seastate definitions may also be required.
Further to the above, it is important to ensure that the mesh of the body geometries is sufficiently fine to evaluate all quantities up to the highest frequencies required without introducing numerical noise (which will appear as large and apparently random oscillations in the data at adjacent frequency points). Convergence tests on the flow solver simulations are advised to determine when a suitable mesh has been refined. Note that high frequency noise in the data may be ignored but only if it occurs after the radiation damping terms have decayed to zero or the added mass terms have stabilised, as this may never converge. The noise may be cut-off by specifying a Radiation Damping Minimum Cut-off Period to be used in the flow solver data processing to produce impulse response functions.
# 4.6.7 Defining the sea state for radiation diffraction hydrodynamics
The radiation diffraction hydrodynamics formulation requires linear sea state definition in a.SEA file format. For details of how to generate a.SEA file using the SEA file generator tool, see 6.12.5.
# 4.6.8 Current limitations
The radiation diffraction hydrodynamics model released in v4.6 does not incorporate the effects of wavecurrent interaction. Wave loads are applied assuming the wave field described in the.SEA file is undisturbed.
# 4.7 Defining the power train
Click the Power Train button on the toolbar to define the following aspects of the power train: • the transmission (see 4.8) (gearbox and shafts),
• any flexible mountings (see 4.9),
• the generator (see 4.10),
• the energy losses (see 4.11), and
• the network (see 4.12) to which the turbine is connected.
# 4.8 Transmission
The Power Train button on the toolbar allows the drive train or transmission to be specified. The transmission consists of the gearbox and shafts. Two alternative ways of specifying the drive train are provided:
Locked speed model (see 4.8.1) Dynamic model (see 4.8.2)
Users with an Advanced Transmission Interface module licence may extend the dynamic model by supplying user-defined gearbox dynamics of arbitrary complexity through a defined DLL interface (see 4.8.3).
# 4.8.1 Locked speed model
This model is available for simple calculations in which the turbine rotor is assumed to rotate at an absolutely constant speed. The rotor speed is specified along with the gearbox ratio, but no other characteristics of the drive train nor of the generator may be specified. Flexible mountings cannot be modelled either. This model is useful for initial aerodynamic design of fixed speed rotors, but for detailed performance and loading calculations the dynamic model is more appropriate.
# 4.8.2 Dynamic model
This model includes the rotational degree of freedom of the rotor. Torsional flexibility of the drive train may be included as an option.
Unless the Direct Drive option has been selected on the Rotor screen (see 4.1), the gearbox ratio must be specified, along with the moment of inertia of the generator rotor. The additional inertia of the high speed shaft may also be specified along with the inertia of the gearbox which is referred to the high speed shaft. If there is a brake on the high speed shaft, its inertia should be included as shown; depending on the location of the brake, it may be more appropriate to include it in the gearbox inertia. Note that the moment of inertia of the turbine rotor is defined by the blade mass distribution (see 3.3) and the hub inertia (see 4.2).
The rotor speed will be governed by the characteristics of the generator (see 4.10), which must be defined in this case. For a variable speed turbine it is also necessary to define an appropriate controller (see 5) to define the speed schedule. If there is a shaft brake (see 5.12) to be used in stopping and parked calculations, specify where it is located on the high speed or low speed shaft. Click the Brakes button to specify the dynamic characteristics of the brakes.
Use the check-boxes to specify if there is significant torsional flexibility in the low speed and/or high speed shaft, and supply the torsional stiffness. If both low and high speed shafts are flexible it is necessary to define the gearbox inertia. Torsional damping may also be specified. For fixed speed turbines, the damping is unlikely to have much effect as it will be swamped by the damping provided by the induction generator. In fact the damping may be sufficient to ensure that there is very little drive train torsional motion, in which case it may be advantageous not to specify any shaft torsion, as the simulations will then run faster.
Note: If not including flexibility in the high speed shaft, gearbox or pallet mounting, the LSS stiffness parameter should reflect the total torsional flexibility between the rotor and the tower top. For most geared transmission configurations this will include the series combination of low speed shaft flexibility and equivalent torsional flexibility of the gearbox mounting arrangement (if not defined separately in the Mounting screen) you should also consider whether the rotor hub, gearbox low speed stage and/or yaw bearing roll present significant additional torsional flexibilities about the low speed shaft axis.
Low speed shaft bending can be modelled in Bladed with two rigid massless shafts connected at a hinged bending point. Enter the total shaft length and the point of bending. The point of bending is entered as a percentage along the shaft length measured from the hub. So for example, a bending point of $0\%$ would mean the hinge flexibility is at the hub, a bending point of $100\%$ would mean the hinge flexibility is at the gearbox. The hinged connection is modelled as a 2D hinge, with flexibility about the hub $"\mathsf{y}^{\prime}$ and $"z^{\prime}$ axes. The bending stiffness and damping of the hinged connection should also be entered.
If required, a slipping clutch may be modelled. This is assumed to be at the generator end of the high speed shaft. The clutch slipping friction should be entered, and any additional stiction which has to be overcome before slipping begins. The high speed shaft inertia must be specified if the slipping clutch is to be used.
The dynamic drive train model may be used in combination with flexible mountings (see 4.9).
Note: including drive train flexibility may reduce the speed of simulations.
# 4.8.3 External DLL for transmission dynamics
Users with a licence for the Advanced Transmission Interface module may tick the “External DLL” checkbox to indicate that a user-defined DLL will be used to define the gearbox dynamics. Three options are available for the scope of the dynamics defined by the DLL:
Gearbox only: the DLL replaces the gearbox ratio in the Bladed model and has a single torsional degree of freedom only. Both the low and high speed shafts in the Bladed model must have torsional flexibility defined. If the shaft brake is defined then it is also considered to be within the scope of the DLL. Note that this affects not only the braking torque (see 5.12) but also the mechanical losses (see 4.11.1). Any flexible gearbox mountings must be modelled within the DLL if required.
Gearbox $+\ {\sf H S S}+$ generator rotor: the DLL replaces the gearbox ratio, high speed shaft and generator inertia in the Bladed model and has a single torsional degree of freedom only. The low speed shaft in the Bladed model must have torsional flexibility defined. If the shaft brake is defined then it is also considered to be within the scope of the DLL. Note that this affects not only the braking torque (see 5.12) but also the mechanical losses (see 4.11.1). The slipping clutch model is not available in this case – of course it can be modelled within the DLL if required. Any flexible gearbox mountings must be modelled within the DLL if required.
Gearbox $+\ {\sf H S S}+$ generator rotor (6 DOF): the DLL replaces the gearbox ratio, high speed shaft and generator mass and inertia in the Bladed model and has full six degrees of freedom. The low speed shaft and gearbox mountings in the Bladed model must have six degrees of flexibility defined. If the shaft brake is defined then it is also considered to be within the scope of the DLL. Note that this affects not only the braking torque (see 5.12) but also the mechanical losses (see 4.11.1). The slipping clutch model is not available in this case – of course it can be modelled within the DLL if required.
For the 6 DOF model it is also necessary to specify the length of the low speed shaft and the location about which the bending is assumed. This bending point is specified as a $\%$ of the shaft length from the hub side. The stiffness and damping parameters in all 6 degrees of freedom can be specified by clicking on the “LSS Stiffness / Damping Parameters” button.
On the mounting tab, the stiffness and damping parameters of the gearbox mounting in all 6 degrees of freedom have to be specified by clicking on the “Mounting Stiffness / Damping Parameters” button. In addition the dimensions of the gearbox need to be entered by specifying the location of the LSS relative to the centre of the mounting.
# For all options:
Within the $\mathsf{D L L},$ the dynamic equations of motion are specified in continuous time in state-space form:
$$
\begin{array}{l}{\dot{x}=A x+B u}\\ {\mathrm{~y}=C x+D u}\end{array}
$$
where $\times$ is a vector of state variables defined within the DLL, u is a vector of torques acting on the DLL shafts, and $\mathsf{y}$ is a vector of output variables. Matrices A, B, C and D need not be constant: in other words the gearbox model may be non-linear and time-varying, and discontinuous changes are allowed.
The DLL interface is specified in detail in Appendix C. The DLL may be written in any suitable language, using either of the two most common calling conventions: __cdecl, which the usual default for code written in C or $\mathsf{C}\!+\!+\!,$ , and __stdcall, which is the usual default for Fortran and Visual Basic; note however that these are only default settings and particular compilers may be different. The calling convention used must be specified on the Transmission screen, as well as the location of the compiled DLL.
User data: If desired, enter any User Data in the free-format field provided. This data will be placed in a text file whose location is passed to the DLL, so that the DLL may read this information if desired. This is useful for passing parameter values to the DLL which are also stored in the project file or the calculation details $(\Phi\mathsf{j})$ file.
# Mechanical loss modelling:
If an external dll is used to represent the gearbox, the mechanical losses as specified in the user interface are not applied. Instead, they are calculated and passed through to the dll where they should be applied appropriately. The calculation of these losses depends on whether the brake is defined or not in the powertrain transmission screen:
• No brake defined: the losses are calculated at the HSS (note that this assumes the gearbox ratio as specified in the user interface)
• Brake at LSS: the losses are calculated at the LSS Brake at HSS: the losses are calculated at the HSS (note that this assumes the gearbox ratio as specified in the user interface)
# 4.9 Drive train mountings
The Power Train button on the toolbar also allows the drive train mountings to be specified.
If desired, torsional flexibility may be specified in the gearbox mounting and/or between the pallet or bedplate and the tower top. This option is only allowed if the dynamic drive train model (see 4.8.2) is specified without the external user-defined DLL facility (see 4.8.3). If the external DLL facility is used then of course it can include its own flexible mountings.
In either case, specify the following parameters:
The torsional stiffness of the mounting.
The damping of the mounting.
In the case of a flexible gearbox mounting, the moment of inertia of the gearbox casing.
If desired, pallet or bedplate nodding stiffness can also be specified.
If either form of mounting is specified, the direction of rotation of the generator shaft will affect some of the loads. If the low speed and high speed shafts rotate in opposite directions, specify a negative gearbox ratio in the drive train (see 4.8) model.
Flexible mountings introduce a high frequency mode which is liable to result in much slower simulations. If the mountings are relatively stiff, for example if they are intended for noise isolation, then they will not influence loads significantly and are best omitted. If they are intended to modify the rotational dynamics, then they can be approximated by a corresponding reduction in shaft stiffness. This will have little effect on rotor and tower loads, but simulations will run faster.
# 4.10 Defining the Electrical Model
The Power Train button on the toolbar also allows the electrical model of the generator and power convertor to be specified. An electrical model must be defined if the dynamic drive train model (see 4.8.2) is specified.
Three catagories of electrical model are possible:
Simple (see 4.10.1) • Electrical Dynamics (see 4.10.3) External DLL (see 4.10.7) (User defined electrical dynamics model)
Simple models are often sufficient for modelling turbine performance and loading while the Electrical Dynamics models also simulates currents and voltages, active and reactive power flows etc., and can be used to calculate electrical flicker coefficients (see 8.18) and to model the effect of network voltage and frequency variations (see 7.20.2). To use the Electrical Dynamics models, a licence for the Electrical Dynamics module is required.
The external DLL option (see 4.10.7) allows a user-defined model for calculating the air-gap torque and electrical power output. This could be anything from a simple mechanical model to a comprehensive electrical transient model representing the electrical dynamics of the generator and also the power converter and electrical network if appropriate.
See also Defining the rotor (4.1), Control (5).
# 4.10.1 Simple electrical models
# 4.10.1.1 Fixed speed induction
This model represents an induction generator directly connected to the grid. It is characterised by:
• Rated slip: the slip speed at rated power as a percentage of synchronous speed. Rated power: the power output corresponding to rated slip. Synchronous speed: the rotational speed of the generator rotor at zero load. Generator time constant: this is the short-circuit transient time constant for the generator. Note: a small time constant may result in slower simulations. If it is very small, specifying a zero time constant will speed up the simulations, without much effect on accuracy.
The following table shows the synchronous speed for some typical cases:
| Fixedspeedpitchregulated | x=deviationofmeasuredpowerfromtheset-point y = pitch angle demand* |
| Variablespeedtorquecontrol (bothstallandpitch | X=deviationofmeasuredgeneratorspeedfromtheset- point |
| regulated) | y = generator torque demand |
| Variablespeedstallregulated power control | x=deviationofshaftpowerfromtheset-point y = generator speed set-point |
| Variable speedpitch | x = deviation of measured generator speed from the set- |
| controller | point |
| y = pitch angle demand* *Note: If a pitch rate actuator (see 5.10) is defined, the rate of change of this pitch angle demand is |
| usedastheinputtotheactuator. |
Various control design techniques are available to help with the selection of the gains $\mathsf{K}_{\mathsf{p}}$ and $\mathbf{K_{i}}$ . However the design of controllers is a specialist task. The main aim is usually to minimise deviations from the set-point without excessive control action and without causing any instabilities. A non-zero integral gain is important to ensure that the steady-state error is zero. This means that the measured quantity settles at the desired value in steady state conditions. The ratio $\mathbf{K}_{\mathbf{p}}/\mathbf{K}_{\mathbf{i}}$ is known as the integral time constant.
For very simple cases only, an automatic calculation of PI gains is available. Click the “Auto tune …” button after having first run a Steady Operation Loads calculation which generates the required aerodynamic characteristics. This may be useful in the early stages of design when the blade aerodynamic characteristics are defined but structural details are not yet available.
It is often useful or necessary to vary the overall gain of the controller depending on the operating point.
This is known as Gain scheduling - see 5.6.
Garrad Hassan will be pleased to undertake the design of any turbine controllers.
The desaturation time constant is used to prevent “integrator wind-up” when the output y is constrained by limits. If this is too long, the controller may take too long to break away from a limit. For example a pitch controlled machine crossing rated wind speed may take too long before the pitch starts acting. On the other hand a very short time constant may result in slower simulations. If the PI controller being modelled is actually implemented in discrete form, as is usual, then the desaturation time constant should be chosen to be somewhat smaller than the discrete controller timestep. Alternatively, specify a zero time constant for instantaneous desaturation.
# 5.6 Gain scheduling
The gains of a closed loop controller such as a PI controller (see 5.5) are designed to give good control (i.e. stable operation with good tracking of the set-point) at one particular operating point, for example at one wind speed. At a different operating point, the gains may need to be modified because the relevant characteristics of the system may be different. This is particularly true of pitch controlled turbines, where the aerodynamic gain changes significantly with pitch angle, and hence with wind speed.
The gain scheduling facility allows the overall gain of each PI controller to be modified as a function $\mathsf{f}(V)$ of some chosen variable $V_{\cdot}$ . Thus the actual proportional and integral gains at any point are given by $\upkappa_{\mathsf{p}}/\mathsf{f}(V)$ and $\kappa_{\mathsf{i}}/\mathsf{f}(V),$ where $\mathsf{K}_{\mathsf{p}}$ and $\mathbf{K_{i}}$ are the proportional and integral gains as entered in the boxes provided.
The variable $V$ is selected by clicking the arrow by Schedule on: and choosing from the list. The variables to choose from are:
Pitch [in radians] (for pitch regulated controllers only). Power [in W] Speed [in $\mathsf{r a d}/\mathsf{s}]$ (generator speed, for variable speed controllers only)
• Wind speed [in $\mathsf{m}/\mathsf{s}]$ (note that in practice, a suitable wind speed signal is unlikely to be available).
There are three possibilities for $\mathsf{f}(V)$ :
• Constant: the gains are fixed, with $\mathsf{f}(V)$ constant as defined by Value.
? Look-up table: $\mathsf{f}(V)$ is defined by a look-up table. Click LUT Data to open the table. Use Add to add points to the table, and double-click on an entry to edit or enter a value. The $\mathsf{X}$ values refer to the variable $V$ in the units shown, and the Y values are the corresponding $\mathsf{f}(V)$ . Polynomial: $\mathsf{f}(V)$ is defined by a polynomial. Click Coefficients to define it. Use Add to add terms to the polynomial, and double-click on an entry to edit or enter a value. Enter a coefficient for each term of the polynomial. $\mathsf{f}(V)$ is the sum of terms of $V$ (in the units defined above) raised to the power of the order $n$ , and multiplied by the coefficient entered for that term.
# Recommendations
For pitch regulated turbines, it is usually useful to apply gain scheduling to the above-rated pitch controller. This applies to both fixed speed and variable speed pitch regulated controllers. This is because the sensitivity of aerodynamic torque to changes in pitch angle is much greater at large pitch angles, and hence large wind speeds, than close to rated. The gain schedule may be defined as a function of pitch angle, and it is often appropriate, though conservative, to set f(pitch angle) to be
proportional to the partial derivative of aerodynamic torque with respect to pitch angle. Once the steady state control parameters have been defined, use the steady operating loads calculation to calculate the partial derivatives at each operating point, using the ‘Calculate pitch and speed change’ option.
For very simple cases only, if the automatic calculation of PI gains is used for the pitch controller, an appropriate gain schedule is also calculated at the same
Tip: in practice, for full-span pitch-regulated rotors pitching towards feather, setting f(pitch angle) to be proportional to pitch angle is often a reasonable first approximation, but this is unlikely to apply to partial span pitch or aileron-controlled rotors, or in the assisted stall case.
Gain scheduling is unlikely to be required for below rated controllers on variable speed turbines. For above-rated stall-regulated variable speed controllers, the need for gain scheduling will depend on the stall characteristics.
# 5.7 Auto tune for internal PI controllers
The tuning of control loops is a specialist task. To tune a PI controller it is necessary to select appropriate values for the proportional gain and the integral gain, and in the case of the pitch controller these gains will vary with the operating point according to a gain schedule. Classical tuning methods based on open and closed loop analysis of the system and its controller may attempt to achieve a particular cross-over frequency and stability margins for example. It is usually necessary to include a number of filters in series with the PI controller to achieve desired characteristics, and also to filter out signal noise, structural resonance frequencies, and forcing frequencies such as multiples of the blade passing frequency.
By ignoring all structural resonances and forcing frequencies it is possible to formulate an automatic calculation of PI gains. Bladed provides such a calculation, which may be appropriate for preliminary calculations at an early stage of turbine design when only the aerodynamic characteristics of the rotor have been defined. The calculation is only available for variable speed pitch regulated turbines with feathering pitch. First, carry out a Steady Operation Loads calculation, which generates the required aerodynamic characteristics. Then on the Control Systems screen, select the controller dynamics for a variable speed pitch regulated controller, and click one of the PI Control definition buttons. The “Auto tune…” button can then be used to launch the automatic calculation. You will be prompted to locate the results of the Steady Operational Loads calculation, and asked to confirm the required cross-over frequency, for which a default value of 1.0 rad/s is provided. In the case of the pitch PI controller, an appropriate gain schedule is also calculated.
The resulting controller is not likely to work effectively once any of the following effects are taken into account:
Tower flexibility
• Blade flexibility
• Drive train flexibility Turbulence
• Non-uniformity of the flow field
• Non-axial flows
• Sensor and actuator dynamics Noise and delays in control loop signals
Since these effects are important, a professional controller design is required for any detailed calculations.
# 5.8 Variable speed control below rated
Select Optimal tip speed ratio or Look-up table, and proceed as follows:
# 5.8.1 Optimal tip speed ratio
Below rated wind speed, a variable speed turbine may try to stay at its optimum tip speed ratio wherever possible, by changing the rotor speed in proportion to the wind speed. This maximises the power coefficient and hence the aerodynamic power available.
This can be achieved in the steady state by setting the generator torque to be proportional to the square of the rotor or generator speed. The Optimal mode gain multiplies the square of the generator speed to give the required generator torque demand. It can be calculated as:
$$
\mathbf{K}_{\mathrm{opt}}=\varpi\mathbf{R}^{5}\mathbf{C}_{\mathrm{p}}\,/\,2\lambda^{3}\mathbf{G}^{3}
$$
where
$\mathsf{K}_{\mathsf{o p t}}\,=\,0$ ptimal mode gain $\begin{array}{r l r l}{\uprho}&{{}}&{{}={}}\end{array}$ air density $\textsf{R}=$ rotor radius $\begin{array}{r l}{\mathsf{C}_{\mathsf{p}}}&{{}=}\end{array}$ power coefficient at $\uplambda$ $=$ desired tip speed ratio G $=$ gearbox ratio
The torque demand is then given by
$$
\mathrm{Q_{d}}=\mathrm{K}_{\mathrm{opt}}\Omega^{2}
$$
where
$\mathrm{Q}_{\mathrm{d}}=\$ generatortorquedemand $\Omega\!=$ measuredgeneratorspeed
Note that energy output may not be maximised by maximising aerodynamic efficiency, because the energy losses may also vary with the operating point. It may therefore be better to track a slightly different tip speed ratio. The desired tip speed ratio and the corresponding power coefficient should be used in calculating the Optimal mode gain as above.
See Automatic calculation of optimal torque-speed curve.
Alternatively the torque-speed curve may be expressed as a look-up table to account for losses or any other effects
# 5.8.2 Look-up table
Click Data, and enter a look-up table of generator torque demand against generator speed, using the Add button to add entries to the table. The View button allows the curve to be visualised.
See Automatic calculation of optimal torque-speed curve.
# 5.8.3 Automatic calculation of optimal torque-speed curve
When operating in the peak region of peak power coefficient, the torque-speed curve can be calculated by means of a quadratic relationship which can be calculated from the turbine parameters once the peak power coefficient and the corresponding optimum tip speed ratio are known.
First use the Performance Coefficients calculation to calculate the $C P-\lambda$ curve for the optimum fine pitch angle. If this is not known, repeat the calculation for different pitch angles to find the angle giving the highest peak CP.
Then on the Control Systems screen with “Variable Speed” selected, click the ‘Calculate …’ button in the below rated section to perform the calculation of the optimal mode gain. The result will either replace the existing value or be used to construct a look-up table if that option has been selected.
You will be prompted to confirm the air density and to locate the Performance Coefficients calculation results.
# 5.8.4 Other parameters
In addition to the above, specify the Minimum generator speed and the Optimal mode maximum speed (this must be no greater than the Maximum generator speed).
The controller will adjust torque demand as above using the optimal mode gain or the look-up table provided the generator speed lies between the Minimum generator speed and the Optimal mode maximum speed. If either of these speed limits is reached, it is used as the speed set-point for the Speed Control by Torque Demand PI controller. The deviation of measured generator speed from this set-point is the input to the controller, and generator torque demand is the output. Thus the generator torque is varied to keep the generator speed at the limit.
Note: This PI controller remains active while the speed is between the limits, but the torque demand output is constrained to lie on the specified torque-speed curve.
In addition to the Optimal mode gain, the Optimal mode maximum speed must be specified. This must be no greater than the Maximum generator speed. The controller will adjust torque demand as above to maintain the desired tip speed ratio provided the generator speed lies between the Minimum generator speed and the Optimal mode maximum speed. If either of these speed limits is reached, it is used as the speed set-point for the Speed Control by Torque Demand PI controller (see 5.5). The deviation of measured generator speed from this set-point is the input to the controller, and generator torque demand is the output. Thus the generator torque is varied to keep the generator speed at the limit.
Note: This PI controller remains active while the speed is between the limits, but the torque demand output is constrained to lie on the quadratic curve defined by the Optimal mode gain.
Note also that energy output may not be maximised by maximising aerodynamic efficiency, because the energy losses (see 4.11) may also vary with the operating point. It may therefore be better to track a slightly different tip speed ratio. The desired tip speed ratio and the corresponding power coefficient should be used in calculating the Optimal mode gain as above.
Further details are given in the Theory Manual.
# 5.8.5 Control in the variable slip case
With a variable slip generator, it is sufficient to set a high torque demand below rated: the generator itself will restrict operation to the nominal slip curve. This can easily be specified using the look-up table option.
# 5.9 User-defined controllers
Although Bladed has a full set of built-in controllers, both for power production control and for supervisory control, a wide range of different control algorithms is used in practice by different turbine manufacturers. The controller details can have a profound effect on turbine loads and performance, so it is important to allow the flexibility to use any control logic the user wishes to define. Also it is worth considering whether the same code can be used for controlling Bladed simulations as is used in the actual turbine controller.
Bladed allows a user-defined controller to be used for any of the following tasks:
Control of blade pitch and generator torque across the whole range of operation, including power production, normal and emergency stops, starts, idling and parked conditions. Control of the shaft brake and generator contactors. • Control of nacelle yaw.
You may use a mixture of user-defined and built-in controllers for the various different controller functions.
Note that the user-defined controller operates on a discrete timestep, like most real controllers. The built-in controllers operate in continuous time, but can be used to approximate the behaviour of a discrete controller as long as its timestep is not too long.
The user-defined controller may be written in any language, as a 32-bit DLL (dynamic link library).
The user-defined controller is ignored in the case of Hardware Test (see 7.17) simulations.
Bladed versions prior to 4.4 used a “swap array” to pass data between Bladed and the external controller. This has been replaced in version 4.4 onwards with a new function style of data communication, making use of “Get” and “Set” functions. The old swap array method is still supported but will not be extended further. Details of the new method can be found in Appendix A with further details available to download from the Bladed website. Details of the old “swap array” method are available to download from the Bladed website.
# 5.9.1 Using a user-defined controller
To use your own controller for any of the control functions, select the External Controller option where appropriate on the Control Systems or Yaw Control windows.
To specify your controller, select the External Controller tab on the Supervisory Control window; on the Control Systems window, Define... buttons are provided for this in appropriate places. Then specify the following information:
Controller code: the compiled.exe or.dll file which runs your controller (a dialogue box allows you to select the appropriate file)
Communication Interval: the controller timestep. At these regular intervals, Bladed will send information to the external controller, and expect to receive information in return. In real life the controller will take some time to calculate its outputs, usually a whole timestep, so it may be appropriate for the controller to always return the results of the previous timestep’s calculation. Noise on measured signals: if desired, click Specify noise to specify any random noise and discretisation errors (see 5.10.3) on the signals passed to the external controller. External Controller Data (optional): The information you provide here will be placed in a text file DISCON.IN which your controller is free to open and read when it starts up (no path is required as the file will be in the Bladed directory, which is the current directory when the external controller starts. You can use this to pass additional parameters to your controller, e.g. gains, look-up tables, etc. There is no need to pass the set-points etc. which are defined in the steady-state part of the Control Systems window, as these will be passed across in the communication file or array. (Note: to enter a tab in this field, press Control $^+$ Tab).
# Important note:
When using a user-defined controller, Bladed evidently cannot guarantee that the performance defined by the steady-state control parameters will actually be achieved on average during a dynamic simulation.
# 5.10 Simulation of Measured Signals
Bladed provides the facility to mimic measured signals – that is signals that are measured using a physical transducer. This is of particular relevance to controller design, when the input signals upon which the controller relies (such as the blade pitch angle) may suffer from noise, lag, discretisation or specific faults.
Any signal available to the external controller which represents a value coming from a physical transducer can have the following properties defined:
1. Noise
2. $1^{\mathrm{st}}$ or $2^{n\alpha}$ order transducer behaviour
3. Discretisation of value (e.g. rotor speed discretised to 0.1 rpm)
4. Sampling period (i.e. discretisation in time)
5. Various inbuilt or time-dependent faults
Signals which can have these properties defined can be identified in the external controller API as they begin “GetMeasured…” (see APPENDIX A Communication Between Bladed And External Controllers for details of how to write an external controller). In the user interface, these signals can be viewed by clicking 'Measured Signals' in the Control Systems window. For each of the transducer signals listed, it's possible to select any of the following signal quality values:
Simulated Value With Transducer Lag As Measured
The external controller will receive values exactly as they occur in the simulation.
Transducer behaviour is added to the signal, along with any faults.
Noise, discretisation, transducer behaviour, and faults are all added to the signal.
# 5.10.1 Transducer Behaviour
Most physical sensors have some form of time-dependent behaviour. Bladed allows the user to simulate both a simple first order lag and second order behaviour:
Note that even if transducer behaviour is defined, it will not be added to the signal if the signal quality is set to “Simulated Value”.
# 5.10.2 Noise
Uniform (rectangular distribution) or Gaussian noise can be added to the signal. In order to be reproducible, a random number seed can be provided. For a rectangular distribution, specify the halfwidth of the distribution (i.e. the maximum error). For a Gaussian distribution specify the standard deviation.
Noise is only applied if the signal quality is “As Measured”.
Note that the noise (the only random element of the measured signal) can be turned off for all measured signals at once with an option on the main signal properties screen.
# 5.10.3 Discretisation
The discretisation step for each signal may also be specified – the signal is discretised after any random noise has been added.
Discretisation is only performed if the signal quality is “As Measured”.
# 5.10.4 Sampling Period
A constant sampling period can be defined. This represents a signal whose value is only updated periodically – and the transducer will return the last sampled value until it takes another sample.
Periodic sampling is only added if the signal quality is “As Measured”.
# 5.10.5 Faults
In order to create robust controllers, Bladed allows the user to replicate a number of faults occurring on various signals upon which it may rely. These can either occur on the first instance of this type of controller, or on all controllers of this type.
Faults are applied if the signal quality is “With Transducer Lag” or “As Measured”.
# 1.1.1.1 Constant Faults
These faults represent manufacturing, assembly or installation faults with the sensor.
ds? If true, the sensor will always return the negative of its ‘true’ value. This is to represent a sensor that has been installed incorrectly.
ion This is a constant value offset, to represent a faulty sensor or one that has been installed incorrectly.
# 1.1.1.2 Time-Dependent Faults
Time-dependent faults occur after a specified amount of time. A note is reported in the message file when the error is activated.
# Change in Noise Characteristics
Once triggered, this will multiply the noise element of a measured signal by a specified amount. If there is no noise defined, this fault will have no effect.
Constant Fail Value
Once triggered, the sensor will return a specified fixed value. A ramp rate to this value can also be specified, in which case the value will linearly ramp between its last ‘good’ value and the constant fail value.
Constant Miscalibration Increasing Miscalibration
Once triggered, a constant offset is applied to the signal.
Once triggered, the signal’s value drifts away from its ‘true’ value. The offset increases linearly over time, but the signal will still vary underneath. Note that there is no limit on the value – it can continue increasing or decreasing indefinitely.
# Harmonic Interference
Once triggered, a sinusoidal element is added to the signal.
The properties of this interference can be defined.
# 5.11 The Pitch Actuator
The dynamics of the pitch actuator are an important part of the control loop dynamics for pitch controlled turbines, and must be defined before dynamic simulations can be carried out for these turbines. Click Control on the main toolbar followed by Pitch Actuator, or use the Specify $->$ Control systems pull-down menu. (For aileron or air-brake control, the pitch actuator is actually an aileron or air-brake actuator.)
Pitch actuator models can be defined as either passive or active. In the case of passive models, the motion of the pitch freedom is prescribed to follow the output of a Laplace transfer function. In an active model, an actuator torque is calculated using one or two PID controllers, and this is then applied to the pitch freedom.
The input demand may be defined as a position or rate demand. This must match with the controller’s behaviour. The user can choose to filter the incoming demand by the use of setpoint trajectory planning (also known as ramp control). This enables step demands (coming for example from a discrete external controller) to be smoothed according to rate and acceleration limits. Furthermore, the rate and acceleration limits will limit the demand signal over multiple controller time-steps, which may be the case if the setpoint trajectory rate and acceleration limits are within the controller rate and acceleration limits.
Response Types
Use setpoint trajectory planning – This uses the acceleration and rate limits of the setpoint trajectory planning to prescribe the pitch freedom. This can only be used when acceleration limits have been defined, otherwise infinite acceleration would be possible and the actuator torque would be undefined.
• $\pmb{\Omega}^{\mathbf{st}}$ order passive $-~1^{\mathrm{st}}$ order passive dynamics are represented by transfer function.
• 2nd order passive – 2nd order passive dynamics are represented by a transfer function.
· Other passive – The user may enter a custom Laplace transfer function. Closed Loop PID o Position Response: PID dynamics convert the position demand into a rate demand. o Rate Response: PID dynamics convert the rate demand into a torque or force demand, depending on the whether the actuator drive type is rotary or linear respectively.
There are three types of position limits:
• The Standard limit switches define the minimum and maximum pitch demands. When the Safety limit switches are tripped, the pitch motor is disengaged so that a rotary actuator drive provides no actuator torque and a linear actuator drive provides no actuator force. The brake is also applied at this time. If the brake has not been defined by the user, an internal brake is used by the calculation. The brake torque will appear in the pitch bearing friction output. The end stops model a hard physical stop on the bearing and provide a stiffness force if the pitch angle has exceeded the end stop position.
The bearing friction model calculates friction based on a constant value plus components based on the magnitudes of the bending moment and axial and radial forces at the blade root. The additional components of the friction value may be defined proportionally to the loads, or as look-up tables for a more detailed model. Stiction is also defined, which is an additional friction torque acting when the pitch is stationary.
The user may choose to not define the actuator drive details, or define them as rotary or linear:
• A rotary actuator system models a system of a pitch motor with a gearbox between the motor and blade ring. The user is able to specify the torque of a pitch brake, which is applied when a limit switch is activated. Torque limits are defined on the motor side of the gear ratio. These can be defined as fixed limits, as a look-up table according to pitch angle (blade side) or as a look-up table according to motor rate (motor side). A linear actuator models a system where linear motion is converted into rotary pitching motion by the use of a ram. The pitch angle at the point of maximum torque is the value of the pitch angle when the ram is perpendicular to the radius from the pitch axis to the connection point of the ram. If the user selects no actuator drive details, the drive is implemented as a rotary system with a unity gear ratio and no inertia.
# 5.11.1 Safety System Pitch Action
Two principle types of pitch action during a safety system event are available. For both of these, the user is able to define torque or force limits (depending on whether the actuator drive is rotary or linear respectively). These limits override the limits that are used in normal operation. If the user defines no torque or force limits in the safety system definition, the normal operation torque or force limits will not be used.
• In a rate demand safety system, the rate demand may be calculated from one of four methods, and this will override the demand from the controller. If the rate is to be passed by the external controller, the external controller must pass the demand by a specific function in the function API. In a torque demand (or force demand for a linear actuator) safety system, the torque or force demand can either be calculated by the external controller, or by the application of a constant torque plus torque terms dependent on the pitch angle (spring torque) and pitch rate (damper torque). The torque or force demand is applied directly to the actuator drive so must be defined on the motor side of the gear ratio in a rotary actuator.
# 5.12 The shaft brake
Click the Brakes button, either on the Transmission screen or in the Supervisory Control section of the Control Systems screen, to define the dynamic characteristics of the shaft brake.
The mechanical shaft brake is used for simulations of stopping or parked rotors. Up to three shaft brake characteristics may be defined, e.g. for different torque settings, or if different numbers of callipers are used. Shaft brake 1 is applied during the in-built normal (see 5.14) and emergency stop calculations (see 5.15), and at the start of a parked simulation (see 5.17). The other brake definitions can be applied through the external controller (see A.1 The Bladed External Controller API) or by means of the safety system (see 5.19).
When the brake is applied, it usually takes a little time to reach full torque. Two ways are provided for specifying how the brake torque develops. Select Linear Ramp or Look-up Table by clicking the appropriate button:
# Linear ramp model
The brake torque increases linearly from zero up to full braking torque. Enter the following data in the boxes provided:
Maximum shaft brake torque: the full braking torque, and Shaft brake ramp time: the time to reach full torque starting from zero.
# Look-up table
The braking torque may be specified as a time history of braking torque against time since the brake application was initiated. Use the Add button to add points to the time history.
See Drive train (4.8.2) to specify the position of the brake. See also Mechanical Losses (4.11.1), which are assumed to result in an additional braking torque.
See also: Normal stop sequence (5.14), Emergency stop sequence (5.15), Parked conditions (5.17).
# 5.13 Start-up sequence
For simulations of start-up, click the Start button on the Control screen and specify the following parameters:
• Initial rotor speed: the rotor speed at the start of the simulation. Initial pitch angle: the pitch angle at the start of the simulation.
The built-in start-up logic may be used, in which case the following parameters should be entered to define the sequence of events:
Initial pitch rate during start-up: the constant pitch rate which will be used during the start-up sequence.
Generator speed at which generator is put on line: As soon as the generator speed reaches this value, the generator is connected to the grid and the power production control (see 5) starts to act. However any pitch angle demanded by the controller will be ignored until the initial pitch ramp is complete, as defined by:
Final pitch angle in start-up mode: The Initial pitch rate during start-up continues to act until this pitch angle is reached, at which point any pitch angle demanded by the power production controller starts to be used.
Alternatively, click Defined by external controller to use your own start-up logic.
# Notes:
1. The sign of the Initial pitch rate must be consistent with the initial and final pitch angles: if the final pitch angle is less than the initial pitch angle, the pitch rate must be negative, and vice versa.
2. Pitch angles and rates apply also to aileron/flap/airbrake deployment angles. They are ignored if the rotor has no aerodynamic control surfaces (see 4.1).
# 5.14 Normal stop sequence
Normal stops use aerodynamic braking and/or the shaft brake to stop the rotor from the normal running condition. For simulations of normal stops, click the Normal stop button on the Control screen. To use the built-in logic for normal stops, specify the following parameters:
• Pitch rate: the constant pitch change rate used during the stop sequence.
? Maximum pitch: the final pitch angle. If the Pitch rate is negative, the Maximum pitch should actually be the most negative pitch angle. The initial pitch angle will depend on the running conditions before the stop occurs. Rotor speed for cut-in of shaft brake: when the rotor has slowed down to this value, the shaft brake (see 5.12) is applied to bring the rotor to rest.
The normal stop begins at the Time to begin a stop specified in Simulation control (see 7.14).
Alternatively, click Defined by external controller to use your own normal stop logic. Note that the Time to begin a stop (see 7.14) does not apply in this case.
# Notes:
1. Pitch angles and rates apply also to aileron/flap/airbrake deployment angles. They are ignored if the rotor has no aerodynamic control surfaces (see 4.1). 2. User-defined supervisory control may also be defined in an external controller (see 5.9).
# 5.15 Emergency stop sequence
The emergency stop calculation models the complete shutdown of a turbine by the safety system in the event of a loss of load due to an unexpected grid loss. Modern turbines attempt a “ride through” in the case of grid loss, so this calculation is rarely used in loads calculations unless it is for a legacy project. See section 5.19 for the functionality to model a modern safety system.
If this calculation were to be used, it models the safety system’s actions to bring the rotor to a full halt. A traditional safety system of this type will attempt to stop the rotor by feathering the blades to utilise aerodynamic braking and/or the use of a brake within the drivetrain. The triggering of these actions can be either directly on the occurrence of a grid loss (the “Grid Loss” trip mode), or by the increase of the rotor speed beyond a certain threshold (the “Overspeed” trip mode).
For simulations of emergency stops, click the Emergency stop button on the Control screen. To use the built-in logic for emergency stops, specify the following parameters:
• Emergency pitch trip mode: Select Overspeed if aerodynamic braking is to be triggered by an overspeed, or Grid loss if it is to be triggered immediately on loss of load. Rotor speed at which pitch motion commences: The overspeed condition which triggers the aerodynamic braking if Emergency pitch trip mode is set to Overspeed. Emergency pitch rate: the constant pitch change rate which is used following the Grid loss or Overspeed.
• Maximum pitch: the final pitch angle. If the Emergency pitch rate is negative, the Maximum pitch should actually be the most negative pitch angle. The initial pitch angle will depend on the running conditions before the stop occurs. Emergency shaft brake trip mode: Select Overspeed if the shaft brake is to be triggered by an overspeed, or Grid loss if it is to be triggered immediately on loss of load. Rotor speed for cut-in of brake (speed control): The overspeed condition which triggers the shaft brake if Emergency shaft brake trip mode is set to Overspeed. Rotor speed for cut-in of brake (for parking): when the rotor has slowed down to this value, the shaft brake is applied to bring the rotor to rest if it has not already been applied as a result of grid loss or overspeed..
The emergency stop begins at the Time to begin a stop specified in Simulation control (see 7.14).
Alternatively, click Defined by external controller to use your own emergency stop logic. Note that the Time to begin a stop (see 7.14) does not apply in this case.
# Notes:
1. Pitch angles and rates apply also to aileron/flap/airbrake deployment angles. They are ignored if the rotor has no aerodynamic control surfaces (see 4.1). 2. User-defined supervisory control may also be defined in an external controller (see 5.9).
# 5.16 Idling conditions
For simulations of an idling rotor, click the Idling button on the Control screen and specify the fixed pitch or aileron angle to be used while idling. Click Enable external controller if any user-defined control logic is to be used.
# 5.17 Parked conditions
For simulations of a parked rotor, click the Parked button on the Control screen and enter:
• the fixed pitch or aileron angle to be used when parked, and the rotor position or azimuth angle when parked (note that this position is used at the start of the simulation, but it is possible for the brake to slip, allowing the azimuth to change during the simulation). A zero azimuth angle means that the first blade is pointing vertically upwards.
Click Enable external controller if any user-defined control logic is to be used.
# 5.18 Yaw control
# 5.18.1 Yaw Dynamics
Four yaw dynamics options are available. Click the Yaw control button on the Control Systems window and select the appropriate option:
None: the nacelle direction will remain fixed.
Rigid yaw: the nacelle follows the demanded nacelle angle (see 5.18.2) precisely.
Flexible yaw: the nacelle yaws passively with respect to the demanded nacelle angle (see 5.18.2) as a result of aerodynamic forces.
Controlled torque: the yaw actuator torque is specified by the external controller (see 5.18.2).
If any yawing is allowed, specify the Yaw friction torque
If Flexible Yaw or Controlled torque is selected, you may specify any
Additional stiction. The nacelle will only start to yaw if the balance of applied yawing torques at the yaw bearing bearing exceeds the friction plus the stiction torque, and it will stop yawing if the balance of torques falls below the friction torque. The applied torques include aerodynamic and inertial Mz torques and either the spring/damper torque (in the case of Flexible yaw) or the Controlled torque.
If Flexible Yaw is selected, specify the:
Yaw Damping (if any), and either the
Yaw Stiffness (if any) for a linear spring model, or select the hydraulic accumulator model.
Note that using the external controller, it is possible to control the yaw damping and stiffness and also to provide additional friction by applying a yaw brake torque.
The following parameters are required to model yaw compliance provided by a hydraulic accumulator system:
Gas volume (V): the volume of gas in the accumulators
Nominal pressure (P): the equilibrium pressure in the hydraulic system
Pump flow per unit yaw (F): the volume of fluid which must pass through the yaw motor to achieve a one radian change in nacelle angle
Yaw torque per unit pressure (Q): the relationship between the pressure difference across the yaw motor and the torque developed at the yaw bearing
Gas law constant: the constant $\upgamma$ for the gas in the accumulator. In the gas law equation,
$\mathrm{PV}^{\gamma}=\mathbb{R}\mathrm{T}$ Specify $\gamma=1$ for isothermal conditions.
The yaw torque $\mathbf{v}$ provided by the hydraulic system is then given by:
$$
\mathbf{Y}=\mathbf{Q}.\mathbf{P}\left({\left[{\frac{\mathbf{V}}{\mathbf{v}_{1}}}\right]}^{\gamma}-{\left[{\frac{\mathbf{V}}{\mathbf{v}_{2}}}\right]}^{\gamma}\right)
$$
where $\mathsf{v}_{1}=\mathsf{V}\mathrm{~-~}\mathsf{F}_{\cdot}\mathrm{{a},~}\mathsf{v}_{2}=\mathsf{V}+\mathsf{F}_{\cdot}\mathrm{{a},~}$ and $\mathfrak{a}$ is the difference between the actual and the demanded nacelle angle.
# 5.18.2 Active Yaw
Three choices are available for defining the demanded nacelle angle in simulations:
None: the demanded nacelle angle will be fixed at at the initial angle. The default is zero, i.e.
pointing North.
Prescribed manoeuvre: specify a yaw manoeuvre by entering: · Time to start yaw manoeuvre: when the demanded nacelle angle starts to move • Required yaw position change: the angle through which it will move • Yaw rate for yaw manoeuvres: the yaw rate to be used.
External controller: if Rigid yaw or Flexible yaw is selected (see 5.18.1), the user-defined controller will generate a yaw rate demand which defines the demanded nacelle angle at any instant. If Controlled torque is selected, the user-defined controller (see 5.9) will generate a yaw actuator torque demand. This will be the same as the actual yaw actuator torque, since the dynamic response of the yaw torque actuator is assumed to be fast, and is therefore not represented in the model.
# 5.19 Safety System
The safety system module allows a series of trips and subsequent actions to be specified to model the turbine safety system. The safety system consists of a number of hard-wired trips and relays, arranged in one or more fail-safe circuits which act independently of the turbine controller, and are intended to bring the turbine safely to rest in the event of a problem which the normal control system cannot handle.
To define the safety system from the Bladed toolbar, use Specify... Control Systems... Safety system.
# 5.19.1 Safety System Circuits
One or more safety system circuits can be defined. These circuits can be tripped by various safety system trips or by the external controller. For each circuit, specify the actions that take place when the circuit is tripped. These actions include fail-safe pitch action, disconnecting the generator, applying shaft brakes, and disconnecting the yaw drive.
# 5.19.2 Safety System Pitch Action
Two options are available for the safety system pitch action: Constant rate demand and Torque demand. These actions override the pitch demand from the controller. There are three possible components to the torque demand:
# Constant torque
Spring torque, which takes the form of a look up table of pitch angle to torque.
Damping torque, which takes the form of a look up table of pitch rate to torque.
In addition, a variable torque or force limit can be specified, as a function of the distance moved by the actuator since the start of the safety system pitch action. This can be used to represent, for example, the decrease in hydraulic pressure in a failsafe hydraulic actuator as the accumulator discharges.
# 5.19.3 Safety System Trips
Six different trips are available. For each trip, specify which safety circuit it will activate.
There are four sensor trips: rotor speed, generator speed, generator power after electrical losses and nacelle vibration, which uses the magnitude of the horizontal nacelle acceleration vector. Specify the sensor trip level and the delay before the circuit is activated.
There is also a generator short-circuit (see 7.20.2) trip, also with a specified delay, and an operator emergency stop button which can be activated at a specified time in the simulation.
Check Only activate safety system once output logging begins to prevent safety system activation during any transients at the beginning of the simulation before any output has been recorded.
# 5.20 LIDAR
LIDAR (light detection, imaging and ranging) is a laser Doppler anemometer system by which laser beams mounted on the turbine measure the wind field ahead of the turbine. This information can be passed to the controller, to implement a feed-forward control system. The instruments collect data by emitting light waves from transducers according to some pre-determined scanning pattern. Particles suspended in the air cause backscatter of the emitted waves. The Doppler shift in the frequency of the backscattered light is used to determine the wind speed along the direction of the LIDAR beam, at a particular focal point. This information can be logged in the control system and used to enhance the control strategy to provide load reduction.
The LIDAR definition can be accessed via Specify- $\cdot>$ Control Systems…->LIDAR. The LIDAR definition allows the specification of multiple beams that take velocity measurements at intervals and pass the data to the external controller. All the options in the LIDAR screen only affect the LIDAR definition and the information passed through to the external controller, and do not affect any other environmental or structural definitions within Bladed.
When using turbulent wind with a LIDAR sensor, the evolving turbulence feature can be used so that the wind variations measured by the LIDAR will change before reaching the turbine. This requires two wind files to be defined and is more realistic than the usual assumption of “frozen turbulence” (see Section 6.7).
# 5.20.1 Laser beam direction
This section explains how a user can define the LIDAR beam direction, relative to a mounting component, in the user interface. The LIDAR beam is specified using a LIDAR axis which describes the orientation of the LIDAR system relative to the mounting node. The beam direction is then defined relative to the LIDAR axis and this can vary depending on the scanning pattern; or remains fixed on the LIDAR axis if the fixed scan pattern is specified.
The user can add multiple LIDAR beams in the user interface. These can be mounted on the nacelle, hub or at a blade station. The coordinate system used depends on the component. The rotating hub coordinate system is used if mounted on the Spinner, the yaw bearing coordinate system is used if mounted on the nacelle and the root axes coordinate system is used if mounted on a blade station. A position offset and the nominal (undeflected) LIDAR axis is specified relative to the mounting component. The LIDAR axis is orientated by a rotational quaternion and is specified using two angles; the inclination angle $\theta$ and the azimuth angle $\psi$ . The transformation is carried out by first inclining the LIDAR axis about the y axis, then rotating about the $\times$ axis by the azimuth angle, as shown in Figure 5- 3.
Figure 5-3: Visualisation of the LIDAR axis defined in the mounting component coordinate system
The direction of each beam will change at run-time according to the motion of its mounting component, the LIDAR axis and the scanning pattern defined by the user or external controller. Static and dynamic deflections of the mounting component will also affect the LIDAR beam’s orientation and position. For instance, if mounted on the spinner, the LIDAR system will be inclined if the hub has a tilt angle. Likewise, if the nacelle is yawed, the LIDAR system will also be yawed accordingly and if mounted on the blade, the LIDAR system will rotate as the rotor azimuth angle varies.
A laser beam can be steered to scan an area in front of the turbine referred to as a scan pattern. The scan pattern describes the variation in the beam direction relative to the LIDAR axis with respect to time. The beam direction is defined by a spherical coordinate system defined using two angles $\alpha$ and $\beta,$ as denoted in Figure 5-4. The angles $\alpha$ and $\beta$ represent the inclination to and the azimuthal rotation about the LIDAR axis.
Figure 5-4: LIDAR beam direction defined relative to the LIDAR axis coordinate system
# 5.20.2 LIDAR scan patterns
The user can apply one of the pre-defined scanning techniques, fixed position, circular scanning or rosette scanning. If the fixed scan pattern is selected, then the beam direction is coincident with the LIDAR axis.
The number of samples per complete scan $N_{s},$ should be specified by the user along with the sampling rate $T_{s}$ . The period of each scan is then given by $T=T_{s}\cdot N_{s}$ .
If a scanning pattern is selected when there is more than one beam defined, the same scanning pattern is assumed for all beams. This limitation can be avoided, if the scan pattern is defined by the external controller.
For more details on the scan patterns that can be used, please consult the Theory Manual.
# 5.20.3 Line of sight wind speed calculation
At each sample interval $t,$ the LIDAR beam orientation is updated and then the wind module is interrogated to obtain the wind speed along the direction of the beam. This section outlines the calculation process that is used for each beam. The line of sight speed reported is relative to the motion of the mounting component.
The focal distance(s) can be set directly in the user interface, or by the external controller. This is the distance from the LIDAR instrument along the beam direction, at which the wind module is interrogated to obtain a line of sight wind speed.
In reality, a single precise wind speed at the focal point is not recorded, but rather the weighted average of velocities around the focal point according to a weighting function. For a pulsed beam, this can be specified by the user, as a look-up table. For a continuous beam, the weighting function is defined by the lens area and beam wavelength (see the Theory manual for further details).
If there is more than one LIDAR beam, then the LIDAR beam sampling can be made sequential, in which case a measurement is only taken from a single beam per LIDAR sample. The order upon which the readings are taken is determined by the order in which the lidar beams were inputted in the user interface (top to bottom is first to last). If simultaneous beam sampling is selected, then wind speed measurements will be taken from all beams on every LIDAR sample.
In order to compute the line of sight velocity reading for beam the following algorithm is used:
1. The beam direction is updated relative to the LIDAR axis as per the scanning pattern.
2. The orientation of the mounting component, the LIDAR axis and the beam direction are used to define the direction of the LIDAR beam in the inertial frame (global frame).
3. The focal position of the beam is computed (in the global frame) using the mounting location of the lidar instrument, the distance along the beam and the direction of the beam.
4. The wind velocity is computed by interrogating the wind module at the focal position, the velocity of the mounting component is subtracted from this value.
5. The dot product of the lidar beam direction and the wind velocity is computed to get the line of sight velocity.
6. If a weighting function is defined, then steps 3 to 5 are repeated at each distance from the focal point and the result is multiplied by the corresponding weighting. The product of the line of sight speed and weighting is summated and then normalised to compute a final line of sight velocity.
# 5.20.4 LIDAR and external controller interface
To make use of the LIDAR feature, the user will need to write an external controller DLL. By default, Bladed will not log any values from the LIDAR system. To output any states or measurements made by the LIDAR system, the relevant channels should be added to the External Controller outputs group. Please consult the external controller documentation regarding this feature. The relevant external controller commands that can be used are documented in the ExternalControllerAPI.chm file, which can be found in the Bladed installation directory. The sections below provide more details on the implementation of the external controller LIDAR commands.
# 5.20.4.1 Setting up LIDAR system using external controller
The external controller can be used to define the following LIDAR system attributes:
1. LIDAR range: distance to the focal point(s)
2. LIDAR beam angles: the scanning pattern
3. LIDAR sampling rate: interval between successive LIDAR measurements
If the sample rate is defined by the external controller, then it is recommended that the value specified be the same as or greater than the external controller time step. The external controller can also inspect some other properties of the LIDAR system, see the ExternalControllerAPI.chm file for further details. Some aspects of the LIDAR model such as the LIDAR type (pulsed/continuous), the weighting function, beam mounting component, offset and centreline orientation cannot be specified by the controller.
# 5.20.4.2 LIDAR outputs from external controller
The external controller is commonly used to output the LIDAR beam orientation and line of sight velocity readings. The orientation data is returned using the following coordinate system defined in Figure 5-5. The angle Y is the rotation about the y-axis between the negative $\times$ -axis of the LIDAR beam. Similarly, the angle Z is the rotation about the z-axis between the negative x-axis and the LIDAR beam. The line of sight velocity reading is the component of the measured wind speed, at a particular focal point, which is parallel to the LIDAR beam as defined by its measured Y and Z angles.
Figure 5-5: Visualisation of the angles Y and Z returned by external controller API calls.
# 6 DEFINING THE ENVIRONMENT
This section describes how to define the environmental conditions in which the turbine operates. The environmental conditions include wind (Section 6.1), sea state (Sections 6.12 to 6.14) and seismic conditions (Section 6.15).
# 6.1 Defining the wind
Click the Wind icon on the toolbar to define the characteristics of the wind field.
For the calculation of steady state parked loads and for all simulations, define the steady-state spatial distribution of the wind. For simulations, define also the time-varying wind characteristics.
The steady-state characteristics of the wind field may include any combination of the following elements:
Wind shear (see 6.2): the variation of wind speed with height. • Tower shadow (see 6.3): distortion of the wind flow by the tower. Upturbine wake (see 6.4): full or partial immersion in the wake of an upstream turbine.
For simulations, wind speed may vary with time in addition to the steady-state spatial characteristics defined above. Click on Time Varying Wind and choose one of the following options:
No variation (see 6.5): for simulations in which the wind speed and direction are constant in time. Single point history (see 6.6): to supply a time history of wind speed and direction which is fully coherent over the whole rotor. 3D turbulent wind (see 6.7): use a 3-dimensional turbulent wind field with defined spectral and spatial coherence characteristics. See Generating turbulence fields. Transients (see 6.8): to use sinusoidal wind speed and direction transients as defined by certain standards, such as IEC 1400-1.
These options are described below, starting at Section 6.5. In each case the wind characteristics are defined at a specified height, and the wind speed at other heights will be modified according to the shear profile. Click Refer wind speed to hub height to ensure that the definitions always apply at the hub height of the turbine.
Click on View wind data for a graphical representation of the wind field. The options available include:
Time history plots for selected positions in the rotor plane 3D carpet plots and animations showing the wind field over the whole rotor plane
The variables which may be displayed include:
Longitudinal, lateral and vertical wind speeds Resultant horizontal wind speed Total resultant wind speed Wind direction and upflow angle.
This data may also be tabulated as for other graphical data (see 9.9).
The time history plots optionally include moving-averaged values if a gust average period is specified. Click Stats to see the mean wind speed and the maximum and minimum values of the wind speed and the moving-average gust wind speed. Click Max Gust Table to view a table of maximum movingaverage gust values at a grid of points, after first specifying the spacing of the grid points required. Note that this calculation may be quite slow.
# 6.2 Wind shear
Wind shear is the variation of steady state mean wind speed with height. If wind shear is required, click the Wind icon on the toolbar and select Wind shear. Then select either the exponential or the logarithmic wind shear model, or provide a look-up table to specify a user-defined shear profile.
# Exponential model:
Enter the wind shear exponent. The wind speed $\mathsf{V}(\mathsf{h})$ at height h above the ground is then given by:
$$
\mathbf{V}(\mathbf{h})=\mathbf{V}(\mathbf{h}_{0})\left({\frac{\mathbf{h}}{\mathbf{h}_{0}}}\right)^{\alpha}
$$
where
${\mathfrak{h}}_{0}$ isthereferenceheight,and $\mathfrak{a}$ isthewindshearexponent.
A zero exponent results in no wind speed variation with height.
# Logarithmic model:
Enter the ground roughness height parameter. The wind speed V(h) at height h above the ground is then given by:
$$
\mathrm{{V}(h)={\bf V}(h_{0})\!\left(\frac{\log(h/z_{0})}{\log(h_{0}/z_{0})}\right)}
$$
where
$\mathfrak{h}_{0}$ isthereferenceheight,and $\mathbf{Z}_{0}$ isthegroundroughnessheight.
# User-defined profile:
Enter a lookup table giving a wind speed multiplication factor as a function of height. The multiplication factor would normally be 1.0 at the reference height.
# 6.3 Tower shadow
Tower shadow defines the distortion of the steady-state mean wind field due to the presence of the tower. Click the Windicon on the toolbar and select Tower shadow, and select a model.
# Potential flow model:
This model is appropriate for rotors operating upwind of the tower. Enter the Tower diameter correction factor, $F_{\mathrm{.}}$ . The longitudinal wind velocity component around the tower is modified using the assumption of incompressible inviscid flow around a cylinder of diameter $F.D_{\prime}$ , where $D$ is the tower diameter at the height where the tower shadow is being calculated.
# Empirical model:
For rotors operating downwind of the tower, use this empirical model, which uses a cosine bell-shaped tower wake. Enter the following parameters:
• Maximum velocity deficit at the centre of the wake as a fraction of the local wind speed. Width of the tower shadow as a fraction of the local tower diameter. Reference position: the distance downwind, as a proportion of the local tower diameter, at which the above parameters are defined. At other distances, the shadow width increases and the velocity deficit decreases with the square root of the distance from the tower.
# Combined model:
This model uses the potential flow model at the front and sides of the tower, and the empirical model on the downwind side wherever it gives a deficit greater than the potential flow model deficit, with a smooth transition between the two. This is useful in simulations where the rotor might yaw in and out of the downwind shadow area.
# 6.4 Upwind turbine wake
There are two ways to model a wake originating from an upwind turbine in Bladed:
The ‘Upwind turbine wake’ tab in the ‘Wind’ screen (to be deprecated)
The ‘Dynamic upwind wake’ screen in Specify- $>$ Wind
From these options, the latter is recommended for site-specific calculations as the user-inputs in general do not need to change between runs in a set of site-specific load calculations, because they are independent of operating condition. The former has several user entries relating to the upwind turbine (e.g. tip speed ratio), which are dependent on operating condition. This formulation will be deprecated in a future release.
In order to define simple, static Gaussian wake, this must be done in the ‘Upwind turbine wake’ tab, where the user must enter:
• Centre line velocity deficit: as a percentage of the local wind speed. Wake half-width: the distance from the wake centre line at which the deficit is reduced to $\mathsf{e}^{-0.5}$ times the centre line value.
The ‘Dynamic upwind wake’ screen allows the user to generate a wake at the start of the simulation based on an eddy-viscosity model and also meander the wake centreline throughout the simulation.
The wake deficit profile is calculated at the start of the simulation and is constant throughout. The user must provide a steady operational loads relevant to the upwind turbine over all operating points. The initial wake profile is calculated from the axial induction factors of the aerodynamic information group in these steady operational loads results at the wind speed that corresponds to the mean wind speed of the current simulation. It is therefore important that aerodynamic information is output at every blade station otherwise the initial wake profile will be very coarse. This information can also be obtained from an aerodynamic information simulation but then the information will not be valid for multiple wind speeds.
The initial wake profile is propagated downstream according to the thin shear layer approximation of the Navier-Stokes equations, which is documented in the theory manual. The distance of propagation and the eventual centreline offset at the modelled turbine is determined by the relative global location (North and West) of the upwind turbine and the simulation mean wind direction, which is shown as an example in Figure 6-1.
Figure 6-1: Propagation of a wake profile from upstream turbine to modelled turbine
There is also the option for a meandering wake where the wake centreline meanders through the simulation. The motion of the wake is calculated by integrating the lateral and vertical components of a user-specified turbulence file. This turbulence file should be low-pass filtered in order to avoid high frequency content in the meandering. The speed of wake propagation relative to ambient wind speed affects the lateral and vertical meander speeds from the file and also affects the length of axis over which the meander is integrated. Therefore, a slower speed of propagation will generally lead to larger meander.
# 6.5 No time variation of wind speed
Use this option to specify the wind conditions for a simulation when no variation of the wind field with time is required. Any steady-state spatial variations defined by wind shear, tower shadow and an upturbine wake will apply.
Click the Wind icon on the toolbar, select Time varying wind, and choose the No variation option.
Enter the following parameters:
• Wind speed: the steady wind speed to be used in the simulation. Height at which speed is defined: the reference height to which the wind speed applies, unless Refer wind speed to hub height has been selected. If wind shear (see 6.2) is defined, the wind speed at any other height will be different. Wind direction: measured clockwise from North - see wind direction (see 6.9). Flow inclination: for non-horizontal flows (e.g. on the side of a hill). A positive value indicates a rising flow.
# 6.6 Single point wind history
Use this option to specify the wind conditions for a simulation when an arbitrary time-varying wind speed and/or direction is required, but no spatial variation is required other than the steady-state characteristics defined by wind shear, tower shadow and an upturbine wake.
Click the Wind icon on the toolbar, select Time varying wind, and choose the Single point history option.
Enter the following parameters:
Height to which speeds relate: the reference height to which the wind speed time history applies, unless Refer wind speed to hub height has been selected. If wind shear (see 6.2) is defined, the wind speed at any other height will be different.
Flow inclination: for non-horizontal wind flows (e.g. on the side of a hill). A positive value indicates a rising wind.
Then use the Add button to add points to the time history. For each point, enter
Time: this should start from zero.
Speed: the wind speed at that time.
Wind direction: measured clockwise from North - see wind direction (see 6.9).
# 6.7 3D turbulent wind
Use this option to perform a simulation in which the turbine is immersed in a turbulent wind field which varies both in space and time. The turbulence is superimposed on any steady-state spatial variations defined by wind shear, tower shadow and any upwind turbine wake. The turbulent variations at any point in the rotor disk have defined spectral characteristics, and the variations at any two points in the rotor disk are correlated by a defined coherence relationship representative of the spatial structure of real atmospheric turbulence. The turbulent wind field may have just the longitudinal component of turbulence, or it may have all three components.
Click the Wind icon on the toolbar, select Time varying wind, and choose the 3D turbulent wind option.
Then enter the following information:
Turbulent wind file name: Click here to select a file which contains an appropriate turbulent wind field: see Generating turbulent wind fields (see 6.10). Click Properties to display the characteristics of the turbulent wind field. The windfield should be high enough and wide enough to envelop the whole rotor, and long enough to allow the simulation to continue for as long as is required. If $U$ is the average wind speed corrected to hub height according to the wind shear, and the simulation is to run for a time $T,$ the turbulence field length should be $U,T+$ Turbine diameter, or just $U.\tau$ if the turbulence file is allowed to wrap around (see below). It may be longer than this, although in this case the mean wind speed and turbulence intensity for the simulation may not match the values selected, as they apply only if the whole file is used. Mean wind speed: the mean wind speed for the turbulent wind at the reference height. If this does not match the wind speed for which the turbulence characteristics were defined, the turbulence field will be scaled appropriately. However, since the dimensionless characteristics of turbulence are not quite invariant with wind speed, this scaling is not strictly valid. Height at which speed is defined: the reference height to which the Mean wind speed applies, unless Refer wind speed to hub height has been selected. If wind shear (see 6.2) is defined, the mean wind speed at any other height will be different. Turbulence intensity: the standard deviation of turbulent wind speed variations as a percentage of the mean wind speed. If using a three-component wind field, specify the turbulence intensity for each component.
。 Wind direction: measured clockwise from North - see wind direction.
。 Flow inclination: for non-horizontal flows (e.g. on the side of a hill). A positive value indicates a rising wind. Allow turbulence file to wrap around: if this is checked, the turbulence file can be recycled indefinitely by looping from the end back to the beginning.
If a LIDAR sensor (see 5.20) is used to provide wind preview information to the external controller, the upstream wind velocities measured by the LIDAR cannot be expected to convect towards the turbine unchanged as implied by Taylor’s frozen turbulence hypothesis. Select ‘Evolving turbulence’ to allow the turbulence to change as it convects towards the turbine, and then provide the following additional information:
Additional wind file name: Select a second file containing a turbulent wind field, which should be identical to the first turbulent wind field apart from the random number seed - see Generating turbulent wind fields (see 6.10). While the turbulence reaching the turbine is still given by the first turbulent wind field, the turbulence measured by the LIDAR is calculated by combining frequency components from both wind fields depending on the upstream distance of the measurement point. Note that this can result in significantly slower simulations, especially if there are many simultaneous LIDAR measurement points and/or many points are defined for the LIDAR weighting function.
For the Height of turbulent wind field, select one of the following options. These do not affect the mean wind speed at any height, only the location of the turbulent wind fluctuations:
• Centred on hub height: The centre of the turbulent wind field is placed at hub height.
• Best fit for rotor and tower: If the turbulence field vertical dimension is less than the height of the turbine, the top of the turbulence field will be located at the height of the top of the rotor, so that the turbulence field extends as far down the tower as possible. If the turbulence field vertical dimension is greater than the turbine height, the turbulence field will start at ground level and will envelop the whole turbine.
Turbulent variations outside the defined turbulence field will be taken from the nearest defined point.
The turbulence is defined at a number of grid points (see 6.10), both in the rotor plane and in the alongwind direction. In between these points, either linear or cubic interpolation can be used to determine the turbulence. Specify one of the following options for the interpolation scheme:
Linear: uses linear interpolation
Cubic in rotor plane only: uses cubic interpolation laterally and vertically, and linear interpolation in the alongwind direction
Fully cubic: uses three-dimensional cubic interpolation.
Cubic interpolation for Lidar measurement: If ‘Evolving turbulence’ has been selected, this option specifies that cubic interpolation in the y-z plane will be used for both wind fields when calculating the wind velocity at any LIDAR measurement point, otherwise linear interpolation will be used. Note that evolving turbulence can result in slow simulations, in which case selecting cubic interpolation for the Lidar measurement would make them even slower.
If desired, specify also a superimposed sinusoidal wind direction transient. This is particularly useful if only a single turbulence component is being used. Enter the following parameters:
Amplitude of direction change: the peak-to peak amplitude of the transient. Note that the specified wind direction (see above) refers to the start of the simulation. The mean wind direction will be different from this if a transient is added. • Start time of transient: the time into the simulation at which the transient starts. • Duration of transient: the duration of the transient from the time it starts to the time it finishes. • Type of transient: half or full wave: see 6.8.
# 6.8 Transients
Use this option to define sinusoidal wind speed and direction transients such as are defined in certain standards, such as, for example, IEC 1400-1. These transients are superimposed on any steady-state spatial variations defined by wind shear, tower shadow and an upturbine wake.
Click the Wind icon on the toolbar, select Time varying wind, and choose the Transients option.
First enter the following data:
Reference height at which the wind speed is defined, unless Refer wind speed to hub height has been selected. If wind shear (see 6.2) is defined, the wind speed at any other height will be different.
Flow inclination: for non-horizontal wind flows (e.g. on the side of a hill). A positive value indicates a rising wind.
Separate transients may be defined for each of the following variables:
• Wind speed: defined at the reference height. Wind direction: measured clockwise from North - see wind direction (see 6.9).
• Horizontal wind shear: defined as the difference in wind speed between the two sides of the rotor plane. The wind speed varies linearly with horizontal distance across the rotor plane. Vertical wind shear: defined as the difference in wind speed between the top and bottom of the rotor plane. The wind speed varies linearly with height. Vertical direction shear: defined as the difference in wind direction between the top and bottom of the rotor plane. The wind direction varies linearly with height.
All defined transients will be used, but if any is not required, specify an appropriate start value and a zero amplitude. For each transient, enter the following information for the variable in question:
Start value: the value of the variable before the start of the transient. Amplitude of change: the largest range from maximum to minimum during the transient. ? Time to start cycle: the time into the simulation at which the transient starts. Time period of cycle: the duration of the transient. Type of cycle: Three types of transient are provided, as specified by various standards. A half wave transient begins with the start value and increases sinusoidally by the amount defined by the Amplitude. A full wave transient begins with the start value and changes sinusoidally by the amount defined by the Amplitude in the first half of the time period, changing back to the start value by the end of the time period. The IEC-2 transient is a more complex shape, as described in the Theory Manual.
# 6.9 Definition of wind direction
The wind direction is the direction from which the wind is coming, measured clockwise from North when looking from above, i.e. East of North.
The default nacelle orientation and the wind direction together define the nacelle yaw angle. This may change during a simulation if any yaw control (see 5.18) is specified, i.e. passive yaw, active yaw control or a yaw manoeuvre. Note that the yaw angle can also be defined explicitly for the steady parked loads (see 7.10) calculation.
Figure 6-2: Default nacelle orientation
The yaw misalignment angle is defined as:
Yaw misalignment $=$ nacelle angle - flow direction (for 0⁰ ≤ nacelle angle $\leq180^{\circ}$ ) Yaw misalignment $=$ (nacelle angle – 360⁰) - flow direction (for $180^{\circ}<$ nacelle angle $\leq360^{\circ}.$ )
Note that the calculated “Nacelle yaw displacement” output variable found under “Nacelle motion”, is defined in terms of the measured nacelle angle (positive sign convention shown in figure 6.2) as follows:
Nacelle yaw displacement $=$ - nacelle angle (for $_{0^{\circ}}$ ≤ nacelle angle $\leq\,180^{\circ}$ ) Nacelle yaw displacement $=360^{\circ}$ - nacelle angle (for $180^{\circ}~<$ nacelle angle ≤ 360⁰)
# 6.10 Generating turbulent wind fields
In order to run a simulation which uses a 3-dimensional turbulent wind field, a suitable turbulence field must first be generated. Once generated, these turbulence fields are stored so that they can be used for future simulations.
Turbulence fields contain time histories of wind speed variations at each of a number of points on a rectangular grid covering the turbine rotor swept area. The time history at each point will have spectral characteristics conforming to one of the available turbulence models. The time histories at any two points in the rotor plane will be correlated with each other in accordance with an appropriate model of the lateral and vertical coherence characteristics of atmospheric turbulence. Linear or cubic interpolation is used between grid points, as well as between successive time points.
To generate a new turbulence field, first define the turbulence characteristics and then generate the turbulence field as described below.
# 6.10.1 Defining turbulence characteristics
Click the Wind icon on the toolbar and select Define turbulence. Then define the following:
| Latitude: Turbulence characteristics are onlyweakly dependent onlatitudeexcept near the equator, where the model breaks down. |
| The appropriate length scales and turbulence intensities are calculated and displayed, although it is possible to specify different turbulence intensities when setting up a simulatio run. The old von Karman model allows either 1 or 3 components of turbulence to be |
| generated. All the turbulence length scales must be entered by the user. This model conformstoESDU74031. |
| Mannmodel:enterthefollowingparameters: Shear parameter (gamma): Zero gives isotropic turbulence; the recommended value for IEC edition 3 is 3.9, which produces a spectrum similar |
| to a Kaimal spectrum. Scale length (L): The scale length is given by L = 0.8 A1 (A1 as defined inIEC 61400-1 Edition |
| 3). The recommended value for IEC edition3 is 33.6mwhen the turbine hub height z > 60 m. FFT points: The number of points used for the FFT |
| in the lateral and vertical directions, e.g. 32 (must be a power of 2). Cubic interpolation is used to find the wind at the requested grid points. Max.lateral/vertical Thisshouldbeappreciablygreater than wavelength therotordiametertoavoidperiodicity in thelateral andvertical directions.A value ofatleastfour times thescale length is recommended. |
To start from an existing turbulent wind field definition, click Import details… to select an existing turbulent wind file. The parameters defining the turbulence in that file will then be loaded.
# 6.10.2 Advanced options
The Advanced options button allows the following additional functionality:
Gust iterator: this option is available only with the ‘Improved von Karman’ model of turbulence (see 6.10.1). It will carry out an iteration in which the surface roughness is varied until a particular gust is achieved somewhere within the rotor swept area. The turbulence intensity will change with surface roughness in accordance with the model. The following parameters are required:
Gust wind speed, tolerance, and averaging time: the iteration will continue until the specified maximum gust, defined as a block average with the specified averaging time, is achieved within the specified tolerance somewhere within the rotor swept area. Gusts of the specified speed and up to the specified tolerance above this speed are accepted.
Hub height and rotor diameter: these define where the rotor swept area is in relation to the lateral and vertical extent of the wind file. Click Get current values to input the values currently available on the Rotor (see 4.1) screen.
Height of turbulent wind field: together with the hub height, this defines the vertical position of the rotor within the wind field – see 6.7.
Minimum and maximum allowed values of surface roughness: the iteration will fail if the required gust cannot be achieved with a surface roughness between these limits.
Minimum and maximum alongwind position of gust: these can be used to ensure that the maximum gust is not too close to the start or end of the wind file.
Wind shear: Wind shear can be taken into account when finding the maximum gust in the wind file. In addition wind shear can be applied to the target gust speed. If this second option is chosen, the target gust speed will vary with height, such that the required gust is higher near the top of the rotor; otherwise the target gust speed is equal at all heights. The wind shear may be specified in terms of a shear exponent or a roughness length. Click Get current values to input the shear model as currently defined on the Wind screen.
Number of files to generate: If required, this calculation can generate more than one file with different random number seeds which match the specified criteria. In this case, the resulting wind files will be called name_n.wnd, where name is the run name selected, and n is a sequential number.
Matching a measured single-point time history: this option is available only with the Kaimal and
von Karman models of turbulence. It generates a wind file in which the wind speed time history at a
specified point in the rotor plane exactly matches a given, perhaps measured, time history. The
coherence function associated with the chosen turbulence model is used to generate appropriate time
histories at all the other points in the plane, but the single-point spectrum associated with the chosen
turbulence model is not used, since the measured data defines this. The measured data may include
more than one component of turbulence: any unmeasured components should be provided as zeros, and
the program will then synthesise these components entirely, using the single-point spectrum of the
chosen turbulence model. The following information must be supplied:
• a file of measured data, with one column for each component specified in the chosen turbulence model (using zeros for any components for which measured data is not available).
• The mean wind speed for the measured data. The sampling frequency of the measured data. The grid point at which the simulated turbulence should match the measured data.
The duration of the measured dataset must be at least as great as the duration of the wind file which is to be created. The sampling frequency should of course be sufficient to capture the wind speed variations in sufficient detail, but need not match the frequency of points specified for the final turbulence file: the measured data is simply re-sampled as required.
# 6.10.3 Generating the turbulence field
Having defined the turbulence characteristics as above, click the appropriate button to generate the turbulence field either Now or In Batch. Alternatively, click the Calculations icon on the toolbar to bring up the Calculations screen (if it is not already present). Select Wind turbulence, and then click either Run now, or Run in Batch to store the calculation for later execution using the batch (see 7.2.7) facility.
# 6.10.4 Turbulence file format
The wind file format is documented in APPENDIX F TURBULENCE FILE FORMAT. This information is useful for users who wish to generate their own turbulence file.
# 6.11 Annual wind distribution
Some post-processing calculations require the annual distribution of hourly mean nspeeds to be defined. This includes the Annual Energy Yield calculation as well as any calculations which produce weighted lifetime values, for example of fatigue damage or damage equivalent loads.
Click the Wind icon on the toolbar and select Annual wind distribution. Select either a Weibull or a user-defined distribution. For the Weibull distribution, enter the annual mean wind speed and shape factor. A shape factor of 2 results in a Rayleigh distribution. For the user-defined distribution, enter the cumulative distribution or exceedance table, starting with an exceedance of 1.0 at zero wind speed. For wind speeds above the last point in the table, the exceedance will be assumed to decay in proportion to $\mathsf{e x p}(-\mathsf{v}^{\mathsf{k}})$ , where v is the wind speed and $\mathsf{K}$ is the number entered as the “Exponent for high wind speeds”.
Click either Cumulative plot or Probability plot to view the wind speed distribution as defined.
To define a joint distribution of wind and waves, select the wind and wave option, and click Define scatter diagram (see 6.15).
# 6.12 Defining Waves
Click the Sea state icon on the toolbar and select the type of wave model required:
• None (see 6.12.1): no waves are required.
• Regular waves (see 6.12.4): choose Linear Airy or Stream Function waves of a single frequency. Irregular waves (see 6.12.2): choose either a standard Jonswap or Pierson Moskowitz spectrum, or enter a user-defined spectrum. This will generate a stochastic wave train as a superposition of linear Airy waves with the appropriate spectral distribution.
For all wave types, enter:
Direction of approach (from North): the bearing from which waves arrive at the tower. Like wind direction, wave direction is defined as the direction which the waves are coming from, and not the direction that the waves are going to. The angle increases positively to the East of North.
Then enter the appropriate data for the chosen model as described below.
In the case of irregular waves, it is also possible to add a “Constrained wave” (see 6.12.3), which guarantees that the irregular wave train will contain an extreme wave of a specified height.
Any specified waves will be ignored if the turbine is not offshore - see 4.5.5.
# 6.12.1 None (no waves)
In this case the tower is not subjected to wave loading. However, if the turbine is specified as being offshore, it will still experience hydrodynamic damping and a drag force from any specified currents.
# 6.12.2 Irregular waves
Irregular waves are generated as a superposition of linear Airy waves with the appropriate spectral distribution. Enter:
Random number seed: an integer to be used as the seed value for the random wave generation. A different seed value will generate a different realisation of the irregular waves, but still conforming to the desired wave spectrum.
To define the wave spectrum, chose either Jonswap / Pierson-Moskowitz spectrum or Userdefined spectrum, and specify the diffraction approximation to use – see below.
# Jonswap / Pierson-Moskowitz spectrum
This standard wave spectrum is described in the Theory Manual, and requires the following parameters:
Significant wave height: corresponding to the average height of the highest one third of the waves in the seastate. Given a time-history of wave height, this parameter can be calculated as four times the standard deviation of the water surface elevation.
Peak spectral period: the period of the most energetic component in the wave spectrum. Peakedness parameter: this parameter controls the width of the frequency band containing most of the energy in the spectrum. It should take a value between 1 and 7. For a Pierson-Moskowitz spectrum, the peakedness parameter should be set to 1.
# User-defined spectrum
If the wave spectrum is known at the site of interest, select this option and then click Define spectrum. A further window opens which allows up to 100 pairs of values of frequency and power spectral density to be entered. Data entry boxes may be created and deleted by clicking on the Add and Delete buttons respectively. Data points are automatically sorted by frequency. Clicking on the Show button will reveal a plot of the spectrum as defined. The values of power spectral density at the lowest and highest frequencies entered should be zero.
# Wave diffraction approximation
Two options are available to account for diffraction effects at small wavelengths comparable to the size of the structure (see Theory Manual):
MacCamy-Fuchs approximation: enter a representative member diameter, or select Auto-define to allow Bladed to select a suitable value.
Simple cut-off frequency: enter a frequency, or select Auto-define to allow Bladed to select a suitable value.
# 6.12.3 Constrained waves
An irregular wave history can be modified to include a prescribed extreme wave at a particular time. Two methods are available (see Theory Manual):
• Linear NewWave: the linear airy waves are constrained in such a way as to produce the desired extreme wave height at the specified time, or • Stream Function: a non-linear stream function wave is blended into the irregular wave history.
In each case, enter:
• Constrained wave height: measured from trough to crest; in the case of Linear NewWave, the trough elevation is taken as the lower of the troughs on either side of the crest. Time of constrained wave: determines at what point in the simulation the constrained wave occurs
For the Stream Function method, enter also the Constrained wave time period.
# 6.12.4 Regular waves
This option generates a regular wave train, also sometimes used as a way of obtaining extreme deterministic waves.
Choose the type of regular wave model:
Linear Airy gives a simple linear wave of the specified period • Stream Function gives non-linear stream function waves, which are more appropriate for extreme waves.
In either case, the following parameters are required:
Wave height: defined from trough to crest.
Wave period.
# 6.12.5 SEA file
This option in the “Tide and SEA file” tab defines the sea-state based on a pre-processed SEA file. A SEA file can be generated by clicking the “Generate SEA File” button. It consists of a list of regular wave components which are linearly superimposed to form the irregular sea state. Each regular wave component is specified in terms of four properties:
Frequency – the reciprocal of the period of the wave. The number of frequency components generated depends both on the “duration” variable $D$ , and on the frequency cut off values selected. The frequency resolution $d f=1/D_{\mathrm{.}}$ , so the longer the duration, the smaller the
frequency steps between adjacent frequency components. Harmonic components with
frequencies that lie outside of the frequency range indicated in the user interface will be omitted. If ‘Automatic Frequency Range’ is selected (this is the default option), $0.05\%$ of spectral energy at each end of the spectrum will be omitted so that only $0.1\%$ of energy is lost overall. Alterantively it is up to the user to specify the range of frequencies to be used. If the user enters their own values, they need to ensure that the most important features of the spectrum are retained, both in terms of frequency range, and energy distribution. In some cases for instance, it may be acceptable to cut off high frequency ‘tails’.
Amplitude – With the deterministic amplitude method, the amplitude is determined by the spectral density. Random amplitudes are calculated according to a Rayleigh distribution with the scale parameter proportional to the spectral density.
• Direction – assigned randomly according to the directional distribution about the mean direction. More details on the directional distributions are given in the theory manual. Phase – a uniformly distributed random variable.
The random number seed is used to initiate the sequence of random numbers used for the phases, directions and amplitudes (if applicable). If regular waves are chosen rather than a JONSWAP spectrum, then there is only one component in the SEA file.
The SEA file format can be edited using standard text editing tools. It is therefore possible to write or even modify existing SEA files externally to Bladed such that constant energy spectra or constrained waves can be simulated using SEA files. While the user is free to select the range and resolution of frequencies used to define the power spectral density of the sea state the user should verify that the required statistics/properties of the input wave spectra are satisfied for the free surface elevation timehistory created (for example, spectral shape, Hmax , etc). It would also be important to make sure that the possible frequency responses of the structure will be covered (i.e. not omitting any important frequencies of the structure response).
Definitions of the fields used to define the JONSWAP spectrum can also be found in section $\underline{{6.12.2}},$ and detailed theory is given in the theory manual, as well as definitions for the different period parameters.
# 6.13 Defining Currents
To define currents, click on the Sea state icon on the toolbar and then select the Currents tab. Three current components may be specified, separately or in combination:
• near-surface current • sub-surface current • near-shore current
If more than one current is selected, the velocity vectors are added linearly. The form of each current profile is described in the Theory Manual. Any specified currents will be ignored if the turbine is not offshore (see 4.5.5).
Click the check boxes to select the required current components, and enter the relevant parameters for each as described below:
# 6.13.1 Near-surface current
This is a current whose strength decreases linearly with depth. Enter the following parameters:
Surface velocity.
• Reference depth: the depth at which the flow velocity reaches zero.
Heading (degrees from North): the direction towards which the current is flowing. Unlike wind and wave directions, current direction is defined as the direction that the current is flowing to, and not the direction that the current is coming from.
# 6.13.2 Sub-surface current
This is a current whose strength depends on height above the sea bed according to a power law or a user defined relationship. Enter the following parameters:
# Surface velocity.
Heading (degrees from North): the direction towards which the current is flowing. Unlike wind and wave directions, current direction is defined as the direction that the current is flowing to, and not the direction that the current is coming from.
Normally a 1/7 power law is used, but there is also an option to enter a different power law exponent x such that $\mathsf{V}(\mathsf{h})=\mathsf{V}(\mathsf{s u r f a c e})^{\ast}\,(\mathsf{h}/\mathsf{s e a}\,\mathsf{d e p t h})^{\wedge}\times,$ where V(h) represents flow speed a distance h above the sea bed. The default is $\mathbf{\Deltax}=\mathbf{1}/7$ .
Alternatively, a custom shear profile can be defined by the user by selecting the “Custom Shear Profile” option. The desired shear profile can then be entered in a look up table.
To define the custom shear profile, the current shear at each height above the sea bed must be specified.
The height above sea bed data must be entered as values normalised by the mean sea depth for the simulation. Note that the mean sea depth for a simulation is the sum of the mean sea depth specified in the support structure definition screen and the tide height for the simulation.
heightaboveseabed Normalisedheightaboveseabed $=$ meansealeveltideheight
The current shear at each height is defined as the mean current speed at that height divided by the mean sea surface current speed.
meancurrentspeedateachheight Currentshear $=$ meansea surfacecurrentspeed
The values in the first row of the look-up table represent the normalised height above sea bed and the current shear values at the sea surface. The values in this first row therefore always each take the value 1. Subsequent data points must be entered in descending order of normalised sea surface height.
# 6.13.3 Near-shore current
This is a current which is constant with depth. Enter the following parameters:
# Current velocity.
Heading (degrees from North): the direction towards which the current is flowing. Unlike wind and wave directions, current direction is defined as the direction that the current is flowing to, and not the direction that the current is coming from.
# 6.14 Tide height
The mean sea level is defined on the Tower screen (see 4.5.5), but for any particular calculation the water depth may be changed by specifying the height of the tide. Click the Sea state icon on the toolbar, select Tide and specify the change in water depth compared to mean sea level.
# 6.15 Scatter diagram
Users with a licence for the Advanced Processing module may use a scatter diagram to set up multiple simulations containing the correct combinations of wind and wave parameters, and to post-process these simulations using the extreme load extrapolation option. To define the scatter diagram, from the toolbar menu select Specify … Sea State … Scatter.
The scatter diagram is used to define the distribution of wave height and period as a function of wind speed, for use in setting up multiple simulations and extreme load extrapolation of off-shore turbines.
The probability density of each wind speed bin is the summation of the probability density in the appropriate set of wave bins. The total probability density should not be greater than 1.0.
# 6.16 Earthquakes
Users with a licence for the Seismic module may run simulations which include an earthquake. From the toolbar menu, select Specify... Earthquake to bring up the Earthquake definition screen. Select the option to specify an earthquake to use in simulations, and browse for an earthquake file. Specify the time in the simulation at which the earthquake will strike, and the principal direction of the earthquake.
An earthquake file is a text file containing two header lines and a time history of earthquake ground accelerations. The header format is:
POINTS n TSTEP t
where n is the number of time points and t is the timestep. The header is followed by two columns of numbers representing time histories of ground acceleration in the principal direction and perpendicular to this direction. Each column must contain n entries.
# 6.16.1 Generating earthquake time histories
Users with a licence for the Seismic module may generate earthquake files from a standard earthquake response spectrum.
Open the Earthquake definition screen, and select Generate new earthquake. Click Import... to start from an existing definition.
Click Define response spectrum to specify the response spectrum as a look-up table, and then enter the following parameters:
Response spectrum damping coefficient: This is the damping coefficient of the structure for which the response spectrum is defined. The damping used in the earthquake generation is not a physical characteristic of the structure or the soil, but is the damping that was used to calculate the seismic response spectrum and as such it is inherent to the spectrum definition and it is not correct to use different damping values for the same response spectrum. It could be understood as the damping coefficient of the earthquake measuring device.
Random number seed: different values will result in different earthquake realisations, but in each case the resulting earthquake will conform to the specification as defined.
Duration of earthquake: The length of time for which the earthqhuake will persist.
Earthquake time step: the timestep of the ground acceleration time histories which will be produced.
The accelerations will be in a given fixed direction, the principal direction of the earthquake. However, if desired, an additional component at $90^{\circ}$ to the principal direction can be specified. This may be proportional to the acceleration in the principal direction (which simply corresponds to scaling and rotating the accelerations by a fixed amount), or independent.
Certain advanced parameters may also be specified, to control in detail the way in which the earthquake is generated:
Accuracy of fit: the number of frequency points can be specified at which the actual response spectrum can be specified,will be compared against the target response spectrum. The allowable deviation from the target response spectrum can also be specified. Filters: a low and high pass filter can be defined to remove frequencies outside the range of interest. This may help to speed up convergence. Shaping function: This controls the overall shape of the earthquake. Two models are included. These are typical for Stiff soils or Soft soils. Some seismic standards specify the length of the Stationary part of the earthquake. These can be entered using the Soft option, along with a final Decay constant.
# 6.17 Point loading
On the Point Loading screen, the user is able to apply a pre-defined loading onto any point on the tower or support structure or the blade. To access the screen, use the Specify and then Point Loading menu item. Bladed uses a file that contains a time-history of loading in a specified format. The timehistory of loading can either have a single loading component and be applied at any horizontal direction or have full six directions of loading (3 forces and 3 moments). The load histories must be specified in the global coordinate system. The user can specify more than one point loading to be applied during a simulation.
• Tower Loading\Blade loading: If either of these boxes are checked then the currently entered settings will be applied during the simulation.
To add a point loading description, press the $\mathbf{+}$ symbol next to the Loading Set. Then enter the following parameters:
Impact data file: Browse for the location of the file containing the specified time-history of loading. The file must specify the number of points in time at the top of the file. For the single component case, the order of columns is (Time//Force), separated by a tab. For the six component case, the order of columns is (Time//Fx//Fy//Fz//Mx//My//Mz). The values in the time column do not need to be equally spaced. The loads should be in SI units i.e. N and Nm. For example:
| Submodel | Default setting in aerodynamics | Default setting in pre-4.8 aerodynamics |
| Drag induction | uo | *uo |
| Momentum theory | Glauert momentum theory | Axial momentum theory* |
| Dynamic wake base | Dynamic wake on blade | Dynamic wake on blade element* |
| Dynamic wake model | element Oye dynamic wake | Pitt & Peters dynamic wake |
| Dynamic tangential induction | On | N/A |
| Skew wake correction | Glauert skew wake model | No skew wake model |
| model No inflow below TSR | 1 | 0 |
| Full inflow above TSR | 2 | 0 |
| Tip loss correction | uo | On |
| Hub loss correction | Off | Off |
| Simplified aerofoil orientation | Off | On* |
| Dynamic stall model | Incompressible Beddoes- Leishman | Pre - 4.8 Beddoes-Leishman |
| Dynamic pitching moment coefficient | uo | On* |
| Include impulsive lift and moment contributions | Off | On* |
| Starting radius for dynamic stall | 25 | 0 |
| Ending radius for dynamic stall | 95 | 100* |
| Separation time constant | 3.0 | 3.0 |
| Pressure lag constant | 1.7 | 1.7* |
| Vortex lift constant | 6.0 | 6.0* |
\* Asterisk indicates options that were hard-coded in the pre-4.8 model.
# 7.19 Imbalances
Use the Imbalances window (see 7.5) to define any mass imbalance of the rotor, or errors in the set angle or the pitch angle of one or more blades. Enter the following data:
• Imbalance mass: the mass of the imbalance. Note: From version 4.0 onwards, this is an additional mass. Radius of imbalance: the radial position of the centre of gravity of the imbalance mass.
• Azimuthal position of imbalance: zero means the imbalance mass centre of gravity is on blade 1.
Error in blade set angle: for each blade, specify the error in set angle. This will rotate both the pitching and non-pitching sections of the blade - see 4.1. Error in pitch angle: for each blade, specify the error in pitch (or aileron/airbrake) angle. This will rotate only the pitching sections of the blade - see 3.2. Error in blade azimuth: for each blade, specify the error in azimuthal position for each blade.
# 7.20 Turbine Faults
Use the Turbine Faults window to specify pitch fault cases, generator or network fault cases, yaw fault cases and transducer fault cases.
From the Bladed toolbar, use Specify... Faults. An external controller fault can be specified in the Other faults tab. This allows the user to define an integer number which is passed through in swap array number 161 at the specified time. It is up to the user defined external controller dll to interpret this number and take the appropriate action.
# 7.20.1 Pitch faults
Enter the following data to specify pitch faults (applies also to ailerons or airbrakes if defined):
Pitch failure mode: for each blade, specify the type of failure for the pitch mechanism. • None: The pitch mechanism is operating normally. • Permanently Stuck at the Pitch of failed blade. • Constant rate runaway. • Free or Constrained by a constant torque and/or the pitch bearing friction. • Seizure at a given time or when passing through a given angle after a given time. Recoverable: specify whether the pitch failure is recoverable. This means that the safety system pitch action will override the pitch fault. Pitch of failed blade: for each blade where the pitch mechanism is Permanently Stuck, specify the angle at which the pitch is stuck Time for pitch failure: enter the time at which the failure will occur. Constant rate demand for constant rate runaway failures. Constant torque demand for Free/Constrained failures. Fail when passing through: angle at which seizure will take place after the Time for pitch failure has passed.
# 7.20.2 Generator and network faults
Four types of generator or network faults can be modelled:
# Grid loss
Specify the time from the start of the simulation for the grid loss to occur.
Generator short circuit
Enter the time from the start of the simulation for the short circuit fault. Different short circuit models are available for synchronous generators as long as a transient model is selected: select the type of fault required. In other cases, a file containing a time history of generator torque during the fault can be provided. Enter the path to the short circuit file. The file should be in ASCII format, and the first line should be an integer specifying the number of points in the time history. The remainder of the file can contain either 2 or 3 columns. Column 1 is the time in seconds from the start of the fault. If the file contains only 2 columns, the second column is the generator torque (Nm). If the file contains 3 columns the second column is the maximum generator torque and the third column is the minimum generator torque (both in Nm). The power production control is allowed to continue to specify a demanded generator torque between these two values.
# Network voltage disturbance
If electrical dynamics have been specified in the generator module, a network voltage disturbance can also be specified. Enter the time for the start of the voltage disturbance, and select either
External Input File or a user defined Look up table. The external input file should be an ASCII file containing an integer specifying the number of points on the first line, and then two columns on the subsequent lines giving time in seconds since the start of the disturbance, and the corresponding voltage disturbance factor. The voltage disturbance factor is expressed as a fraction of the nominal network voltage.
# Network frequency disturbance
If electrical dynamics have been specified in the generator module, a network frequency disturbance can also be specified. Enter the time for the start of the frequency disturbance, and select either External Input File or a user defined Look up table. The external input file should be an ASCII file containing an integer specifying the number of points on the first line, and then two columns on the subsequent lines giving time in seconds since the start of the disturbance, and the corresponding frequency disturbance factor. The frequency disturbance factor is expressed as a fraction of the nominal network frequency.
# 7.20.3 Yaw faults
If a yaw system is defined, three types of yaw fault can be specified:
# Constant rate
Enter the constant rate for a yaw runaway, and the position of any end stop if required. The end stop is defined clockwise from north.
# Constant torque
Enter the constant yaw failure torque
# Free/Constrained
Enter any linear stiffness, linear damping and yaw friction which are active during the yaw failure.
# 7.20.4 Transducer faults
Transducer faults can be modelled for both the power transducer and the generator speed transducer. Enter the time for failure and the constant value which the transducer reports after it has failed. Note that a transducer failure can be used to cause an overspeed.
If an external controller failure is specified, at the given time channel 161 of the swap array will take the value specified. This can be used to activate a fault within the external controller.
# 7.21 Coordinate systems and outputs
A large amount of output is potentially available from some simulations. Click the Outputs button on the Calculations screen to define which of these outputs are required. The outputs are grouped into blade, tower and other outputs.
The Aerodynamic information, Performance coefficients and Power curve calculations are unaffected, as they produce a pre-defined set of outputs.
# 7.21.1 Blade coordinate systems and outputs
To configure blade outputs, click Blade outputs on the Calculation Outputs Specification screen.
Blade outputs include aerodynamic information (including distributed aerodynamic loadings), blade loads and motions. For each of these categories, the information may be generated at any or all of the blade stations (see 3.1). Click Select Output Stations to determine which information is required at which stations. Click Add to define additional points where interpolated loads will be output.
Having defined the blade stations required, specify for each type of output whether that information is required on the First blade, All blades, or not at all (None).
The loads can be specified independently about the following axes: Principal axes, Root Axes, Aerodynamic Axes and User Axes. Within each group, the loads are labeled $\mathsf{M}\mathsf{x}_{\prime}$ My, Mxy, Mz, Fx, Fy, Fxy, Fz.
It’s important to note that the Principal Axes and Root Axes stations coincide with the underlying finite beam element model nodes. The Principal Axes coordinate vectors align with the finite beam elements. The Root Axes coordinates have the same origin rotated to align with the blade root.
The Aerodynamic and User Axes potentially have origins that do not coindicide with the underlying finite element beam model nodes. These load outputs are found by transforming the finite element loads (in the Principal Axes coordinates) to the Aerodynamic or User Axes coordinate centre. The tranformation accounts for additional Mx and My bending moments that are generated at the Aerodynamic or User Axis centre by the element axial force Fz in the Principal Axes system, due to the offset in the aerofoil plane between these two coodinate centres. This effectively estimates the load that would have occurred if the load was carried through an axis at the Aerodynamic Axis or User Axis centre. There is some approximation in this method as the Aerodynamic and User Axes loads are not the true load path. Any changes to the the blade dynamics (e.g. deflections or loads) that might result from such a change in load path are not accounted for.
Principal axes: The positive z-axis follows the local deflected neutral axis at each blade station towards the blade tip. The positive y axis is defined by the principal axis orientation. The positive $\times$ axis is orthogonal to the y and z and follows the right hand rule. For output loads, the origin of the axes is on the neutral axis at each local deflected blade station. (see diagram below)
Figure 7-3: Blade principal axes coordinate system
Note that there is a subtle difference between the “principal axis” frame and the “blade local element frame”.
The “blade local element frame” is orientated so that the X vector in this coordinate system points directly between adjacent nodes on the blade. The other two coordinate system vector directions are determined by the “principal axis twist” angle. This is explained in more detail in DNV GL technical note UKBR-110052-T-31-A
The “principal axis” frame is used for load output in Bladed. The principal axis orientation is calculated by taking the average orientation of the two blade elements at the node where the elements join. This is illustrated below. The two adjoining local element frames are shown in green and red. The principal axes output frame is shown in blue.
Note that the element local coordinate system has its $\times$ direction along the element, unlike the “principal axes” coordinate system which has z along the element axis.
Figure 7-4: Relationship between blade "local element axes" and "principal axes coordinates"
Root axes: The orientation of the axes is fixed to the blade root and does not rotate with either twist or blade deflection. The axis set does rotate about the z axis with pitch. For output loads, the origin of the axes is on the neutral axis at each local deflected blade station.
Figure 7-5: Blade root axes coordinate system
• Aerodynamic axes: The $\times$ axis is perpendicular to the local aerodynamic chord line and the positive y axis is aligned along the local aerodynamic chord line from leading edge to trailing edge. The z-axis is parallel to the local deflected neutral axis at each blade station and increases towards the blade tip. For output loads, the origin of the axes is on the chord line at $25\%$ chord from the leading edge at each local deflected blade station.
Figure 7-6: Blade Aerodynamic axes coordinate system
User axes: The origin of the axes is specified as percentages of chord, parallel and perpendicular to the chord at each blade station. The user can specify whether the z-axis is parallel to the root axis or the local neutral axis, and independently whether the y-axis is aligned to the principal axis orientation, the aerodynamic twist or the root axis.
Figure 7-7: Position of the user defined output axis
Note that an additional set of the blade root loads, using a coordinate system fixed to the blade root in the hub (i.e. inboard of the pitch bearing) are output as part of the Rotating Hub output group. This coordinate system is shown in Figure 7-9.
The table below compares the blade outputs between Bladed versions 3.x and 4.x
Table 7-2: Blade loads output coordinate systems
| Origin | y-axis | z-axis |
| (v3.85) | Blade loads Mx/My | At deformed pitch axis | Fixed to blade root and doesn't rotate with pitch or twist (=always tangential to rotor plane coned) | Parallel to rotor plane (coned) |
| Blade loads Flap/edge | At deformed pitch axis | Parallel to chord (twist angle) | Parallel to rotor plane (coned) |
| (v4.x) | Principal axes | At deformed neutral axis | Determined by principal axis orientation Parallel to blade root Y axis. | Parallel to deformed neutral axis. Affected by blade mounting cone / sweep angles |
| Root axes | At deformed neutral axis | (doesn't rotate with twist or deflection, i.e. tangential to rotor plane coned whenboth pitch angle and blade set angle are 0 and no blade mounting cone or sweep are present. Rotates about z axis with pitch) | Parallel to blade root Z axis. Parallel to rotor plane (coned) when no blade mounting cone or mounting sweep present. Affected by blade mounting cone / sweep angles |
| Aero axes | At 25% chord (deformed) | Parallel to chord (aerodynamic twist) | Parallel to deformed neutral axis. Affected by blade mounting cone / sweep angles |
| User axes | Anywhere in blade section, user defined | Determined by principal axis orientation, OR | Parallel to deformed neutral axis Affected by blade mounting cone / sweep angles OR |
| Fixed to blade root as in root axes (rotates with pitch) , OR | Parallel to rotor plane (coned) when no pre-cone or sweep present. Affected by blade mounting cone/ sweep angles |
| Blade root from rotating hub | At blade root centre (on pitch axis) | Parallel to chord (aerodynamic twist) Fixed to blade root and doesn't rotate with pitch (=always tangential to rotor plane coned) | Parallel to rotor plane (coned). NOT affected by blade mounting cone / sweep angles |
# The coordinate system for the blade deflections is as follows:
z-axis Radially along the blade root Z axis
x-axis Perpendicular to $Z,$ and pointing towards the tower for an upwind turbine, or away from the tower for a downwind turbine.
y-axis Perpendicular to blade axis and shaft axis, to give a right-handed co-ordinate system independent of direction of rotation and rotor location upwind or downwind of the tower.
# Load configuration file:
If the blade manufacturer has provided a load configuration file, this can be used to produce custom output for that particular blade.
# 7.21.2 Hub coordinate system
To configure hub outputs, click Other Outputs on the Calculation Outputs Specification screen.
The co-ordinate system for the hub load and deflection outputs from the calculations is based on the ‘GL’ convention, with some modifications as specified below.
Hub loads in fixed frame of reference:
XN Along shaft axis, and pointing towards the tower for an upwind turbine, or away from the tower for a downwind turbine (the picture shows an upwind turbine).
ZN Perpendicular to XN, such that ZN would be vertically upwards if the tilt angle were zero.
YN Horizontal, to give a right-handed co-ordinate system independent of direction of rotation and rotor location upwind or downwind of the tower.
Figure 7-8: Co-ordinate system for stationary hub loads
Hub loads in rotating frame of reference:
XN Along shaft axis, and pointing towards the tower for an upwind turbine, or away from the tower for a downwind turbine (the picture shows an upwind turbine).
ZN Perpendicular to XN, such that ZN would be aligned with blade 1 axis if the cone angle were zero.
YN Perpendicular to XN and ZN, to give a right-handed co-ordinate system independent of direction of rotation and rotor location upwind or downwind of the tower.
Origin At hub centre (intersection of blade and shaft axes).
Figure 7-9: Co-ordinate system for rotating hub loads
# 7.21.3 Anti-clockwise and downstream rotor coordinate systems
The rotor and blade output coordinate systems for all permutations of clockwise, anti-clockwise, downstream and upstream rotors are shown in the following four figures.
Note that for simplicity these diagrams do not include blade mounting cone, blade mounting sweep. The blade pitch and set angle are shown at zero in these diagrams.
Figure 7-10: Blade and rotor coordinate systems for a clockwise upstream rotor
Figure 7-11: Blade and rotor coordinate systems for an anticlockwise upstream rotor
Figure 7-12: Blade and rotor coordinate systems for a clockwise downstream rotor
# Anti-Clockwise Downstream
Figure 7-13: Blade and rotor coordinate systems for an anticlockwise downstream rotor
# 7.21.4 Tower outputs
Click Tower outputs on the Calculation Outputs Specification screen.
Tower loads may be generated at any or all of the tower stations (see 4.5). Click Select Output Stations to determine at which stations the loads are to be output. For the monopile tower model, click Add to define additional points where interpolated loads will be output.
Having defined the tower stations required, specify for each individual output whether that information is required. For the loads, this can be specified independently for each of the forces and moments.
There is also an option to Refine deflections. With this option disabled, the tower deflection outputs are modal deflections. Modal deflections give a good estimate of overall tower motion, but may not be very precise in predicting the small deflections at foundation stations. This can lead to a poor estimate of the foundation reaction loads given the applied loads on the rest of the turbine.
With the Refine deflections feature enabled, the tower deflections are re-calculated at each output time step using the underlying finite element model with the external loads applied. The refined deflections give a more accurate estimate of the deflections at the foundation nodes, hence the foundation reaction forces are more accurate, especially for non-linear foundations. With Refine deflections selected, Bladed will iterate at each output time step to find the foundation applied loads based on the refined deflections. This will ensure that the applied loads, tower deflections and foundation reactions correspond correctly.
Note that the modal deflections are used to estimate the foundation loads when solving the structural system at each integrator time step. To use the refined deflections at every time step would require iteration on each time step, causing a significant increase in simulation time. The assumption with this modelling choice is that error in foundation load estimate due to using modal deflections doesn’t affect the overall turbine dynamics significantly. It is therefore reasonable to only calculate the foundation reaction loads based on refined deflection on each output time step.
# 7.21.5 Tower co-ordinate systems
The co-ordinate system for tower outputs is based on the GL convention, as described below.
XT Pointing South.
ZT Pointing along deflected tower centre line.
YT Pointing East.
Origin At each tower station.
Note that for steady-state calculations the wind is deemed to come from the North.
Figure 7-14: Co-ordinate system for tower loads (monopile only) and deflections
Note that the coordinate system system for tower loads moves with tower modal deflections, so that the coordinate system remains aligned with the deflected member axis.
# Loads and deflections for multi-member towers
The loads are output with reference to the local member coordinate system for each member. The member x-axis is always aligned along the member. The member z axis is perpendicular to the member x-axis and aligned according to the direction cosines for the member z-axis as specified in the tower screen. These are the direction cosines of the z-axis relative to the global GL coordinate system. For example, a vertical member with z-axis direction cosine of 0.0 in the $\times$ -direction, 1.0 in the y-direction and 0.0 in the z-direction would be a member where the local $\times$ -axis corresponds to the GL z-axis, the local y-axis corresponds to the GL $\times$ -axis and the local z-axis corresponds to the GL y-axis.
The deflections are output in the GL coordinate system.
# Bladed multi-member output convention: local x-y plane:
# Bladed multi-member output convention: local x-z plane:
Figure 7-15: Multi-member tower coorindate output convention
# 7.21.6 Pitch system direction convention
Figure 7-16: Pitch system sign convention
The same sign convention applies to the applied pitching moment, pitch bearing friction and pitch actuator torque. A positive applied pitching moment means that the blade aerodynamic, gravitational and other applied forces act in such a way as to drive the pitch in the positive (feathering) direction.
# Pitch system loads and deflections:
For pitch angles, rates and accelerations, the positive direction is the direction in which the leading edge moves upwind, i.e. towards the feathered position:
# 7.21.7 Yaw bearing output
The yaw bearing is located at the “nacelle node” as specified in the support structure screen. For monopile towers, the yaw bearing is assumed to be located at the top tower station.
The co-ordinate system for yaw bearing loads is the same as for the top tower station except that it rotates with the nacelle yaw angle.
Yaw bearing output is defined opposite to the GL z-axis, with clockwise from North being positive.
# 7.21.8 Variables that follow “rotor direction is positive” convention
There are a number of other variables for which the positive direction is the ‘normal running’ direction. This applies to controller and drive train variables such as rotor and generator speeds, torques and azimuthal positions, drive train and generator torques etc.
# 7.21.9 Other general outputs
• Summary information: includes principal operational and environmental indicators.
• Software performance: useful for diagnosing slow simulations, and finding a suitable step length for Real Time Test simulations. Specific node outputs: This is an advanced feature, intended for users with some knowledge of the Multibody Dynamics approach used within Bladed. See the description below.
Clicking the Specific node outputs… button will open the Node Outputs screen, which allows you to add, delete or edit entries in the node outputs list. This is a list of structural nodes at which you want to see kinematic (position, velocity, acceleration) or loads outputs. To see a tree diagram of node identifiers and the components they are connected to, run Bladed once without altering the project details (a 1-second simulation is enough), and then open the verification $(.{\mathfrak{S V E}})$ file in a text editor. Any node and component identifiers you enter must be exactly as they appear in this tree, and the component must be connected directly to the node. For each node, choose the type – Loads or kinematics – and the coordinate system in which you want the outputs to be expressed. Specify a name for the output group; this is the name that will appear in the list of outputs in the dataviewer. You can also choose to have the outputs calculated at a position (Offset) other than that of the node itself. This option should be used with care, as it must represent a position in space that is physically on the turbine structure, otherwise the outputs will be meaningless.
# 7.21.10 Environmental information
These outputs describe the wind speed and direction, as well as the sea surface elevation and seismic motions, if present. All speeds are relative to the hub or rotor motion, so any motion of the nacelle is reflected in these outputs.
Hub wind speed magnitude: This is the overall relative flow speed magnitude experienced by the hub. Cup anemometer wind speed: This is the horizontal component of the hub wind speed magnitude. Hub longitudinal wind speed: The component of the horizontal wind speed aligned with the hub longitudinal direction (includes rotation due to yaw angle).
Hub lateral wind speed: The component of the horizontal wind speed aligned with the hub lateral direction (includes rotation due to yaw angle).
Hub vertical wind speed: The component of the flow speed magnitude in the vertical direction. Wind direction at hub: Direction, as an angle to North (clockwise looking down).
Wind upflow at hub: Upflow, as an angle to the vector of flow direction in the horizontal plane. Positive angle is upwards.
Rotor average longitudinal wind speed: Component of the overall flow speed along the hub X axis, averaged over the rotor plane. The rotor is effectively treated as a non-rotating disc.
Rotor average longitudinal wind direction: Direction, as an angle to North (clockwise looking down), averaged over the rotor plane as above.
Sea surface elevation: Instantaneous sea surface Z in global coordinates, at global $x{=}0,\,\Upsilon{=}0$ . Ground positions, velocities, accelerations: Seismic motion of the ground at the turbine base, in global coordinates.
# 7.21.11 Aerodynamic outputs
In this section the outputs of the aerodynamics module are explained. To understand the outputs, the definitions of the rotor axial/tangential direction are defined in the figures below. Note that “axial” indicates in direction of normal vector $\mathfrak{n}^{\prime\prime}$ and tangential indicates in the direction of tangent vector $\mathrm{"t"}$ . Further note that the radial vector $\"{\Gamma}^{\prime\prime}$ assumes an un-coned rotor.
Figure 7-17: Geometry of rotor including rotor and blade coordinate system
Figure 7-18: Aerofoil section with velocities and aerodynamic loads. The blade is pointing out of the paper and moves towards the right, while the flow is coming from below
Table 7-3: Aerodynamic outputs in Bladed
| Variable name | Explanation | Equivalentvariable name in pre-4.8 model |
| Axial inductionfactor | Velocity induced byrotorin axial direction normalized with the negative free stream flow speed | Axial inflow factor |
| Tangential induction factor | Velocity induced by rotor in tangential direction normalized with the negative rotational speed times local radius | Tangentialinflowfactor |
| Axial induction velocity. | Velocity induced byrotor in axial direction | n.a. |
| Tangential induction velocity. | Velocity induced by rotor in tangential direction | n.a. |
| Tip/hub loss factor | Magnitude ofPrandtl tip/hublossfactor | Tip/hub loss factor |
| Inflow angle | Angle between effective flow speed velocityvector and tangentialdirection of rotor plane (Φ in the figure) | Inflow angle |
| Angle of attack | Anglebetweeneffectiveflowspeed vector and local chord line (a in the figure) expressed the quarter chord point | Angle of attack |
| Angle of attack rate of change. Relative axial flow | Timederivativeofangleofattack expressed at the quarter chord point Axial component of relative flow speed | n.a. Incident out of plane |
| speed Relative tangential | vector. Tangential component of relative flow | velocity. Incident in-plane |
| flow speed Axial aerodynamic | speed vector Axial component of aerodynamic load | velocity. Out of plane |
| loading. Tangential | per unit length. Tangential component of aerodynamic | aerodynamic loading |
| aerodynamic loading Aerodynamic twist | load per unit length. Aerodynamicmoment aboutthe elastic | In plane aerodynamic loading. Aerodynamic twist |
| loading Chordwise | axis per unit length. Orientation of moment axis is in direction of unit vector "r "in Figure XX | loading |
| Aerodynamic loading Normal Aerodynamic loading. | Aerodynamic force in direction of the chordline Aerodynamicforce normal to the chord line | n.a. n.a. |
| Local longitudinal/lateral/v ertical flow speed | Components of the flow speed at each blade station in global co-ordinates including tower shadow exluding induction from rotor. | n.a. |
| Minimum Pressure Coefficient | Minimum pressure coefficient on the suction surface of the hydrofoil at the given angle of attack | n.a. |
# 7.21.12 Pitch Actuator outputs
• Pitch Angle – The current pitch angle of the blade
• Pitch rate – The current rate of pitching of the blade
• Pitch acceleration – The current acceleration of pitching of the blade
• Demand Pitch Angle/Rate – The input demand from the controller into the actuator system External Pitch Bearing Moment – This is the load that comes from the blade due to aerodynamic and inertial loading, acting on the bearing. When the rotor is clockwise, the sense of the load is opposite to that of GL Blade z-axis, but is the same sense for an anti-clockwise rotor. Pitch Bearing Friction – The total friction (kinetic or static) acting on the pitch freedom. Pitching Inertia – The total pitching inertia consisting of the outboard blade inertia plus the referred motor rotational inertia if a rotary actuator is defined. Pitch actuator torque (rotary actuator drive or no drive defined) –The torque provided by the pitch motor. Defined on the motor side of the gear ratio. Pitch actuator force (linear actuator drive) –The force provided along the line of the actuator ram. Controller demanded pitch angle/rate (setpoint trajectory planning defined) –The input demand from the controller before being limited by setpoint trajectory planning. Limited demanded pitch angle/rate (setpoint trajectory planning defined) –The demand after it has been limited by setpoint trajectory planning. Motor position/rate/acceleration (When a flexibility in the actuator drive is defined) – Kinematics the motor side of the gear ratio and drive flexibility
# 7.22 Specifying calculation options
A number of machine and wind features, once defined, are available as options which can be switched on or off in some of the steady state calculations and simulations. Click the Show Options button in the bottom left corner of the Calculations screen to display the options.
Initially, all options are switched on by default, but if any are switched off for a given calculation, the setting will be remembered. An option displaying a yellow or red light will not be used even if selected – see 7.2.3. The options available are:
Energy losses (see 4.11).
• Drive train flexibility (see 4.8.2).
• Drive train mounting flexibility (see 4.9).
• Teeter restraint (see 4.4).
• Teeter motion: see hub (see 4.2).
• Passive yaw motion (see 5.18).
• Prescribed yaw manoeuvre (see 5.18.2).
• Imbalances (see 7.19) and Faults (see 7.20).
• Nacelle and tower windage loads.
• Tower shadow (see 6.3). Wind shear (see $\underline{{6.2}}$ ). Upturbine wake (see 6.4).
# 7.23 Multiple calculation setup
Users can define sets of calculations at one time, varying one or more parameters for each calculation. Select the type of calculation on the Calculations screen, and then from the main toolbar select
# Calculation … Multiple calculation.
Multiple calculation setup is available for the following calculation types:
• Generation of turbulent wind files, for a range of mean wind speeds (using MultiSetup)
• All simulations, for various parameter ranges (using MultiSetup)
• Post processing of variables with multiple locations, such as blade or tower stations (using pre-4.7 Multiple calculation setup)
For the MultiSetup feature released with Bladed 4.7, please see the MultiSetup user manual.
# 7.23.1 PostProcessing Multiple Calculation Setup
For each parameters, a list of values can be defined, provided the parameter is appropriate for the currently selected calculation, and the module from which the parameter is taken is defined.
At the top of the screen, the list of parameters is displayed along with a status label. The status label can be one of the following:
Not available: The parameter is not appropriate for the currently selected calculation, or the module that contains that parameter has not been defined.
Not active: The parameter is appropriate for the calculation and its module has been defined, but the user has not chosen to vary it as part of the set of multiple calculations. Clicking in the checkbox will activate this parameter.
X Selected: This parameter has X different values that will be used in combination with other selected parameters in the set of multiple calculations. Clicking on the Select option will allow the list of values to be edited.
Active but not available: This parameter has a set of values defined by the user, but it is not appropriate for the currently selected calculation or its module is not defined.
# Viewing calculation details
Once the lists of parameters have been defined, the set of calculations can be viewed by clicking on the View calculations details option. This will bring up a list summarising the set of calculations and presenting the parameters that vary between each one. This screen can also be used to specify the format of the path and runname to be used for the calculations. The parameter value, or a list of letters or numbers, can be used as part of the path or runname. The initial letter or number is displayed in the list box next to the parameter name, and can be changed by selecting it. This letter or number can be used in the path as a folder, or as part of the runname. The first calculation defined will use the initial letter or number displayed, and for subsequent calculations this letter or number will be incremented. It is also possible to choose which calculations to include in the set, and to change some parameters in this screen.
Once the user is happy with the set of calculations, they can be added to the current batch list by clicking on the Add calculations to Batch option.
# 8 POST-PROCESSING
Bladed provides an integral post-processing facility for analysis of calculation results. The following calculations are provided:
Basic statistics (see 8.1) to calculate the mean, minimum, maximum, standard deviation, skewness and kurtosis of a signal. Fourier harmonics (see 8.2) to calculate the fourier components of a signal at multiples of rotor rotational frequency. Periodic components (see 8.3) to separate out the periodic and stochastic parts of a signal. Extreme predictions (see 8.4) to estimate lifetime extreme loads from a sample time history. Auto spectrum (see 8.5) to calculate the auto-spectral density (frequency spectrum) of a signal.
Cross spectrum (see 8.6) to calculate cross-spectral density, coherence and transfer functions between any two signals.
• Probability density (see 8.7) to calculate the probability distribution of a signal.
• Peak analysis (see 8.8) to calculate the probability density of signal peaks and troughs.
• Level crossing analysis (see 8.9) to calculate the probability of crossing any particular threshold. Rainflow cycle count (see 8.10) for cycle counting of a signal for fatigue analysis, and calculation of damage equivalent loads. Fatigue analysis (see 8.11) to calculate fatigue damage. Annual energy yield (see 8.13) to calculate the annual energy yield as a function of mean wind speed from a power curve. Channel combination and tabulation (see 8.14) to combine and scale a number of signals (useful to produce a combined stress signal for fatigue analysis). Ultimate loads (see 8.16) to find the maximum and minimum values of loads and concurrent values of other loads. Ultimate load cases (see 8.17) to identify the load cases producing maximum and minimum values of specified loads. Flicker (see 8.18) to calculate the flicker severity due to voltage variations on the network caused by the turbine. Linear model (see 8.19) to convert the output of the Model Linearisation (see 7.11) calculation into a state-space model suitable for control design, for example using Matlab [2]. Offshore Code Checking (see 8.21)via direct links from Bladed to external third party code checking capabilities.
Post-processing calculations may be run from the Calculation screen, by selecting the calculation required and pressing Run now or Run in Batch, or from the Post-processing screen (obtained by clicking the Analyse icon on the toolbar) by pressing the Execute now or Add to Batch buttons.
# 8.1 Basic statistics
This calculation generates the mean, minimum, maximum, standard deviation, skewness and kurtosis of a signal.
Click the Select... button (see 8.20) to define the signal to be processed. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime.
Note that if the skewness and kurtosis are not required, the remaining statistics may be obtained directly from the Data View facility.
# 8.2 Fourier harmonics
This calculation generates the Fourier components of a signal at multiples of rotor rotational frequency.
Click the Select... buttons (see 8.20) to define the signal to be processed and the signal representing rotor azimuth. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case. Then choose either automatic or manual selection of parameters. Automatic selection is usually adequate, but if entered manually, the parameters required are:
Number of bins: the number of azimuthal bins to be used in the calculation (minimum 4, maximum 144). More bins will generally give a more accurate result, but do not choose more bins than the number of signal samples in one revolution of the rotor.
Number of harmonics: the number of harmonics to be calculated (minimum 1, maximum one quarter of the number of bins).
# 8.3 Periodic components
This calculation separates out the periodic and stochastic parts of a signal.
Click the Select... buttons (see 8.20) to define the signal to be processed and the signal representing rotor azimuth. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case. Then choose either automatic or manual selection of data. Automatic selection is usually adequate, but if entered manually, the parameter required is:
Number of bins: the number of azimuthal bins to be used in the calculation (minimum 4, maximum 144). More bins will generally give a more accurate result, but do not choose more bins than the number of signal samples in one revolution of the rotor.
# 8.4 Extreme predictions
This obsolete calculation was used to estimate lifetime extreme loads from a sample time history. Use Extreme load extrapolation (see 8.20) instead.
Click the Select... buttons (see 8.20) to define the signal to be processed and the signal representing rotor azimuth. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case. Then enter the following information:
Time spent at specified condition: the total amount of time during the turbine lifetime for which the conditions of the sample time history are considered representative.
Number of points per segment (power of 2, maximum 4096)
• Percentage overlap
• Window (select) Remove trends (check box)
Then choose either automatic or manual selection of data. Automatic selection is usually adequate, but if entered manually, the parameter required is:
Number of bins: the number of azimuthal bins to be used in the calculation (minimum 4, maximum 144). More bins will generally give a more accurate result, but do not choose more bins than the number of signal samples in one revolution of the rotor.
# 8.5 Auto spectrum
This calculates the auto-spectral density (frequency spectrum) of a signal.
Click the Select... button (see 8.20) to define the signal to be processed. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime. Then enter the information required, as follows:
Number of points per segment (power of 2, maximum 4096) See 8.5. • Percentage overlap Window (select) Remove trends (check box)
# 8.5.1 Options for spectral analysis
All calculations involving spectral analysis use a Fast Fourier Transform technique with ensemble averaging. To perform the spectral analysis, the signal is divided into a number of segments of equal length, each of which contains a number of points which must be a power of 2. The segments need not be distinct, but may overlap. Each segment is then shaped by multiplying by a ‘window’ function which tapers the segment to zero at each end. This improves the spectrum particularly at high frequencies. A choice of windowing functions is available. Optionally, each segment may have a linear trend removed before windowing, which can improve the spectral estimation at low frequencies. The final spectrum is obtained by averaging together the resulting spectra from each segment.
The information required is therefore as follows:
Number of points: the number of datapoints per segment. Must be a power of 2, maximum 4096. More points will give better frequency resolution, which may be important especially at low frequencies. However, choosing fewer points may result in a smoother spectrum because there will be more segments. If in doubt, 512 is a good starting point.
Percentage overlap: the overlap between the segments. Must be less than $100\%$ . $50\%$ is often satisfactory, although $0\%$ may be more appropriate if a rectangular window is used.
Window: this may be
(a) rectangular (equivalent to not using a window)
(b) triangular:
$1-\left|2\mathbf{f}-1\right|$
(c) Hanning:
$(1\,{-}\cos(2\pi\mathrm{f}\,))/2$
(d) Hamming:
$0.54\!-\!0.46\!\cos(2\pi\!\mathrm{f}\,)$
(e) Welch:
$1\!-\!(2\!\mathbf{f}-\!1)^{2}$
Where f is the fractional position along the segment (0 at the start, 1 at the end). One of the last three windows (which are all quite similar) is recommended.
Trend removal: If checked, a linear trend is calculated for each segment and removed from it before windowing, this is usually desirable. If left unchecked, the mean is calculated instead for each segment and removed from it before windowing.
# 8.6 Cross spectrum
This calculates the cross-spectral density, coherence and transfer functions between any two signals.
Click the Select... buttons (see 8.15) to define the two signals to be processed. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime. Then enter the information required, as follows:
Number of points per segment (power of 2, maximum 4096)
• Percentage overlap
。 Window (select) Remove trends (check box)
See 8.5.1
# 8.7 Probability density
This calculation generates the probability distribution of a signal, and also, for comparison, the Gaussian or normal distribution with the same mean and standard deviation.
Using the Revs at Level option, it is also possible to bin the signal based upon a signal other than time. In this case the Azimuth signal will also need to be defined. If rotor azimuth is selected for this, the load level will be binned against the number of revolutions, which is useful in gearbox and bearing design.
Click the Select... button (see 8.20) to define the signal to be processed. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime.
For each variable to be processed, enter the following data:
Minimum value: the start of the distribution (should be less than the signal minimum) (see 8.12). Maximum value: the end of the distribution (should be greater than the signal maximum) (see 8.12).
Number of bins: suggest 20. A higher number (maximum 144) will give finer resolution, provided there is sufficient data.
Remove mean (check box): If this is selected, the distribution will be representative of deviations from the mean, and so the distribution mean will be zero. If Multiple Channels is selected, the means will not be removed.
# 8.8 Peak value analysis
This calculation generates the probability distribution of the peaks and troughs of a signal.
Click the Select... button (see 8.20) to define the signal to be processed. button to define the signal to be processed. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime.
For each variable to be processed, enter the following data:
Minimum value: the start of the distribution (should be less than the signal minimum) (see 8.12).
Maximum value: the end of the distribution (should be greater than the signal maximum) (see 8.12). Number of bins: suggest 20. A higher number (maximum 144) will give finer resolution, provided there is sufficient data.
# 8.9 Level crossing analysis
This calculation generates the frequency with which a signal crosses given levels.
Click the Select... button (see 8.20) to define the signal to be processed. button to define the signal to be processed. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime.
For each variable to be processed, enter the following data:
• Minimum value: the start of the distribution (should be less than the signal minimum) (see 8.12). Maximum value: the end of the distribution (should be greater than the signal maximum) (see 8.12). Number of bins: suggest 20. A higher number (maximum 144) will give finer resolution, provided there is sufficient data.
# 8.10 Rainflow cycle counting
This calculation generates the rainflow cycle count for a stress history. A suitable stress history can be generated from the relevant loads using the Channel combination (see 8.14) calculation.
Click the Select... button (see 8.20) to define the stress signal to be processed. button to define the signal to be processed. If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime.
For each variable to be processed, enter the following data:
Minimum value: the start of the stress distribution (should be less than the minimum stress) (see 8.12).
Maximum value: the end of the stress distribution (should be greater than the maximum stress) (see 8.12).
Number of bins: suggest 20. A higher number (maximum 128) will give finer resolution, provided there is sufficient data.
Minimum range: the smallest signal range to be counted as a cycle. A non-zero value may be useful to remove the effect of any spurious noise on the signal.
If you select Calculate Equivalent Loads, enter up to 10 inverse S-N slopes to use for the calculation of equivalent loads. These are the sinusoidal loads which would produce the same fatigue damage at the specified S-N slope if the frequency of the sinusoid is as specified.
# 8.11 Fatigue damage estimation
This calculation generates fatigue damage estimates from a stress history or a previously generated rainflow cycle count, by taking account of the fatigue properties of the material. A suitable stress history can be generated from the relevant loads using the Channel combination (see 8.14) calculation.
Click the Select... button (see 8.20) to either the stress signal to be processed, or a previously generated rainflow cycle count (see 8.10) (not available for Multiple Channels). If you have selected Multiple Channels (see 8.15) you will be able to specify a number of load cases and variables to be processed in a single calculation, and the results for each load case will then be stored as additional outputs of that load case, and/or accumulated over the turbine lifetime.
Then enter the material properties as follows:
# Fatigue model
• S-N curve Select either the log-log or the look-up table option. The fatigue properties are then entered as follows: If Log-log relationship is selected: enter the inverse slope $(\mathsf{m})$ and intercept (c) of the S-N curve, such that the stress range (S) giving N cycles to failure is given by: $\,\mathsf{l o g}(\mathsf{S})=\mathsf{l o g}(\mathsf{c})\cdot(1/\mathsf{m})\,\mathsf{l o g}(\mathsf{N})$ If Look-up table is selected, click the define fatigue data look-up table button for a pop-up window. Use Add or Insert to add points to the look-up table, and enter the stress values and the corresponding number of cycles to failure by double-clicking on the appropriate table entries. The stress entries must be monotonically decreasing, with the corresponding cycles to failure monotonically increasing. Goodman correction Use the check box to enable the Goodman correction if required. If required, enter the ultimate strength of the material, as a stress.
If a stress history rather than a previously generated rainflow cycle count is to be processed, enter the following data which is required for the cycle counting:
Minimum value: the start of the stress distribution (should be less than the minimum stress) (see 8.12). Maximum value: the end of the stress distribution (should be greater than the maximum stress) (see 8.12). • Number of bins: suggest 20. A higher number (maximum 128) will give finer resolution, provided there is sufficient data. Minimum range: the smallest signal range to be counted as a cycle. A non-zero value may be useful to remove the effect of any spurious noise on the signal.
# 8.12 Setting bin limits
Several calculations require minimum and maximum values to be entered for the bin range. These values should encompass the entire data range.
For automatic calculation of suitable bin limits, set the minimum and maximum to be equal, or leave them both blank.
Alternatively, the actual minimum and maximum of the signal may be found by running the Basic statistics (see 8.1) calculation.
# 8.13 Annual energy yield
This calculation generates the annual energy yield of a turbine from either a steady state power curve (see 7.8), a dynamic power curve as output by a previous Annual Energy Yield calculation or a series of power production simulations, as a function of annual mean wind speed. The wind distribution (see 6.11) could be defined either as a Weibull, Rayleigh, or a user defined distribution of wind speeds. The wind and wave scatter diagram option could also be used to define the distribution.
Click the Select... button (see 8.20) to define the power curve to be used. Then enter the following data:
Minimum wind speed: the cut-in wind speed for the turbine
Maximum wind speed: the cut-out wind speed for the turbine
Turbine availability: the average availability, assumed to be uncorrelated with wind speed.
Define the wind speed distribution (see 6.11) by clicking the Define button. If desired, select the option to scale the wind speed distribution by a series of factors; the energy yield will then be calculated as a function of annual mean wind speed while retaining the shape of the wind speed distribution. In this case, enter
• Starting annual mean wind speed End annual mean wind speed Annual mean wind speed step
Energy yield will be calculated for each mean wind speed in this range
# 8.14 Channel Combination and Tabulation
Use this calculation to combine and scale a number of signals. For example, this can be useful for generating a stress history for a particular point in the turbine structure, by expressing it as a linear combination of the various loads acting on the component. The stress history can then be used to calculate fatigue (see 8.11) damage. The scale factors used for the linear combination should be such that the resulting signal is in the same stress units as are assumed in the fatigue damage calculation.
If Combine variables across different load cases is selected, it is possible to combine up to six signals, which can be from different calculations – Section 8.14.5. Usually however, the signals to be combined will all come from the same calculation, so this option should be left unselected. This will give three options:
Channel Combination, which allows the choice of:
an Old Style of channel combination which allows channel combinations from Bladed versions 3.51 and older to be re-used (click Import details... and select whether to import the channel combination details, the list of load cases, or both), and
a new style of channel combination (if neither check-box is selected) which provides much greater flexibility and ease of use.
Channel Tabulation, which allows a number of variables to be output to a single file.
Matrix Combination, which allows multiple outputs to be produced from multiple inputs through matrix multiplication.
In each case, first define the Combined Signal File No (from 1 to 260; up to 260 combined signal files may be generated for any load case), and enter a description for the file.
In the Tabulation case, there is a choice to keep the output files with the results of each original calculation, or to place them in a new location. If New directory is chosen, select a directory which will be used to replace any part of the file path which is common to all the runs which are to be processed.
Then click Channels and Load Cases to define the list of variables to be processed, and the list of runs for which the processing will be repeated. Each run should contain all the variables to be processed.
# 8.14.1 Multiple processing option – Channel combination
Select Variables, and click Add Variable to define the variables to be processed. Click Add Calculation to set up an equation for processing the variables. Any number of equations may be defined. Each equation will generate a new intermediate or output variable, which will be denoted #1, #2 etc. Each equation may process any of the input variables (which will be denoted $\boldsymbol{\mathfrak{G}}1$ , $^{\Phi2}$ etc.) together with any of the intermediate variables #1, #2 etc. Any of the intermediate variables which are given an output name will be output to the results file.
To edit an equation, highlight any part of it and use the editing options in the panel on the right. These can be used to enter specific values, such as input $(\Phi)$ or intermediate (#) variables or constants, to apply binary arithmetic or logical operators, or to apply unary operators. These include the conditional operator IF, and user-defined nonlinear functions, which are entered as look-up tables – the calculation will use linear interpolation between table values, and for values beyond the end of the table the nearest point in the table is used.
To create a calculation that refers to its own previous value simply use the # value from the current equation. You do not need to use the prev() function here as the created variable already refers to its previous value when used in its own equation: So if you use $\"\#1+1"$ in the equation of #1, this will add $^{+1}$ to its previous value, which is initialised to zero at the start.
For example, in order to find out if and how many times a variable has been greater than a specified value for an uninterrupted period, you can do something of this sort:
In this example, we are calculating how many times the pitch angle exceeds 5 deg (0.0873rad) for an uninterrupted 5 sec (100 time steps).
Figure 8-1: Multiple processing: Channel combination
Use Resolve Angles to generate a whole set of equations representing the resultant of two orthogonal loads resolved into different angles. Select the $_{0^{\circ}}$ and $90^{\circ}$ loads from the list on the right of the screen – for example, the hub My and $\mathsf{M}z$ loads could be selected. Then define a range of angles to resolve over, and provide a name for the output variables if they are to be output. If the $_{0^{\circ}}$ and $90^{\circ}$ loads are $\mathsf{V}_{0}$ and ${\sf V}_{90},$ then for each angle an equation is created which generates $\mathsf{V}_{0}\,\mathsf{c o s}\theta\,+\,\mathsf{V}_{90}$ sin. The output name for each variable (if specified) will automatically have the appropriate angle appended.
Finally click Load Cases to set up a list of load cases for which the resulting set of channel combinations will be repeated.
# 8.14.2 Multiple processing option - Tabulation
Select Variables to define the list of variables to be tabulated. For each variable, select the desired units to be used for the tabulation.
Finally click Load Cases to set up a list of load cases for which the resulting set of channel combinations will be repeated.
When the calculation is run, the output may be chosen to be in ASCII or binary format. Selecting ASCII format will tabulate the selected variables as columns in a tab-delimited text file, with column headings describing each variable.
# 8.14.3 Multiple processing option – Matrix combination
The matrix combination calculation is available for example to set up an influence matrix, by defining a set of output loads as a linear combination of a set of input loads.
Select Variables, and click Add Variable to define the variables to be processed. Click Add Matrix to set up a matrix for processing the variables. Click Add Output to specify the outputs from the matrix combination. Any number of matrices may be defined. The size of the matrix is determined by the number of input variables and the number of outputs.
To edit a matrix, either type directly into the matrix grid or copy and paste into this grid.
Finally click Load Cases to set up a list of load cases for which the resulting set of matrix combinations will be repeated.
Once a matrix combination has been set up, the Vector Combination option becomes available. This allows outputs from a matrix combination to be processed further as a vector. Click Define details to set up the vector combination.
The vector combination is set up in a similar way to the channel combination. Each equation may process the vector of outputs from each matrix (denoted M1, M2 etc) to produce a further vector of outputs (denoted V1, V2 etc). Click Add Variable to defined additional variables which may also be used in the equation if required.
# 8.14.4 Multiple processing option – Old Style channel combination
Select Variables and click New to define a new output signal, and enter a description for it. Select one of the available units for the signal if appropriate.
For each output variable, click Add Variable… to select the required input variables. Specify any factor, offset and other unary operators to be applied to each input signal before it is combined with the other input signals. The Factor is applied first, then the Offset, then any Unary operators are applied. If several unary operators are specified, they must be separated by | (a vertical bar). They will be applied in order, starting from the right-most. Allowed operators are:
SIN sine (of a value in radians)
COS cosine (of a value in radians)
ABS absolute value
SQRT square root
INV reciprocal
$+\mathsf{n}$ Add a number n
$-\mathsf{n}$ Subtract n
$\times\mathsf{n}$ Multiply by n
$/n$ Divide by n
^n Raise to the power n
where n is a real number, which may be negative. For example, SIN|x-3.2|^-0.2|ABS| $\mathbf{+0.1}$ would calculate sin(-3.2( $|x{+}0.1|^{-0.2})\rangle$ ) from the input signal x.
Then select how the resulting input signals are to be combined. This may be by • addition,
• squaring and adding, then taking the square root of the result, or
• multiplying.
Finally click Load Cases to set up a list of load cases for which the resulting set of channel combinations will be repeated.
# 8.14.5 Single channel combinations
From Channel combination and Tabulation, select Combine variables across different load cases to generate a new variable not attached to any existing load case. You will then be asked for a directory and run name for the output when you start the run. In this case, click the numbered channel selection buttons (see 8.20) to define up to six signals to be combined. The signals may be from different runs or load cases if desired. For each signal, enter a scale factor and an offset. Enter a description for the combined signal, and choose units if appropriate.
The scale factors (ai) are applied before the offsets (bi), i.e. the result of the calculation can be expressed as:
$$
\mathbf{y}=\sum\left(\mathbf{a}_{\mathrm{i}}\mathbf{x}_{\mathrm{i}}+\mathbf{b}_{\mathrm{i}}\right)
$$
if Simple Addition is selected. If Square-Add-Square Root is selected, the result is
$$
\mathbf{y}=\sqrt{\sum\left(\mathbf{a}_{\mathrm{i}}\mathbf{x}_{\mathrm{i}}+\mathbf{b}_{\mathrm{i}}\right)^{2}}
$$
Enter a description for the combined signal, and select one of the available units for the combined signal if appropriate.
# 8.15 Multiple Processing
A number of post-processing calculations allow a whole list of variables to be processed across a whole list of load cases. If appropriate, the results are accumulated over the turbine lifetime (or any other desired period) assuming a particular wind speed distribution, and a frequency of occurrence for transient events. Otherwise the results are stored as additional outputs for each load case.
In the window for the required post-processing calculation, click Multiple Channels to bring up the Multiple Processing window. Use the Add Load Case and Add Variable buttons to build up the required lists of load cases and variables. Add Many Load Cases allows all the calculations in a selected directory (including subdirectories if desired) to be added in one go. In most cases, multiple variables can be selected in one go by dragging over a number of variables, or holding down the Control key and clicking on the required variables.
It is important to ensure that the output (see 7.21) specifications for all the load cases are the same, i.e.
the same outputs and the same blade and tower stations have been requested.
Load cases tab: If accumulating results over the lifetime, the Load Cases window is loaded with all the relevant load cases, and a weighting is applied to each load case to reflect its contribution to the lifetime. For transient (T) load cases such as stops and starts, enter the Occurrences per Year of this load case. For stationary (S) cases, i.e. load cases representing an ongoing situation or state of the turbine, there is a choice three weighting methods:
• “One run per bin”: Specify the wind speed bin represented by this load case, by giving the bottom and top of the bin. Eventually the whole wind speed range, starting from $0~\mathsf{m}/\mathsf{s},$ should be represented. The wind regimecurrent (see 6.11) also needs to be defined so that the appropriate weightings can be calculated.
• “Hours per year”: In this case, the number of hours per year for which each load case is representative is specified. “Pre-defined bins”: In this case, the bins are defined independently, as is the wind regime (see 6.11), and the details of each load case are automatically interrogated to allocate it to one of the bins. If a bin contains more than one run, the time in the bin is split equally between each simulation in the bin. For offshore turbines, a scatter diagram (see 6.15) may be used if desired to define bins representing the joint probability distribution for wind speed, wave height and wave period.
At the bottom, enter the turbine lifetime. Note that it is possible to treat a stationary load case as transient, or vice versa, by clicking on the $\mathbf{\Delta}^{\bullet}\mathbf{s}^{\prime}$ or ‘T’ indicator. At the bottom, enter the turbine lifetime, and ensure a wind speed distribution (see 6.11) is defined (not required if all the load cases are transient). The length of transient simulations will not be included in the turbine lifetime.
Variables tab: If available, for each variable, enter the number and range of the bins to be used. If the Minimum and Maximum are both left as zero, a bin range will be calculated which is suitable across all the load cases. For rainflow cycle and fatigue analysis, it is possible to specify a minimum range below which cycles are not counted.
# 8.16 Ultimate Loads
This calculation scans a number of load cases to find the maximum and minimum values of a specified set of variables, and records the simultaneous values of all the other variables in the list.
From the Ultimate Loads window, click Define Channels and Load Cases to open the Multiple Processing window. Use the Add Load Case and Add Variable buttons to build up the required lists of load cases and variables.
For users with the Advanced Post Processing module, a list of Load Case Groups can be set up along with their corresponding safety factors. A check box on the Load Cases tab enables these to be assigned to the load cases. During processing the safety factors will then be applied. Safety factors may also be assigned to individual variables, and these will override the group safety factor.
Each load case should have a unique identifier. If Load Case Groups have been defined, the identifier will attempt to set itself to a sensible default, which the user is free to change if required. For each variable, specify whether its maximum and minimum are to be found.
If Load Case Groups have been used, an additional check box on the Variables tab allows the safety factor associated with the extreme load to be reported in the extreme loads tables.
If a blade manufacturer has provided a load configuration file, this can be used to produce custom output for that particular blade.
An additional feature available to users with the Advanced Post Processing module is to combine blade loads across all blades. With this option the maximum and minimum loads across all blades will be reported.
The output from this calculation consists of an ASCII text file with many rows and columns. The file name will be $