Morten

Dear Vinh Le Thanh, I inserted the coordinate into Google Earth, and I agree that the roughness pattern at this site is quite complex with many small tiles and a large variation in surface roughness. From the pictures I also saw that fields are flooded in part of the year, and open water surfaces will reduce the roughness roughness. For WAsP applications you should, however, apply a roughness typical for the entire year as this is more relevant for the yearly energy production. It would be a big job to digitize a roughness map for this site. Instead, you could try to press ‘File| Import| from Web-database| GWA Map Warehouse – roughness’ in the WAsP map editor to download a roughness map. The resolution is not ideal, so you may need to add more details for the area close to your windfarm and met mast. But you don’t have to edit the roughness far from the sites of interest, as these details should not affect the WAsP results much. With best regards, Morten Nielsen

The Lib Interpolation applies simplified version of the method described in Nielsen, M. (1999), A method for spatial interpolation of wind climatologies. Wind Energ., 2: 151-166. doi:10.1002/(SICI)1099-1824(199907/09)2:33.0.CO;2-5 where it is referred to as LibInt LT.

It depends a little on the job you need to do. Maybe you want to study a particular problem occurring for a specific wind direction or maybe you need to check turbine safety in general? For the latter you could try an IEC 61400-1 site assessment, which includes a list of wind conditions to check. You start by creating and selecting a group of turbines sites and their hub height. You also need to create an extreme wind climate by the WAsP Climate Analyst and import this in the WEng project. Then you open ‘Tools| Prepare data for WAT’ from the main menu. This tool will calculate the fifty-year extreme winds for all turbine positions plus tables of wind conditions (shear, turbulence, inclination angle, etc.) for a number of wind directions. You can direct these results to an Excel file and use them as they are or import them in WAT to complete the IEC site assessment. Cheers, Morten

The basic wind input to the flow model of WAsP Engineering is a single wind condition. You may insert multiple winds in the project, but usually you have to select only one of these as reference for the calculations. Quite often you also need to specify a height and a site or a group of sites. Some of the ‘tools| scripts’ will iterate over all wind in your project, e.g. ‘Summary site report…’ and ‘Turbulence report…’, or even create a number of winds with different direction and report for these, e.g. ‘Wind speedup and deflection’. WEng can also operate on extreme wind climates (EWC). This is a little like the mean wind climates in WAsP, i.e. you specify an observed EWC for the reference site and then predict the EWC for other sites. There are some video demonstrations of basic operations in WEng at https://www.youtube.com/playlist?list=PLrqli3B3J9m4NtU7d373DVWFQMI2iinpb Cheers, Morten

Hi LinkeshD, You can create a WAsP WTG file by clicking on 'Tools| WAsP Turbine Editor' from the main menu inside WAsP. I usually start by preparing a performance table inside Excel with three columns (speed, Power, Ct) and then copy this table (without headers) to clipboard. Back in the turbine editor I select the 'Enable Edit' option and adjust the 'Table size' to the number of data points. Then I enter the upper left cell of the empty table and paste the performance table from the clipboard. Next step is to edit the 'description' string, rotor diameter, hub height and air density. If multiple performance tables are available, I add them one by one and finally save the WTG file to disk. Starting from WAsP 12 we can interpolate and extrapolate power performance tables to air densities at individual turbine sites. To reduce uncertainty it will be best if the starting point is a file with performance tables for air densities close to the local conditions. Cheers, Morten

A note for third-party readers: The assumption in my first reply was wrong. The real problem is that WAsP applies an automatic routine to detect whether a turbine is pitch or stall regulated. It simply consider a turbine to be pitch-regulated when the power curve has at least three points equal to the maximum. Unfortunately, this routing fails for measured power curves with a bit of noise in the power-curve data. It should not be a problem with power curves supplied by manufactures.

Hi Ebazus, Normally WAsP Engineering calculates for specific wind conditions, not sectors. However, the tool called 'prepare data for WAT' produces a result table for the same number of sectors as defined in the supplied wind atlas file, which is a result from WAsP. These calculations are made with one reference wind for each sector with a wind direction equal to the center angle of each sector. The reference winds are generalized winds, i.e. for flat terrain with surface roughness z0=0.05m, and they may deflect over real terrain. Thus the flow inclination angle (or tilt) are calculated for a case where the local wind direction is a bit off the sector center angle, especially in complex terrain. Cheers, Morten

Hi Roella, The Weibull probability distribution describes the mean wind speed at the turbine position. If you have an observed time series of wind speeds at the right position and height above terrain you can fit a Weibull distribution to the data and then estimate the annual energy production by a probability-weighted integral of the power curve of all wind speeds. With a program like WAsP you can observe the wind climate at a reference mast, then use a flow model to predict wind climates at your turbine sites and finally use these turbine-specific wind climates for the AEP estimates. WAsP operates with different Weibull distributions for individual wind sectors as the relative speedup between reference site and sites of interest and the corrections for wake effects of neighbouring turbines depend on wind direction. So the AEP calculation is a more detailed than in the turbine editor. In any case the observed wind data should cover a whole number of years to avoid seasonal bias. Best regards, Morten

Hi Victoria, Good to hear that you were able to solve the problem. Cheers, Morten

Hi Victoria, Mark forwarded this question to me, but I am afraid that I cannot help much. All I know is that the flow model of WAsP Engineering operates with a local coordinate system which must be Cartesian with metric coordinates. However, if you try to connect to our servers you also need to describe the map projection and datum for your map projection, e.g. UTM 30 and WGS84, so the system can transform your local coordinates to geographical coordinates. I am not familiar with WindPro. It might operate with the same map conventions as WEng, but it is also possible that some information needed by WEng is provided automatically without troubling the user. Maybe you have a user guide for WindPro, which can show you the correct way. Not sure what a 'position type' is, so it is probably a term specific for WindPro. Cheers, Morten

Hi, The ‘Prepare data for WAT tool’ calls some WAsP DLLs to add mean climate predictions (Weibull distributions) to the result table for each turbine site. That is why it asks for a wind atlas file and a map in vector format. The format of the wind atlas file has changed over the years and I suspect that this could be the reason for the problem. I did some experiments and it seems like the WAT launcher tools reads the old LIB format without problems. It also reads a GWC file exported from WAsP 11 correctly but I did not manage to get it to read a GWC file exported from WAsP 12. The GWC format differences are the barometric reference data added in WAsP 12.0 and parameters for the background flow profile added in WAsP 12.1. Since you write about temperature and pressure information, I am guessing that you are in fact using WAsP 12, so your problem is probably related to the modified GWC file format. A work-around solution could be to apply the LIB format instead. If you are worried about the accuracy, then try to compare the Weibull distributions in your WAsP project and the ones which are written in the input file for WAT. I am not sure about the technical reason for the problem, and I rather not ask the programmer who is on vacation right now. Maybe WEng 4 can only work with the DLLs from WAsP 11 or maybe it gets confused when you, like me, both have WAsP 11 and WAsP 12 installed on the same PC. Anyway, thanks for reporting this. I shall forward the message to the programmers. Cheers, Morten

Hi Stefan, The extrapolation is done in two steps. First we correct the wind speeds of the reference points of the original power curve. For a decrease in air density, this shifts the power curve to the right, but we take care of not moving the point at the cut-out wind speed. We then interpolate the shifted power curve and find new values at wind speeds with regular spacing. I have seen one case where this interpolation failed because the interpolation function was forced into unrealistic undulations. The problem was an unconventional additional reference point near cut out inserted by the user. It did not use to disturb WAsP but fooled the air density correction method. If you like me to investigate the problem then please send a WAsP workspace file illustrating the problem to waspsupport@dtu.dk. Best regards, Morten

Dear Sinisa, IEC 61400-12-1 on power performance measurements has an informative Annex M discussing how to correct a measured power curve for effects of turbulence. However, it does not depend on ratios of standard deviations of turbulent fluctuations in different directions (sigmaU, sigmaV and sigmaW). I am therefore guessing that your actually ask about the turbulence structure correction parameter CCT, which is specified in IEC 61400-1 on turbine safety. Actually you can calculate this in two alternative ways – by terrain assessment or by a formula based on sigma-ratios. You can make either assessment with our WAT program, which is a postprocessor of results from WAsP and WEng. WAT calculates CCT by probability-weighted averages of standard deviations of turbulence components in a number of sectors as calculated by WEng. The data for WAT is calculated by a tool called ‘Prepare data for WAT’ inside WEng 4. Learn more in videos no. 2, 3 and 6 at https://www.youtube.com/playlist?list=PLrqli3B3J9m4H0hOHzumZgkqJ5UXlHvYA I am not sure, but I think the preparation of data for WAT worked the same way with WEng 3.1 If not, you may have to upgrade or make equivalent assessments manually using the turbulence calculated for a number of wind directions inside WEng. With best regards, Morten Nielsen

WAsP CFD is not free. However, if you are a student at DTU then please ask your supervisor to contact the WAsP group about your needs.

The file format of the WAsP workspace file has changed, because of the new air-density corrections of the power curve in WAsP 12. We will prepare an update to Fuga, which supports the new workspace format. If you are in a hurry you could work with input from WAsP 11. Regards, Morten Nielsen

Hi Seenu, The ruggedness index (RIX) is a measure of the terrain complexity in the area around a site. Delta-RIX (dRIX) is the difference between the RIX numbers at the site of interest and at the reference mast. Terrain complexity leads to uncertainty in linearized flow models, such as the one used in WAsP. But errors related to generalization of the wind climate at reference mast data and errors related to climate prediction at the site of interest will often compensate each other when dRIX is close to zero, i.e. when the RIX numbers for the two sites are comparable. A positive dRIX value means that the terrain around the site of interest is more complex than near the reference mast and the model is likely to over-predicted the wind speed. Best practice is to install a reference mast at the site of interest, let it measure for 2-3 years, and use the local data for the wind resource estimate. You could also try the WAsP CFD model, which generally gives better predictions in complex terrain than the default flow model in WAsP. Best regards, Morten

Hi Seenu, I am not sure what the 'tiresome refresment' problem is, but I suppose it is a technical matter and I think it will be better to write to waspsupport@dtu.dk than to discuss at the WAsP forum. Please describe the problem in sufficient detail in your mail to the help desk and include a test file or similar information so we can investigate the problem. Cheers, Morten PS: We are closed between Christmas and New Year.

Hi Phil, The programmer is not here today, but to save time I suggest that you already now send a mail to our help desk at waspsupport@dtu.dk Please include the map you are working with in a state where we can see the problem. If you are not able to save the map before the problem occurs, then please send some screen dumps, or maybe a video, so we can reproduce the problem here. Cheers, Morten

Hi Phil, I just asked our Map Editor programmer, who reminded me that roughness consistency checks are optional and may be deactivated under 'Tools' in the main menu on the main Map Editor window. Cheers, Morten

Your assumption is correct.

The main principle is first to compute the two types of statistics mentioned above based on the observations and then to select the A and k parameters which give the same statistics when calculated by the Weibull probability model. This is done for every wind sector. The situations with calm weather need special treatment, since we cannot trust the wind direction sensor when the wind speed is close to zero. Calm data are therefore distributed proportionally over the wind sectors using the frequency of occurrence determined by winds above the calm threshold.

Good morning, Contours with 5-m vertical separation could be a good general choice, but unless you are very pressed for time you should also inspect the result in the map image and maybe make some adjustments. Try to zoom in and check whether the generated contours have lots of parallel line segment in the cells of the original SRTM map. If so, you could probably do with less vertical separation as WAsP will get almost the same height information when it interpolates your contour map. On the other hand, you risk losing information on hill top if you set the vertical separation too low, e.g. to 20m – that is unless you manually add exact spot-height information in the map editor, which actually is recommended. Read more in the http://www.wasp.dk/support-and-services sections called ‘maps’, ‘SRTM-maps’ and maybe also at http://www.wasp.dk/support-and-services/faq#weng__working-with-maps-in-weng The horizontal length scales of the terrain features is also important for the flow at hub height. The near-surface flow will always follow the terrain, but undulations with small horizontal length scales will have less influence on the flow at hub height than undulations with a long length scale. So you might need less vertical resolution lower in a landscape with big round hills. A very large contour map fill will slow down the WAsP map pre-processing. A good tip is to have more resolution near the points of interest than at far distance. You can do this with the map editor, e.g. by inserting a high resolution map for the wind-farm area inside a map generated from SRTM downloads. Click ‘File> Import>SRTM’ in the map editor to open a SRTM download window. This has ‘File location’ at the top from where it is possible to change the SRTM download link or use maps stored locally. The default link is https://dds.cr.usgs.gov/srtm/version2_1/SRTM3/ which point towards version 2 maps. You can also work with version 3 maps, but then you need to register at NASA and log in with a user name before you download with the Map Editor. Read more about this in the Map Editor help file. Cheers, Morten

WAsP is fitting Weibull distributions to data by matching two types of statistics: 1) The mean of the cube of the wind speed = 2) The probability of winds above the observed mean wind speed P_model(u>u1) = P_obs(u>u1) where u1 = This procedure does not guarantee a match of the mean wind speed and the fit is generally less accurate at low wind speeds than at high wind speeds. The focus on the high wind-speed range is intentional, since this is where turbines are operating. Cheers, Morten

Hi, Students at the Danish Technical University (DTU) can learn about WAsP in course 46200 Planning and Development of Wind Farms. This three-week 5 ETCS credit course is held each year in January and it covers the same topics as the WAsP course for wind-energy professionals - plus useful knowledge on wind project economics, grid connection, environmental impacts and IEC site assessments. If you are not a DTU student and can't afford the online course, you should read the WAsP help file section called Quick Start Tutorial. It explains the WAsP user interface and demonstrates how to compose a WAsP project using sample input files installed on your PC together with WAsP. Be sure also to install the programs called WAsP Climate Analyst and WAsP Map Editor. These programs will help you when you need to prepare wind statistics and maps for your own projects in the file formats preferred by WAsP. With best regards, Morten Nielsen

Hi CA1, Thank you for the feedback. While writing these I also saw your email to WAsP support and realized that you might have a problem with the WAsP installation. Still, I will keep this general advice, for other readers at this forum. First it is important that turbine positions and the map inside WAsP apply the same projection and datum. If not, you must either transform the map or the turbine site list. This can be done outside WAsP using either the Map Editor (tools>transform>projection) for the map, or the WAsP Projection Transformer (File>transform XYZ file) for the turbine positions. When you input the aligned data, the turbine positions should display correctly inside WAsP. WAsP needs to know the applied map projection and datum in order to make a translation before displaying the information correctly in Google Earth. You can right-click on the map and select ‘Set map projection’ if the projection or datum is unknown, but sometimes projection is included in the map information coming from the Map Editor. As far as I know WAsP will not try to translate the map relative to the turbine positions, i.e. you can use ‘Set map projection’ to provide missing information on projection and datum but not to fix a difference between map and turbine coordinates. This must be done before you load the data into WAsP as described above. I must admit that I too have observed differences between the turbine positions displayed in Google Earth and turbines visible in the GE satellite images. I am not sure how big these should be before it is alarming, but I think you can expect precision around 10 m. If you choose the right projection but false datum, you may see differences on the order of 100m. If you have checked everything and still see significant offsets, then please send a sample project to WAsP support for inspection. Turbine positions in GE are sometimes displayed by 3D models and sometimes by icons. This depends on the zoom factor. Cheers, Morten