Wind Resource Estimation Using a Fine-Scale Weather Model a report by Anna Hilden Meteorologist, MSc, Country Manager for Denmark at StormGeo AS
Predicting the energy output from a wind farm in complex terrain has long been a challenge. The wind conditions can vary significantly across a wind farm area in the mountains or near a rugged coastline, and even if you have measurements from one or two meteorological masts at the proposed site, these will most often not be sufficient for obtaining an accurate estimate of the potential production. At early stages of wind farm development, when a developer first picks the sites that are promising for wind energy production, the challenge is even greater, as the nearest meteorological stations may be hundreds of kilometres away where the wind conditions can be very different.
Within the last few years a new approach to estimating the wind resources in complex terrain has emerged. This approach makes use of dynamical atmospheric models – weather models – to simulate the wind and weather over an area in very high horizontal resolution for a certain time period in the past. The result is a reproduction, or hindcast, of the state of the atmosphere at any given point in space and time during the simulation period. From this data, all sorts of information about the wind conditions and potential for wind power production can be derived, including wind resource maps, wind statistics, expected power production and more.
The idea of using weather models for this purpose is not new; however, it is only in the past few years that the price and availability of computer resources have made these so-called mesoscale modelling methods applicable in practice. Several companies and research institutions now use them and mesoscale modelling is gradually gaining acceptance in the industry as a useful and necessary tool for wind resource estimation in complex terrain.
An Example
As an example of what mesoscale modelling can provide, Figure 1 shows a map of the average wind speed over a mountainous area in Norway at a level above ground corresponding to a typical wind turbine hub height. The area is approximately 200 kilometres wide and the horizontal resolution of the hindcast is 1 kilometre (for commercial reasons the details are not disclosed here). It is clearly seen that the
Anna Hilden works with the private weather service provider StormGeo, where she holds positions as a Country Manager for Denmark and a Key Account Manager for customers in the renewable energy sector. She has previously worked as a Product Manager for wind farm SCADA systems at Vestas Wind Systems and as a Software Developer and Project Manager at the Danish Meteorological Institute. Mrs Hilden holds a master’s degree in meteorology and mathematics from the University of Copenhagen.
E:
anna.hilden@stormgeo.com
wind speed varies considerably across the area, corresponding to the features of the terrain – mountain ridges, valleys, forests and coastline.
From a map like this, a wind farm developer can quickly determine which locations have the best potential for wind energy production. When combining the data with other aspects, such as regulatory conditions, land ownership, the proximity to the power grid and to roads, etc., the wind farm developer can then select a site or smaller area for further investigation.
As the next step, the developer will then extract time series of the simulated wind and other weather parameters in different heights at the locations of interest, maybe even at individual possible turbine positions. Based on these so-called virtual measurement series, he or she can investigate wind speed distributions and wind roses and calculate the expected wind power production from each turbine assuming a given wind turbine make and type. If wind measurements are available from one or more locations in the hindcast domain, these measurements can be used to calibrate the time series, as described below. To ensure that the results are representative in the long term, data can be combined with long-term data series of a coarser resolution, as explained below. Wake effects can also be accounted for.
Figure 2 presents examples of the information that such an analysis can provide. The examples are for a 20-turbine wind farm. As input to the calculations, the analyst has specified the locations of the wind turbines, as well as the make, type and hub height of the turbines.
The top graph in Figure 2 shows the expected average annual production from each of the 20 turbines. Note that the expected production varies considerably across the wind farm, with the most productive turbine generating nearly 15% more energy than the least productive one. In this case, one would probably try out a different wind farm layout or skip some of the turbines altogether.
The middle graph in Figure 2 shows how the production from the proposed wind farm would have been in each year, 20 years back in time. The figures are calculated using the hindcast data in combination with long-term data sets as described below. Information like this is important for wind farm owners and financing institutions, as it tells them how much their income may vary from year to year due to changing wind conditions.
In the bottom graph it is shown how the production is expected to vary over the year. Since the demand for power changes according to time of year, so does the value of the wind energy produced. Depending on power purchase schemes, this information may be relevant when estimating the possible revenue from the wind farm.
76 © TOUCH BRIEFINGS 2010
Wind
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124