Assessment of the main methods for AEP prediction in wind farms

In this thesis the accuracy of the main program (WAsP) used nowadays for the prediction of a wind farm energy production, together with the analysis of a 2D RANS code for the same goal, is investigated. In particular this work aims to understand, how to predict in the best way the AEP of a wind farm, the accuracy of different methods to reach this goal, and the pro and cons about using WAsP or a RANS code for the study of a wind farm. WAsP is used for the study of 2 wind farms and the results, in terms on energy production, are compared with the real data. A sensibility analysis of the program inputs is done as well to increment the accuracy of the solutions. A RANS code is also used to predict the energy production of the wind farm, and the results are compared with the real data and with the prediction of WAsP. A main result obtained is the wrong prediction of the energy distribution among the turbines, and the lower accuracy of WAsP, compared with a good accuracy of the 2D RANS code, but with higher computational cost. The good prediction of the code can be the incentive for a further improvement of it, and the results obtained can be used to suggest faster way of calculation according to the needs