Abstract:
Wind forecasting for the commercial-scale wind farm is vital for grid management issues,
energy trading, tariff adjustment, and maintenance issues. Forecasting of the wind farm
located in the complex terrain causes a major challenge. The challenge for the prediction
of wind resources for that wind farm immensely increases if inter-farm wakes affect wind
farms and under the severe variations of seasonal changes. The Weather Research and
Forecasting (WRF) model is used for the mesoscale wind resource forecasting for the
wind farm under the effect of immense and complex wakes from neighboring wind farms.
The test case wind farm is located in the complex terrain of the Jhimpir wind corridor,
Sindh, Pakistan. Simulation for forecasting was done for two cases i.e. Without inter-farm
wakes and With Inter-farm wakes, during the prediction of wind resources for the wind
farm. A significant reduction in error assessment parameters has been observed for Case
2. For the month of June (Summer), the mean absolute errors in wind speed prediction
were reduced by 7.7 %. In the month of January (Winter), 14 % of error reduction in mean
absolute error was observed. The power predicted was improved by 15 % and 26 %, for
June and January, respectively. However, the forecasting skill of the WRF is deteriorated
in the winter. The Pearson correlation factor asses the forecasting skill of the WRF for
wind power prediction, the value of correlation in June is 0.82 and for winter its value is
0.27.
