Short-term wind speed forecasting model based on D-S evidence theory
Zhou Jun · Dianli zidonghua shebei · 2013
A combined short-term wind speed forecasting model based on D-S evidence theory is proposed.The forecasting models of time series,BP neural network and support vector machine are adopted to respectively forecast the wind speed.Based on the analysis of forecast errors,D-S evidence theory is applied to fuse these three models.The wind speed data for several days before are taken as the fusion samples to calculate the corresponding basic trust distribution functions,which are then fused.The results of fusion are taken as the weights of the wind speed forecasting model and the wind speed of the day to be forecasted is calculated.Simulative results show that,the proposed combined forecasting model has smaller forecasting error and better effect.