A Cross-Validated Combined Forecasting Method And Its Application in Power Wind

Abstract

A Cross-Validated combined forecasting method is proposed to retain more data information of wind power at each time level, so that the wind power forecasting results at different time scales meet the aggregation constraints. We used the actual data of a wind farm in Shandong Province to verify the effectiveness of the method. The results show that compared with the existing method, forecasting with Cross-Validated combined increases accuracy over conventional forecasting, particularly in different time levels with larger sampling interval.