Combined approach for short-term wind power prediction: A case study of the east coast of China

Power systems with high wind power experience increased variability and uncertainty. Therefore accurate forecast technique is essential to cope with the uncertainty in day-ahead electricity market. Statistical forecast model is considered as a powerful technique for wind-power forecast, however, the forecast accuracy significantly drops as the forecast horizon grows. Focusing on wind speed persistence, a hybrid multi-step-ahead (HMS) method is proposed, which composed of direct multi-step-ahead prediction (DMS) approach and indirect multi-step-ahead prediction (IMS) approach in time-series forecasting. To validate the effectiveness of the proposed method, one-month period of real data, which is collected from five operating wind farms in the east coast of China, is used for test. The persistence model (PM) is used as a benchmark with the new forecast method. Test results show that the proposed method is accurate and effective.