New Energy Output Forecasting Method Considering Multi-zone Numerical Weather Forecast and Error Identification

Abstract

In recent years, new energy has achieved rapid development in the whole world. However, with the gradual increase in the installed capacity of new energy, because of the uncertainty and volatility of new energy output, new energy has brought many impacts and challenges to the operation of the power system. In order to guarantee the safe and stable operation of the power system, the power system has to provide sufficient flexibility to cope with the volatility and uncertainty of the new energy output. Improving the prediction error accuracy of new energy output can effectively reduce the uncertainty of new energy output, so as to reduce the demand for power system flexibility, reduce the operating cost, and improve the system operation reliability. This paper utilizes the data mining technology to establish the prediction error identification module. Based on the data feature extracted from the wind power prediction input data, the potential abnormal wind power prediction error can be identified, and the original wind power output prediction result can be corrected, so as to improve the predication accuracy of the wind power generation. At the same time, in the prediction and identification process, the wind speed prediction data and historical wind power output data of neighbouring regional wind farms will also be utilized to improve the prediction accuracy of wind farm output in this region. The predicated time scale is 12 hours ahead of time, and the predicted result is the wind power output of the next 12 hours (a predicted resolution of 1 hour, and a total of 12 predicted values).