Medium and Long-term Forecast of Wind Power Trend Based on Regional Similarity

Abstract

Accurate and efficient medium and long-term forecast of wind power can provide technical support for efficient development and utilization of wind resources. Taking into account the regional characteristics of wind resources, the regional similarity factor is introduced into the study of wind power forecasting, and the long-term dependence of wind power, the Long Short-Term Memory method is selected for medium and long-term forecasting of wind power trend, a case study is carried out in five provinces of Northwest China. The results show that the error is reduced by an average of 20.80% compared with the forecast of individual stations, which verifies the effectiveness of the proposed method. Different area division methods result in different effects on improving the prediction accuracy. This study provides a new method and reference for medium and long-term wind resource prediction.