基于NWP相似性分析的超短期光伏发电功率预测

According to the fact that photovoltaic plants have similar generation power under similar weather conditions, an Ultra-short-term power forecasting method based on NWP similarity analysis is proposed. The proposed method uses the Pearson correlation coefficient to find weather forecast data similar to the predicted time, and estimates the power in the predicted time based on the actual power of the similar time. The proposed method can efficiently forecast the generated power based on the weather forecast data. Compared with the neural network, the proposed method has a better effect, especially in the period of large data fluctuations, which has higher reliability.