基于CKF-SVR的光伏电站发电功率预测方法研究

A Study on Power Prediction Method for Photovoltaic Power Station Based on CKF-SVR Power

由于光伏发电易受外界环境的影响,发电功率会产生剧烈波动,因此,发电功率的准确预测对于电网的稳定高效运行有着重要的实用价值。文章提出了基于容积卡尔曼滤波的支持向量机回归算法,用于分布式光伏电站的的发电功率预测,将SVR和融合后的CKF-SVR进行对比,结果后者对发电功率的预测精度有明显提高,能够为电站的优化调度提供指导。

Because photovoltaic power generation is vulnerable to the infl uence of external environment,the power generation will fl uctuate violently.Therefore,the accurate prediction of power generation has important practical value for the stable and effi cient operation of power grid.This paper proposes a support vector machine regression algorithm based on volumetric Kalman fi lter for power generation prediction of distributed photovoltaic power stations.The results show that the latter can improve the prediction accuracy of power generation and provide guidance for the optimal scheduling of power stations.