基于提升小波-BP神经网络的光伏阵列短期功率预测

摘要

The power of the PV array is a Non-stationary random process,influenced greatly by the radiation,temperature as well as uncertain external surroundings.Improving the accuracy of photovoltaic power system short-term prediction,especially the ultra-short-term forecast accuracy,has significant implications for the improving the operation and management efficiency of the photovoltaic power system.This article proposed a slip algorithm about the output of the DC side power,combining a lifting wavelet transform with BP neural network theory,to predict the ultra-shortterm power of the photovoltaic power array.The test results show that the method of ultra-short-term power forecast has good precision,and can be applied to complex weather conditions,such as the sunny,cloudy,rainy days.%提高光伏阵列的短期功率预测的精度,对光伏电站运营管理效率具有重要作用.文章提出了一种提升小波变换与BP神经网络相结合的直流侧功率输出预测滑移算法,对光伏阵列的超短期功率进行预测.实验结果表明,文章所提出的算法对超短期功率预测具有较高的精度,适用于晴天、多云、阴雨等复杂天气条件.