基于OOA优化BP神经网络的光伏发电功率预测

Power Prediction of Photovoltaic Power Generation Based on OOA Optimised BP Neural Network

准确预测光伏发电功率是保障电网平稳运行的关键环节,但由于光伏发电系统输入变量较多,造成标准BP神经网络系统对发电功率预测的精度不理想。基于OOA优化BP神经网络的预测模型(OOA-BP),可以提高预测精度。将总辐照度、直射辐射、散射辐射、气温、气压和湿度6种类型数据作为OOA-BP神经网络预测模型的输入量,并将其预测结果与标准BP神经网络进行对比。结果表明:相较于标准BP神经网络模型,OOA-BP神经网络预测模型通过优化阈值和权值,可以有效提升光伏发电功率的预测精度和预测效率。

Accurate prediction of photovoltaic(PV)power generation is a key link to ensure the smooth operation of the power grid,but due to the large number of input variables of the PV power generation system,the accuracy of the standard BP neural network system for power generation prediction is not ideal.The prediction model based on OOA optimised BP neural network(OOA-BP)can improve the prediction accuracy.Six types of data,namely,total irradiance,direct radiation,scattered radiation,air temperature,barometric pressure,and humidity,were used as inputs to the OOA-BP neural network prediction model and their prediction results were compared with the standard BP neural network.The results show that compared with the standard BP neural network model,the OOA-BP neural network prediction model can effectively improve the prediction accuracy and prediction efficiency of photovoltaic power generation by optimising the thresholds and weights.