基于RBF神经网络模型的电力系统短期负荷预测

Short Term Load Forecasting of Power System Based on RBF Neural Network Model

电力系统短期负荷对电力企业的经济效益和社会效益都有一定影响。因此文中建立了基于RBF神经网络的电力系统短期负荷模型。用历史负荷数据作为训练样本,用训练好的神经网络进行电力系统短期负荷预测,并与BP神经网络进行对比。RBF神经网络的平均误差为2.09%,最大误差为4.77%,相比于BP神经网络精确度较高,有利于电力系统合理地进行调度规划工作。

The short-term load of power system has a certain impact on the economic and social benefits of electric power enterprises.Therefore,a short-term load model of power system based on RBF neural network was established in this paper.Using historical load data as training samples,short-term load forecasting of power system was carried out by trained neural network,and compared with BP neural network.The average error of RBF neural network is 2.09%,and the maximum error is 4.77%.Compared with BP neural network,the accuracy of RBF neural network is higher,which is conducive to the rational dispatch planning of power system.