基于PSO-RBF的短期电力负荷预测模型

Short⁃term power load forecasting model based on PSO-RBF

精确的短期负荷预测对电网经济运行至关重要。为了提高电力系统负荷的预测精度,提出一种基于相似日和PSO优化RBF神经网络的短期负荷预测方法。该方法以RBF神经网络为短期电力负荷预测的基础算法,结合灰色关联分析法、K-means算法,通过DBI指数筛选出相似日集合,进一步结合PSO对RBF神经网络的参数进行优化。结果表明,所提方法预测精度优于单一RBF神经网络方法,在K-means算法选取的相似日基础之上,PSO-RBF预测模型的MAPE仅为0.77%,能精准预测待预测日负荷值,使其具有更广泛的应用价值。

Accurate short⁃term load forecasting is critical to the economical operation of the grid.In order to improve the forecasting accuracy of load for power system,a short⁃term load forecasting method based on similar days and PSO optimized RBF neural network was proposed.This method uses the RBF neural network as the basic algorithm for short⁃term power load forecasting,and combines the grey relational analysis method and the K-means algorithm.Selecting the similar day set through the DBI index and combining the PSO further optimizes the parameters of the RBF neural network.The results show that the prediction accuracy of the proposed method is better than that of the single RBF neural network method.On the basis of the similar days selected by the K-means algorithm,the MAPE of the PSO-RBF prediction model is only 0.77%,which can accurately predict the load value of the day to test,making it more applicable.