基于模糊C-均值聚类分析与BP网络的短期负荷预测

The Short-term Load Forcast Using C-means Clustering and BP Network

提出了一种基于模糊Cˉ均值聚类分析与BP(Backˉpropagation)网络的短期负荷预测方法.通过模糊Cˉ均值聚类分析将历史负荷数据分成若干类,建立相应的BP网络模型,用LM(LevenbergˉM arquardt)优化法进行训练.找出与预测日相符的BP网络,预测一天中96点的负荷.实际负荷预测结果表明,该方法具有较好的训练速度和较高的预测精度.

This paper presents a short-term load forecasting method using fuzzy c-means clustering analysis and BP neural network. The historical load data are divided into several categories using fuzzy c-means clustering analysis, the corresponding BP neural network is built and then the Levenberg-Marquardt optimization to train the network is empleyed. The category coincident is found out with that of the daily load to be forecasted, and then the 96 points daily load is forecast with the corresponding BP network. The actual load forecasting results shows that the proposed method possesses faster training speed and greater forecasting accuracy.