考虑多重影响因素的负荷同时系数预测方法

Load Simultaneous Factor Forecast Considering Multiple Factors

负荷同时系数的选取是合理开展电力系统负荷预测工作的基础。目前,实际工程中的小区同时系数的选取缺乏理论依据,选取结果不能合理匹配小区的实际用电需求。针对该类问题,该文提出一种考虑多重影响因素的负荷同时系数预测方法。通过分析负荷同时系数的主要影响因素,构建同时系数影响因素指标体系。基于模糊K均值(K-means)聚类算法和径向基函数(radial basis function,RBF)神经网络相结合的方法,对样本集进行聚类分析,根据聚类结果预测样本的负荷同时系数。该方法可以大大提高样本的预测精度。对城市小区配电网规划具有指导意义。

Load simultaneous factor selection is important for power system load forecasting reasonably. Because of lacking theory basis, load simultaneous factor selection selecting cannot match actual electricity demand of urban community reasonably. This paper presents a method to forecast the load simultaneous factor which considers multiple effects. Through analyzing the main effect factors of simultaneous factor, index system of simultaneous factors' influence factors is constructed. Clustering analysis of the sample set is formed based on the method which combined fuzzy K-means clustering algorithm and radial basis function neural network, and the load simultaneous factors can be forecasted by the clustering results. The prediction accuracy can be improved by this method and sample, and the actual city distribution network can be guided.