基于QPSO_RBF的电力系统短期负荷预测

针对径向基函数(RBF)网络在电力系统短期负荷预测中存在的问题,提出一种量子粒子群优化(QPSO)算法训练RBF网络的方法,在确定网络隐含层节点个数后,将RBF网络各个参数编码成学习算法中的粒子个体进行优化,由此可在全局空间中搜索最优适应值的参数。用优化后的网络进行负荷预测,结果表明,该方法与传统的负荷预测方法相比,减少了训练时间并提高了预测精度,具有较好的应用前景。

Short-term electric power load forecasting based on QPSO_RBF

According to the problems of radial basis function(RBF) network in electric system short term load forecasting,this paper puts forward a method that quantum-behaved swarm optimization(PSO) algorithm train RBF neural network.After confirmed the number of nodes in hidden layer,all network parameters are coded to individual particles to optimize learning algorithm.Then,the parameter can search optimal-adaptive value in global space.Using the optimized network to load forecast result proves that this method not only reduces the training time but also improves the precision of prediction than traditional network algorithm.So it possesses best potential application in the field of short-term load forecasting.This project is supported by the National Natural Science Foundation of the Education Department of Henan Province(No.20047002).