摘要
根据电力系统中长期负荷的特点和径向基函数(RBF)神经网络的非线性辨识功能,将RBF神经网络应用于中长期负荷预测的数据预处理,具体讨论了空缺数据的补全以及失真数据的查找和修正,并提出了一种改进的基于RBF神经网络的中长期负荷预测模型。实际算例的分析表明,所提出的基于RBF神经网络的缺损数据处理方法和改进的中长期负荷预测模型是可行和有效的。
According to the load characteristics of electric power system in medium and long term and the nonlinear identification function of radial basis function(RBF) neural network, this paper applies RBF neural network to data pretreatment for medium and long-term load forecasting. The complement of the vacant data and the searching and correcting of the distortional data are discussed. An improved model based on RBF neural network for medium and long-term load forecasting is presented. The feasibility and validity of the method and model presented in this paper are proved by practical examples.
出处
《电力系统及其自动化学报》
CSCD
北大核心
2006年第1期15-19,共5页
Proceedings of the CSU-EPSA
基金
高等学校博士学科点专项科研基金资助项目(20020561004)
关键词
中长期负荷预测
数据预处理
人工神经网络
径向基函数
medium and long-term load forecasting
data pretreatment
artificial neural network (ANN)
radial basis function(RBF)