摘要
高预测精度的短期负荷预测对于坚强电网非常重要,根据电力负荷特性的变化规律,提出了一种改进的基于径向基函数神经网络的短期负荷预测方法,应用经GA优化的SVM多核径向基函数去提取有用数据,提高了基于RBF神经网络的短期负荷预测精度。以美国加州春季负荷为输入数据,应用MATLAB仿真说明改进算法的优越性和其鲁棒性。
High precision short-term load forecasting is very important to the strong power gird.In this paper,an improved Radial Basis Function（RBF） neural network short-term load forecasting method is proposed,which is according to variable law of power load characteristics.Using multi-core GA-optimized radial basis function SVM to extract useful data,it improves the accuracy of short-term load forecasting based RBF neural net work.The robustness and advantages of this improved forecasting strategy is demonstrated with Matlab simulating test by using data collected from the spring load in California,United States.
出处
《长春工程学院学报（自然科学版）》
2011年第2期21-25,共5页
Journal of Changchun Institute of Technology：Natural Sciences Edition
基金
吉林省教育厅科研项目(2009259)
关键词
RBF神经网络
短期负荷预测
支持向量机
遗传算法
RBF neural network
short-term load forecasting
support vector machine
genetic algorithm
