基于混沌时间序列及神经网络的大同电网负荷短期预测研究  

Short-Term Load Forecasting Based on Chaotic Time Series and Artificial Neural Network for Datong Power Systems

作　　者：刘振宇[1,2] 冯华[1] 杨仁刚[2] 

机构地区：[1]山西农业大学信息科学与工程学院,山西太谷030801 [2]中国农业大学信息与电气工程学院,北京100083

出　　处：《中北大学学报（自然科学版）》2014年第6期695-698,共4页Journal of North University of China(Natural Science Edition)

基　　金：国家高技术研究发展计划(863计划)(2012AA050217)

摘　　要：精确的负荷预测对电力定价、实时负荷调整起着十分重要的作用.提出利用径向基函数神经网络模型,通过采用混沌时间序列的方法,对大同地区的全网用电负荷数据进行了短期预测验证.通过数据预处理产生混沌序列,进而通过相空间参数完成了相空间重构.最后通过RBF网络预测得到了相应的预测值,并与实际值进行了对比.结果表明:得到的预测结果与实际结果几乎重合,证明了该方法的高度预测能力与适应能力.Accurate load forecasting played a very important role in electricity forecast and real-time load ad- justment. The Radial basis function neural network model was proposed to make the prediction of short-term load for the whole network of Datong area by using the chaotic time series. The chaotic series were generated by data preprocessing, and then the parameters in the phase space completed the phase reconstruction, and fi- nally the corresponding predicted value was obtained by RBF neural network prediction. Through the compar- ison with actual value, the predicted results and the real results almost overlap, which shows that the method is highly of predictive ability and adaptive capacity.

关 键 词：负荷预测 混沌时间序列 RBF神经网络 电网 

分 类 号：TM715[电气工程—电力系统及自动化] TP18[自动化与计算机技术—控制理论与控制工程]