基于小波-单神经元组合模型的电力系统负荷预测  

Power System Load Predicting Based on Wavelet and Single Neuron Combination Model

作　　者：曹兴霖[1] 邱道尹[1] 

机构地区：[1]华北水利水电学院,河南郑州450011

出　　处：《华北水利水电学院学报》2011年第4期77-80,共4页North China Institute of Water Conservancy and Hydroelectric Power

基　　金：河南省教育厅自然科学基金资助项目(2009A520016)

摘　　要：小波分析是一种新兴的数学工具,它能任意地提取负荷序列的细节.通过使用小波分析,可以在任何水平上分析负荷序列,它对信息成分采取逐渐精细的时域与频域处理,尤其对突发与短时的信息分析具有明显的优势.为了将小波分析用于负荷预测,提出了一种基于小波分解和单神经元的电力系统负荷预测方法.通过小波变换把负荷序列分解为不同频段的子序列,再对这些子序列采用单神经元模型进行预测,最后综合得到负荷预测的最终预测结果.并通过算例进行了验证.Wavelets analysis is a new mathematical tool which can abstract the details of the loading series at will. By using wavelet analysis,loading series can be analyzed at any level. It applies the gradual fine time-domain and frequency-domain processing to information elements,and especially has obvious advantages for sudden and short-time information analysis. This paper introduces the wavelets analysis into the load forecast,and a prediction method of power system load based on wavelet decomposition and single neuron is proposed. Through the wavelet transform,the load sequence is decomposed into different frequency subsequences, then the subsequences are forecasted respectively using appropriate single neural networks. Finally the ultimate forecasted results are integrated to get. Furthermore, the superiority of the arithmetic is investigated and the effectiveness of this approach is tested.

关 键 词：电力系统 小波分析 负荷预测 单神经元 

分 类 号：TM715[电气工程—电力系统及自动化]