A Compounding Model for Short-term Electrical Load Forecasting Based on Wavelet Transform

• 来源期刊: 电源世界
• 学科: 工学
• 关键词: forecasting WAVELET ANN AR
• 年，卷（期）: 2002,（7）
• 页码范围: 27-30
• 页数: 4页
• 分类号: TM715

摘要

In this paper the wavelet analysis theory is used in short-term electrical load forecasting.In order to concentrate the informationthat influence the load data,daily load data is decomposed by Daubechies wavelet filter.The front several wavelet components are usuallymuch bigger than others.The artificial neural networks(ANN)forecasting model in which consider the influence of climatic factor is built forthe front ones;For the other wavelet components the AR forecasting model is built.By this way the model provided in this paper is efficientand decreases the influence of shortcomings of every method used in this model.