摘要
电力系统负荷预测是电力生产部门的重要工作之一 ,作者利用 BP神经网络进行电力系统短期负荷预测 ,在保证有足够的训练样本的前提下 ,对预测模型进行合理分类 ,构造了相应于不同季节的周预测、日预测模型 ,并对输入变量的选择 ,特别是温度的选取问题 ,进行了讨论。在神经网络训练的过程中 ,往往会出现过拟合的现象 ,给预测的结果带来不利的影响 ,为此在训练过程中 ,将样本随机地分离为训练集和测试集来防止这个问题。典型算例的计算表明 ,该方法是有效的。
Load forecasting is an important task in the production of electric energy. In this paper, BP artificial neural network is applied in short term load forecasting.Under the condition of possessing enough training samples the models for forecasting are reasonably classified,the weekly and daily load forecasting models for different seasons are constructed. The selection of input variables, especially the selection of temperature, is discussed. In the training of neural network the over fitting often appears which affects the result of forecasting. To prevent this problem the entire data set is divided into training set and validation set. The result of typical calculation examples shows that the presented method is effective.
出处
《电网技术》
EI
CSCD
北大核心
2002年第2期10-13,18,共5页
Power System Technology
基金
国家杰出青年科学基金资助项目 ( 6 970 0 2 5 )
国家自然科学基金资助项目 ( 5 993715 0 )~~
关键词
电力系统
短期负荷预测
神经网络
预测模型
short term load forecasting
BP artificial neural network
correlation analysis
over fitting
electricity market
