基于人工神经网络和模糊集的电力系统短期负荷预测方法  

A Method of Short-term Load Forecasting Based on Artificial Neural Network and Fuzzy Logic

作　　者：李杨[1] 李晓明[1] 黄玲[1] 陈岭[1] 舒欣[1] 

机构地区：[1]武汉大学电气工程学院,湖北武汉430072

出　　处：《华中电力》2007年第2期1-4,8,共5页Central China Electric Power

摘　　要：综合考虑到温度、日期类型和天气等因素对短期电力负荷的影响,提出了一种将人工神经网络(ANN)RBF模型和模糊逻辑相结合的短期负荷预测方法。该方法将电力负荷分为周期性的基本负荷和受多种因素影响的变动负荷两部分,对于周期负荷用ANN进行预测,采用负荷预测中比较精确的RBF算法;变动负荷采用模糊逻辑对天气因素、温度、日期类型分别做不同的模糊处理,然后利用模糊推理规则对基本负荷预测结果进行修正。通过典型算例与普通BP法预测结果相比较,结果表明该方法具有较高的预测精度。In order to consider the factors such as temperature,date type,weather status and etc which influence the short-term electric load forecasting, this paper provides a method based on radial basis function neural network combined with fuzzy logic. Short-term system load can be subdivided into periodic basic load and variable load influenced by number of factors. ANN is used to forecast the periodic load, here we adopt the R.BF model which is relative precise in load forecasting;, we use fuzzy logical to give different fuzzy deals to the weather factor temperature and date-types, then update the elementary result of load forecasting with the fuzzy reasoning rules. By Comparing between typical example and usual forecasting result by BP model, the result indicates that this method has high accuracy and some factual values.

关 键 词：负荷预报 人工神经网络 模糊集 电力系统运行 

分 类 号：TM715[电气工程—电力系统及自动化]