基于模糊集的电力系统神经网络负荷预报研究  

Load forecasting of power system using artificial neural network based on fuzzy set

作　　者：刘涤尘[1] 夏昌浩[2] 胡翔勇[2] 刘黎明[1] 

机构地区：[1]武汉大学电气工程学院,湖北武汉430072 [2]三峡大学电气信息学院,湖北宜昌443002

出　　处：《武汉大学学报（工学版）》2002年第4期68-71,共4页Engineering Journal of Wuhan University

摘　　要：提出了电力系统短期负荷预报基于模糊集的神经网络方法 .该方法计及了天气和日期特征量 ,具有训练时间短预测精度高的特点 .采用两种学习算法 ,依据模糊集概念用某地区电网实际数据建立样本集后 ,对ANN进行了训练 ,通过分析比较得出了优化模型 .A short_term load forecasting approach using artificial neural network based on fuzzy set is presented. The weather variables and date variables are considered; therefore the approach is better at respects of training time and forecasting accuracy. After training specimen set is established based on actual data of a region, two algorithms are used to train ANN in order to acquire more better results. The calculation results show that this approach is practical and effective.

关 键 词：模糊集 神经网络 短期负荷预报 电力系统 BP算法 

分 类 号：TM715[电气工程—电力系统及自动化]