基于BP神经网络的家庭用电量预测与分析  

Forecast and Analysis of Household Electricity Consumption Based on BP Neural Network

作　　者：缪庆庆 林涛 张海静 韩小岗 樊相臣 MIAO Qingqing;LIN Tao;ZHANG Haijing;HAN Xiaogang;FAN Xiangchen(Jining Power Supply Co.,Ltd.,State Grid Shandong Electric Power Company,Jining 272100,China;State Grid Shandong Electric Power Co.,Ltd.,Jinan 250000,China;State Grid Shandong Comprehensive Energy Service Co.,Ltd.,Jinan 250000,China)

机构地区：[1]国网山东省电力公司济宁供电公司,山东济宁272100 [2]国网山东省电力公司,山东济南250000 [3]国网山东综合能源服务有限公司,山东济南250000

出　　处：《现代建筑电气》2020年第10期14-17,22,共5页Modern Architecture Electric

基　　金：国网山东省电力公司科技项目(5206061800AU)。

摘　　要：针对居民电力消费过高、电网存在峰值负荷等问题,提出一种基于BP神经网络的家庭用电量预测模型,利用MATLAB工具,通过对居民12个月用电量分析,不断调整各层神经元权值,降低误差信号,掌握居民用电规律,实现对家庭用电量的准确预测,并根据预测模型制定一套科学、合理的用电方案。试验结果表明:基于BP神经网络的家庭用电量预测准确率为90.8%,达到精准预测与分析的目的。Aiming at the problems of high power consumption of residents and peak load in power grid,a household electricity consumption prediction model based on BP neural network was proposed.Using MATLAB tools,the weights of neurons in each layer were adjusted continuously through the analysis of residents’electricity consumption in 12 months,which reduced the error signal,and grasp the rules of household electricity consumption.The accurate prediction of household electricity consumption was realized.A set of scientific and reasonable electricity consumption scheme based on the prediction model has been worked out.The experimental results show that the household electricity consumption prediction based on BP neural network has an accuracy rate of 90.8%,which achieves the purpose of accurate prediction and analysis.

关 键 词：BP神经网络 用电量预测 用电行为分析 数据预处理 

分 类 号：TU855[建筑科学]