基于神经网络BP算法的局部电网短期负荷预测系统  

Design and Application of Short-term Load Forecasting Tool Based on BP Algorithm of Neural Network in Local Grid

作　　者：邹治锐 高坤 朱伟 姚境 唐伟斌 ZOU Zhirui;GAO Kun;ZHU Wei;YAO Jing;TANG Weibin(State Grid Changde Power Supply Company,Changde 415000,China)

机构地区：[1]国网湖南省电力有限公司常德供电公司,湖南常德415000

出　　处：《湖南电力》2020年第2期74-77,共4页Hunan Electric Power

摘　　要：电网负荷受天气、节假日、生活方式等多方面影响,短期呈现随机性,长期来看,又有一定的规律可循。选择合适的短期负荷预测模型,将有利于提高短期负荷预测的准确率,极大方便调度机构的短期负荷预测工作。本文基于神经网络BP算法的局部电网短期负荷预测,通过采集局部电网数据样本,获得大量数组,再使用神经网络BP算法进行自适应学习,获取各因素与负荷之间的非线性关系,预测局部电网负荷变化趋势,提前调控电网方式,降低局部电网主变、线路运行风险,确保电网安全运行。The load of power network is influenced by weather,holidays and life style,which is random in short term and regular in long term.Choosing a suitable short-term load forecasting model will help to improve the accuracy of short-term load forecasting and greatly facilitate the short-term load forecasting work.In this paper,short-term load forecasting tool based on BP algorithm of neural network is used to acquire a large number of arrays by collecting data samples from local power network.The non-linear relationship between each factor and load is obtained by the neural network BP algorithm self-adaptive learning.The changing trend of load in local power network is predicted,the mode of power grid is regulated in advance,the risks of main transformer and line operation are reduced,and the safe operation of power grid is ensured.

关 键 词：短期负荷预测 局部电网 神经网络 BP算法 MATLAB 

分 类 号：TM714[电气工程—电力系统及自动化]