基于BP神经网络的电场测量误差补偿研究  

Research on Error Compensation of Electric Field Measurement Based on BP Neural Network

作　　者：刘靓雯 胡春璇 李国润 蒋琪 窦佳豪 郭俊玲 王建渊[3] LIU Jingwen;HU Chunxuan;LI Guorun;JIANG Qi;DOU Jiahao;GUO Junling;WANG Jianyuan(Department of Electric Power Engineering,Xi'an Electric Power College,Xi'an 710032,China;School of Electrical Engineering,Xi'an Jiaotong University,Xi'an 710049,China;School of Electrical Engineering,Xi'an University of Technology,Xi'an 710048,China)

机构地区：[1]西安电力高等专科学校电力工程系,陕西西安710032 [2]西安交通大学电气工程学院,陕西西安710049 [3]西安理工大学电气工程学院,陕西西安710048

出　　处：《电工技术》2024年第24期151-154,共4页Electric Engineering

基　　金：国家自然科学基金资助项目(编号51807160)。

摘　　要：为研究环境温湿度对输电线路工频电场测量值的影响,并降低电场测量值的误差,通过对测量系统等效电路进行分析,从环境温湿度对测量系统木支架含水率的影响入手,利用有限元仿真研究木支架在不同温湿度条件下对测量电场的影响。采用BP神经网络对不同温湿度环境下的测量电场误差样本进行训练,研究结果表明,通过BP神经网络对测量值的补偿可以有效减小木支架受温湿度影响产生的电场测量误差,从而为高温高湿环境下的紧急工频电场强度测试提供解决方案。In order to study the influence of ambient temperature and humidity on the power frequency electric field measurement of transmission lines,and reduce the error of electric field measurement value.Based on the analysis of the equivalent circuit of the measuring system,starting from the influence of ambient temperature and humidity on the moisture content of the measuring system,the finite element simulation is used to study the influence of the wood support on the measuring electric field under different temperature and humidity conditions.BP neural network was used to train the measurement field error samples in different temperature and humidity environments.The results show that compensation of measured value by BP neural network can effectively reduce the error caused by temperature and humidity of wood support,so as to provide a solution for emergency power frequency electric field strength testing under high temperature and high humidity environment.

关 键 词：电场强度 误差补偿 有限元 BP神经网络 

分 类 号：TM726[电气工程—电力系统及自动化]