广东电力2025，Vol.38Issue(11)：67-74,8.DOI:10.3969/j.issn.1007-290X.2025.11.007
基于L-M神经网络算法的断路器合闸动作时间在线估测方法
Online Estimation Method of Circuit Breaker Closing Action Time Based on L-M Neural Network Algorithm
摘要
Abstract
This paper proposes a principle for estimating the closing action time of circuit breakers based on the auxiliary switch action time to address the problem of difficulty in accurately measuring the closing action time of circuit breakers during pre-breakdown.The analysis results of the operation data of circuit breakers on site show a strong correlation between the auxiliary switch action time and the closing action time of circuit breakers,and their relationship can be represented by a linear equation.The paper also proposes a method for predicting the action time of auxiliary switches based on L-M neural network algorithm.The model is trained using on-site operation data,and a prediction model between the action time of auxiliary switches and environmental temperature parameters is obtained.The simulation results show that the prediction accuracy error of the prediction model for auxiliary switches is within±0.43 ms,meeting the requirements of phase-selection closing.The proposed method for predicting the closing action time of circuit breakers can accurately estimate the closing action time online,providing theoretical and technical support for improving the accuracy of phase-selection closing.关键词
选相合闸/合闸动作时间预测/L-M神经网络/辅助开关/现场运行数据Key words
phase-selection closing/prediction of closing action time/L-M neural network/auxiliary switch/on-site operation data分类
信息技术与安全科学引用本文复制引用
戴龙成,倪辉,张一帆,黄智慧,董恩源,于家英,窦俊廷..基于L-M神经网络算法的断路器合闸动作时间在线估测方法[J].广东电力,2025,38(11):67-74,8.基金项目
国网宁夏电力有限公司科技项目(5229DK23000A) （5229DK23000A）
