基于改进BP神经网络的轨道电路故障预测方法研究  

作　　者：李博文 

机构地区：[1]中国铁路武汉局集团有限公司武汉电务段,武汉430000

出　　处：《科技创新与应用》2024年第26期151-155,共5页Technology Innovation and Application

摘　　要：ZPW-2000A是铁路信号领域的重要组成部分,在保证列车安全运行过程中发挥着重要作用。为更好地预测轨道电路发生故障的概率,该文提出一种改进BP神经网络算法对轨道电路的故障进行预测,以蜻蜓算法对初始BP神经网络的权值和阈值进行优化改进,结合电务车间采集的轨道电压数据对改进后的BP神经网络进行训练,并对轨道电路的轨出1、轨出2电压值进行预测分析,得到轨道电路发生红光带的概率和趋势。同时将改进BP神经网络模型与现有预测模型进行对比分析,仿真结果表明,改进BP神经网络模型能够更为准确地预测轨道电路故障概率,提高设备的安全性和可靠性。ZPW-2000A is an important part of railway signal field and plays an important role in ensuring the safe operation of trains.In order to better predict the failure probability of the track circuit,this paper proposes an improved BP neural network algorithm to predict the fault of the track circuit.The dragonfly algorithm is used to optimize the weight and threshold of the initial BP neural network.Combined with the track voltage data collected in the power workshop,the improved BP neural network is trained,and the voltage values of the track circuits 1 and 2 are predicted and analyzed.The probability and trend of red band in track circuit are obtained.At the same time,the improved BP neural network model is compared with the existing prediction model,and the simulation results show that the improved BP neural network model can predict the track circuit fault probability more accurately and improve the safety and reliability of the equipment.

关 键 词：轨道电路 红光带 故障预测 蜻蜓算法 BP神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]