基于GA-BP神经网络的列车轮对振动信号研究  

Research on Vibration Signal of Train Wheelset Based on GA-BP Neural Network

作　　者：云善起 薛鑫 牟茂源 刘媛媛 YUN Shanqi;XUE Xin;MOU Maoyuan;LIU Yuanyuan(Qingdao JARI Industrial Control Technology Co.,Ltd.,Qingdao 266520,China)

机构地区：[1]青岛杰瑞工控技术有限公司,山东青岛266520

出　　处：《机械制造与自动化》2023年第6期147-150,共4页Machine Building & Automation

摘　　要：轮对作为列车行驶的关键走行部件,对其故障的实时检测与精确诊断具有重要意义。通过小波包算法对轮对振动信号进行分解,提取轮对故障特征向量,以此作为输入参数,建立GA-BP诊断模型,采集振动信号样本对模型进行训练和测试。试验结果表明:GA-BP算法能够对轮对故障做出精确诊断。设计了轮对故障监测软件,实现了轮对信息监测的实时化与可视化。Wheelset as the key running part of a train,is of great significance for real-time fault detection and accurate diagnosis.Wavelet packet algorithm is used to decompose the wheel set vibration signal and extract the wheelset fault feature vector,with which as the input parameter,the GA-BP diagnosis model is established,and the vibration signal samples is collected to train and test the model.The test results show that the GA-BP algorithm can accurately diagnose the wheel set fault.The wheel set fault monitoring software is designed to realize the real-time and visualization of wheel set information.

关 键 词：列车轮对 振动监测 小波包算法 GA-BP算法 故障诊断 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置]