基于加权改进D-S证据融合理论的地铁车辆转向架轴承故障诊断方法

Fault diagnosis method for bogie bearings of the metro vehicle based on improved weighted D-S evidence theory

地铁车辆转向架轴承的故障诊断方法大多采用振动、温度、声信号等单一物理量作为信息来源,存在智能诊断准确率低、可靠性差等问题.提出基于D-S证据融合理论的智能故障诊断方法.该方法以轴承运行中的振动加速度信号和温度信号为基本信息来源,采用加权改进D-S证据理论融合算法,诊断轴承健康状态,并在发生故障时判断故障类型.为了验证该诊断方法的可行性和有效性,以硬件装置为载体,基于轴承故障试验平台,对某型号地铁列车转向架故障轴承进行测试.试验结果表明:该轴承故障诊断方法能有效地诊断轴承的故障状态和故障类型.

The existing fault diagnosis methods for bogie bearings of the metro vehicle mostly use single physical quantity like vibration signal,temperature signal and acoustic signal as information source,thus having shortcomings like low accuracy and poor reliability of intelligent diagnosis.To improve this deficiency,an intelligent fault diagnosis method for bogie bearings of the metro vehicle based on improved weighted D-S evidence theory is proposed.This method,taking the radial vibration acceleration signal and temperature signal of bogie bearings as basic information source,uses the improved weighted D-S evidence algorithm to diagnose the health condition of bearing and find out the fault type when fault occurs.In order to verify the effectiveness and feasibility of the proposed method,a test platform for bearing fault diagnosis is designed and built,and the faulty bogie bearings used in a particular kind metro are tested on this platform.Test results show that the proposed fault diagnosis method for bogie bearings can availably identify the fault condition and fault type of bearing.