基于VMD和PSO-SVM的汽车传动轴系故障诊断  

Fault Diagnosis of Transmission Shaft System of Automobile based on VMD and PSO-SVM

作　　者：吴飞[1] 丁军 刘苏行 鲁雄 Wu Fei;Ding Jun;Liu Suhang;Lu Xiong(School of Mechanical and Electronic Engineering ,Wuhan University of Technology,Wuhan 430000,China)

机构地区：[1]武汉理工大学机电工程学院

出　　处：《机械传动》2019年第8期120-124,149,共6页Journal of Mechanical Transmission

基　　金：国家自然科学基金(51675393)

摘　　要：针对传动轴系振动信号故障特征难以提取的问题和进行故障诊断时难以获得大量故障样本的实际情况,提出了一种基于VMD和PSO-SVM相结合的传动轴系故障诊断方法。首先,将传动轴系振动信号进行VMD分解,得到本征模态函数IMF;然后,计算IMF的能量值和对应的能量熵值;最后,用粒子群优化(PSO)优化支持向量机(SVM)的参数,并将归一化处理后IMF的能量值及能量熵值作为特征向量,输入到PSO-SVM中来判断传动轴系的工作状态和故障类型。实验结果表明,该方法故障诊断准确率达到94. 44%,可以准确、有效地对传动轴系进行故障诊断。For the problem that it is difficult to extract fault features of vibration signals of transmissionshaft system and the actual situation that it is difficult to obtain a large number of fault samples in fault diagno-sis,a fault diagnosis method for transmission shafting based on variational mode decomposition(VMD)sand par-ticle swarm optimization support vector machine(PSO-SVM)is proposed. Firstly, the vibration signal of trans-mission shaft system is subjected to VMD decomposition, and intrinsic mode function IMF is obtained. Then, theenergy value of IMF and the corresponding energy entropy are calculated. Finally, Particle swarm optimization(PSO)is used to optimize the parameters of support vector machine(SVM), and the energy value and energy en-tropy of normalized IMF are input into the PSO-SVM to identify the working state and fault type of transmissionshaft system. The experimental results show that the accuracy of the method is 94.44%, and it can diagnose thefault of transmission shaft system accurately and effectively.

关 键 词：传动轴系 故障诊断 变分模态分解 能量熵 粒子群优化支持向量机 

分 类 号：U472[机械工程—车辆工程]