Research On Fault Diagnosis Method Based On Variational Mode Decomposition

With the development of science and technology, due to the integration degree of mechanical equipment is higher and higher, lead to close contact between components. If one part fails to work properly, the phenomenon of “pull one hair and the whole body is affected” will be happened. Then, the whole equipment is caused to be damaged, productiveness and benefit of factories is heavily impacted. Generally, fault diagnosis has three steps, acquiring the vibration signals, feature extraction, working condition recognition. The second step is so crucial that was deeply studied in this paper.The fundamental principle and algorithm of Variational Mode Decomposition(VMD) was studied, and the chaos particle swarm optimization algorithm was used to solve the problem that parameters of VMD were evaluated difficultly. Then, aimed at test functions, a comparison of the standard particle swarm optimization algorithm and the chaos particle swarm optimization algorithm were processed by numerical simulation. The search efficiency of the chaos particle swarm optimization was higher, which was verified.In this paper, a fault diagnosis method for rolling bearing based on chaos particle swarm optimization algorithm was proposed to optimize VMD parameter. Firstly, the chaos particle swarm optimization algorithm was took adventage of to confirm parameters of VMD. Secondly, rolling bearing fault signals were processed by VMD, the effective components were selected to reconstruct through the cross correlation coefficient, feature vectors of different signal types were extracted. Thirdly, feature vector sets of training samples were clustered by Kernel Fuzzy C-Means(KFCM), clustering centers of different signal types were obtained, the principle of the minimum Euclidean distance was adopted as the recognition method of feature vector sets of testing samples. In the same situation of diagnose problem and data, the clustering results and fault recognition rates were compared between VMD and Empirical Mode Decomposition(EMD), the superiority of VMD was verified.Machinery fault simulator magnum was the experiment setup, the signal acquisition system was build based on Lab VIEW, different types of rolling bearing fault signal were obtained. The parameters of VMD were optimized by the chaos particle swarm optimization algorithm, which consumed too much time. So fault diagnosis of rolling bearing based on VMD denoising was put forward, in this method, a simple and effective way of determining parameters of VMD was included. In the same situation of Diagnose problem and data, the clustering results and fault recognition rates were compared between VMD and EMD, the results showed that the effectiveness and timeliness of VMD was better than EMD.