Application of Feature Fusion Using Coaxial Vibration Signal for Diagnosis of Rolling Element Bearings

Jing Jiao,Di Pei,Zhunqing Hu,Jianhai Yue,Baoqing Li

doi:10.1155/2020/8831723

Abstract

The research of rolling element bearings (REBs) fault diagnosis based on single sensor vibration signal analysis is very common. However, the information provided by an individual sensor is very limited, and the robustness of the system is poor. In this paper, a novel fault diagnosis method based on coaxial vibration signal feature fusion (CVSFF) is proposed to fully analyze the multisensor information of the system and build a more reliable diagnostic system. An ensemble empirical mode decomposition (EEMD) method is used to decompose the original vibration signal into a number of intrinsic mode functions (IMFs). Then the autocorrelation analysis is introduced to reduce the random noise remaining in IMFs. After that, the Rényi entropy is calculated as the feature of bearings. Finally, the features of coaxial vibration signal are fused by a multiple-kernel learning support vector machine (MKL-SVM) to classify bearing conditions. In order to verify the effectiveness of the CVSFF method in REB diagnosis, eight data sets from the Case Western Reserve University Bearing Data Center are selected. The fault classification results demonstrate that the proposed approach is a valuable tool for bearing faults detection, and the fused feature from coaxial sensors improves fault classification accuracy for REBs.

Highlights

Information fusion is a technology that merges data to obtain more consistent, informative, and accurate information than the original raw data that are mostly uncertain [5]
Because a shaft and bearing inner race are rigidly connected, the shaft plays a role of vibration transmission between coaxial bearings. e fault signals of coaxial bearings are usually similar in a frequency domain. erefore, the coaxial vibration signal feature fusion (CVSFF) algorithm proposed in this paper takes vibration sensors of coaxial bearings as data sources
In order to verify the sensitivity of coaxial signal features, we did the same diagnosis based on single sensor

Summary

Methods

If the periodic signal contains random noise, the autocorrelation function can be used to reduce noise. Autocorrelation function is applied to noise reduction of IMFs, so as to retain the useful periodic signals in IMFs and reduce random white noise.

Experiments Analysis and Discussion

Findings

Method