Detection of bearing damage by statistic vibration analysis

E A Sikora

doi:10.1088/1757-899x/124/1/012167

Abstract

The condition of bearings, which are essential components in mechanisms, is crucial to safety. The analysis of the bearing vibration signal, which is always contaminated by certain types of noise, is a very important standard for mechanical condition diagnosis of the bearing and mechanical failure phenomenon. In this paper the method of rolling bearing fault detection by statistical analysis of vibration is proposed to filter out Gaussian noise contained in a raw vibration signal. The results of experiments show that the vibration signal can be significantly enhanced by application of the proposed method. Besides, the proposed method is used to analyse real acoustic signals of a bearing with inner race and outer race faults, respectively. The values of attributes are determined according to the degree of the fault. The results confirm that the periods between the transients, which represent bearing fault characteristics, can be successfully detected.

Highlights

Feature extraction of mechanical fault signals is vital for early fault diagnosis of mechanical equipment
It is shown that values of excess in selected bands are potentially a powerful tool for quantitative evaluation of the bearing condition
The main advantages of this parameter are insensitiveness to changes in speed and load bearing and an ability to indicate damage extent and the tendency to damage spread

Summary

Introduction

Feature extraction of mechanical fault signals is vital for early fault diagnosis of mechanical equipment. A rolling bearing is a key component in various electromechanical devices, and it plays an important role in the entire system. Regardless of size, will lead to a series of failures in the parts of the connection. The non-linear and non-stationary characteristics of the fault vibration signals and the interference of the random noise complicate the feature extraction process. Allowing total characteristic of the system state, the signal processing method is generally applied to fault diagnosis [1].

Published under licence by IOP Publishing Ltd

Findings

Conclusion