Bearing faults identification and resonant band demodulation based on wavelet de-noising methods and envelope analysis

International Conference on Materials Technology and Energy

Abdelrhman, Ahmed M.^1 ; Sei Kien, Yong^1 ; Leong, M. Salman^2 ; Hee, Lim Meng^3 ; Ali Al-Obaidi, Salah M.^2

Curtin University Malaysia, CDT 250, Sarawak, Miri

98009, Malaysia^1

Institute of Noise and Vibration, University Technology Malaysia, Malaysia^2

UTM Razak School of Engineering and Advanced Technology, Malaysia^3

关键词: Bearing fault diagnosis;    Case Western Reserve University;    Continuous wavelet transforms;    Different resolutions;    High frequency components;    Low-and high-pass filters;    Multi variate analysis;    Time and frequencies;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/217/1/012031/pdf
DOI  :  10.1088/1757-899X/217/1/012031

来源: IOP

【 摘 要 】

The vibration signals produced by rotating machinery contain useful information for condition monitoring and fault diagnosis. Fault severities assessment is a challenging task. Wavelet Transform (WT) as a multivariate analysis tool is able to compromise between the time and frequency information in the signals and served as a de-noising method. The CWT scaling function gives different resolutions to the discretely signals such as very fine resolution at lower scale but coarser resolution at a higher scale. However, the computational cost increased as it needs to produce different signal resolutions. DWT has better low computation cost as the dilation function allowed the signals to be decomposed through a tree of low and high pass filters and no further analysing the high-frequency components. In this paper, a method for bearing faults identification is presented by combing Continuous Wavelet Transform (CWT) and Discrete Wavelet Transform (DWT) with envelope analysis for bearing fault diagnosis. The experimental data was sampled by Case Western Reserve University. The analysis result showed that the proposed method is effective in bearing faults detection, identify the exact fault's location and severity assessment especially for the inner race and outer race faults.