A novel unsupervised learning method for intelligent fault diagnosis of rolling element bearings based on deep functional auto-encoder

2020 ◽  
Vol 34 (11) ◽  
pp. 4367-4381
Author(s):  
Anas H. Aljemely ◽  
Jianping Xuan ◽  
Farqad K. J. Jawad ◽  
Osama Al-Azzawi ◽  
Ali S. Alhumaima
Keyword(s):  
Fault Diagnosis ◽  
Unsupervised Learning ◽  
Rolling Element Bearings ◽  
Learning Method ◽  
Intelligent Fault Diagnosis ◽  
Rolling Element

A NEW GEARBOX FAULT DIAGNOSIS METHOD BASED ON LUCY–RICHARDSON DECONVOLUTION

2015 ◽  
Vol 39 (3) ◽  
pp. 593-603
Author(s):  
Xinghui Zhang ◽  
Jianshe Kang ◽  
Hongzhi Teng ◽  
Jianmin Zhao
Keyword(s):  
Fault Diagnosis ◽  
Experimental Studies ◽  
Experimental Tests ◽  
Rolling Element Bearings ◽  
Envelope Analysis ◽  
Rolling Element ◽  
Negative Effect ◽  
Diagnosis Method ◽  
Good Identification ◽  
Incipient Fault Diagnosis

Gear and bearing faults are the main causes of gearbox failure. Till now, incipient fault diagnosis of these two components has been a problem and needs further research. In this context, it is found that Lucy–Richardson deconvolution (LRD) proved to be an excellent tool to enhance fault diagnosis in rolling element bearings and gears. LRD’s good identification capabilities of fault frequencies are presented which outperform envelope analysis. This is very critical for early fault diagnosis. The case studies were carried out to evaluate the effectiveness of the proposed method. The results of simulated and experimental studies show that LRD is efficient in alleviating the negative effect of noise and transmission path. The results of simulation and experimental tests demonstrated outperformance of LRD compared to classical envelope analysis for fault diagnosis in rolling element bearings and gears, especially when it is applied to the processing of signals with strong background noise.

Sign in / Sign up

Export Citation Format

Share Document