A New Intelligent Fault Diagnosis Method and Its Application on Bearings

Fault diagnosis is vital in manufacturing system, however, fault diagnosis is divided into three stages: signal preprocessing, feature extraction and fault classification, which destroys the relationship between each stage and causes a part of the loss of fault information. The feature extraction process depends on the experimenter’s experience, and the recognition rate of the shallow diagnostic model does not achieve satisfactory results. In view of this problem, this paper proposes a method, the first step is converting raw signals into two-dimensional (2-D) images, the step can extract the features of the converted 2-D images and eliminate the impact of expert’s experience on the feature extraction process. Next, an intelligent diagnosis algorithm based on convolutional neural network (CNN) is proposed, which can automatically complete the feature extraction and fault identification of the signal. The effectiveness of the method is verified by using bearing data. Test with different sample sizes and noise signals to analyze their impact on diagnostic capabilities. Compared with other mainstream algorithms, this method has a higher recognition rate and can meet the timeliness of fault diagnosis.

[1]  Brigitte Chebel-Morello,et al.  Linear feature selection and classification using PNN and SFAM neural networks for a nearly online diagnosis of bearing naturally progressing degradations , 2015, Eng. Appl. Artif. Intell..

[2]  M. H. Mathias,et al.  Condition-based monitoring system for rolling element bearing using a generic multi-layer perceptron , 2015 .

[3]  Fanrang Kong,et al.  Fault diagnosis of rotating machinery based on the statistical parameters of wavelet packet paving and a generic support vector regressive classifier , 2013 .

[4]  S. Joe Qin,et al.  Process data analytics in the era of big data , 2014 .

[5]  Yaguo Lei,et al.  Deep Convolutional Transfer Learning Network: A New Method for Intelligent Fault Diagnosis of Machines With Unlabeled Data , 2019, IEEE Transactions on Industrial Electronics.

[6]  Liang Guo,et al.  Machinery health indicator construction based on convolutional neural networks considering trend burr , 2018, Neurocomputing.

[7]  Wenliao Du,et al.  Wavelet leaders multifractal features based fault diagnosis of rotating mechanism , 2014 .

[8]  K. Loparo,et al.  Bearing fault diagnosis based on wavelet transform and fuzzy inference , 2004 .

[9]  Long Zhang,et al.  Bearing fault diagnosis using multi-scale entropy and adaptive neuro-fuzzy inference , 2010, Expert Syst. Appl..

[10]  Yang Yu,et al.  A roller bearing fault diagnosis method based on EMD energy entropy and ANN , 2006 .

[11]  Diego Cabrera,et al.  Fault diagnosis in spur gears based on genetic algorithm and random forest , 2016 .

[12]  Krzysztof Marasek,et al.  Deep Belief Neural Networks and Bidirectional Long-Short Term Memory Hybrid for Speech Recognition , 2015 .

[13]  Feng Jia,et al.  An Intelligent Fault Diagnosis Method Using Unsupervised Feature Learning Towards Mechanical Big Data , 2016, IEEE Transactions on Industrial Electronics.