Deep learning and its applications to machine health monitoring

A survey on Deep Learning based bearing fault diagnosis

A smart factory is a highly digitized and connected production facility that relies on smart manufacturing. Additionally, artificial intelligence is the core technology of smart factories. The use of machine learning and deep learning algorithms has produced fruitful results in many fields like image processing, speech recognition, fault detection, object detection, or medical sciences. With the increment in the use of smart machinery, the faults in the machinery equipment are expected to increase. Machinery fault detection and diagnosis through various deep learning algorithms has increased day by day. Many types of research have been done and published using both open-source and closed-source datasets, implementing the deep learning algorithms. Out of many publicly available datasets, Case Western Reserve University (CWRU) bearing dataset has been widely used to detect and diagnose machinery bearing fault and is accepted as a standard reference for validating the models. This paper summarizes the recent works which use the CWRU bearing dataset in machinery fault detection and diagnosis employing deep learning algorithms. We have reviewed the published works and presented the working algorithm, result, and other necessary details in this paper. This paper, we believe, can be of good help for future researchers to start their work on machinery fault detection and diagnosis using the CWRU dataset.

/pdf/bearing-fault-detection-and-diagnosis-using-case-western-1af0gxj849.pdf

Bearing Fault Detection and Diagnosis Using Case Western Reserve University Dataset With Deep Learning Approaches: A Review

Fault diagnosis of rotating machinery plays a significant role in the industrial production and engineering field. Owing to the drawbacks of traditional fault diagnosis methods, such as heavily dependence on human knowledge and professional experience, intelligent fault diagnosis based on deep learning (DL) has aroused the interest of researchers. DL achieves the desirable automatic feature learning and fault classification. Therefore, in this review, DL and DL-based intelligent fault diagnosis techniques are overviewed. DL-based fault diagnosis approaches for rotating machinery are summarized and discussed, primarily including bearing, gear/gearbox and pumps. Finally, with respect to modern intelligent fault diagnosis, the existing challenges and possible future research orientations are prospected and analyzed.

/pdf/deep-learning-based-intelligent-fault-diagnosis-methods-19edgqmyk6.pdf

Deep Learning-Based Intelligent Fault Diagnosis Methods Toward Rotating Machinery

Since 2006, deep learning (DL) has become a rapidly growing research direction, redefining state-of-the-art performances in a wide range of areas such as object recognition, image segmentation, speech recognition and machine translation. In modern manufacturing systems, data-driven machine health monitoring is gaining in popularity due to the widespread deployment of low-cost sensors and their connection to the Internet. Meanwhile, deep learning provides useful tools for processing and analyzing these big machinery data. The main purpose of this paper is to review and summarize the emerging research work of deep learning on machine health monitoring. After the brief introduction of deep learning techniques, the applications of deep learning in machine health monitoring systems are reviewed mainly from the following aspects: Auto-encoder (AE) and its variants, Restricted Boltzmann Machines and its variants including Deep Belief Network (DBN) and Deep Boltzmann Machines (DBM), Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). Finally, some new trends of DL-based machine health monitoring methods are discussed.

Deep Learning and Its Applications to Machine Health Monitoring: A Survey

In the rolling bearing fault diagnosis, the vibration signal of single sensor is usually nonstationary and noisy, which contains very little useful information, and impacts the accuracy of fault diagnosis. In order to solve the problem, this paper presents a novel fault diagnosis method using multivibration signals and deep belief network (DBN). By utilizing the DBN’s learning ability, the proposed method can adaptively fuse multifeature data and identify various bearing faults. Firstly, multiple vibration signals are acquainted from various fault bearings. Secondly, some time-domain characteristics are extracted from original signals of each individual sensor. Finally, the features data of all sensors are put into the DBN and generate an appropriate classifier to complete fault diagnosis. In order to demonstrate the effectiveness of multivibration signals, experiments are carried out on the individual sensor with the same conditions and procedure. At the same time, the method is compared with SVM, KNN, and BPNN methods. The results show that the DBN-based method is able to not only adaptively fuse multisensor data, but also obtain higher identification accuracy than other methods.

/pdf/bearing-fault-diagnosis-based-on-deep-belief-network-and-5116uw53d9.pdf

Bearing Fault Diagnosis Based on Deep Belief Network and Multisensor Information Fusion

This paper presents an approach to implement vibration signals for fault diagnosis of the rolling bearing. Due to the noise and transient impacts, it is difficulty to accurately diagnosis the faults with traditional methods. So a new type of learning architecture for deep generative model called deep belief networks (DBN) is applied. Since the unsupervised learning ability in DBN, it can extract the features from the raw data layer by layer. This article mainly research how to construct the encoder using DBN which can minimize the energy between the output and input vibration signals. Compared with existing diagnosis techniques, the proposed method can learn a good representation of features with higher accuracy. The results show that DBN can more comprehensively retain the data features in pattern recognition.

/pdf/fault-diagnosis-of-rolling-bearing-using-deep-belief-3s67zf774v.pdf

Fault Diagnosis of Rolling Bearing using Deep Belief Networks

Purpose




The aim of this paper is to solve the problem of low accuracy of traditional fatigue crack growth (FCG) prediction methods.




Design/methodology/approach




The GMSVR model was proposed by combining the grey modeling (GM) and the support vector regression (SVR). Meanwhile, the GMSVR model parameter optimal selection method based on the artificial bee colony (ABC) algorithm was presented. The FCG prediction of 7075 aluminum alloy under different conditions were taken as the study objects, and the performance of the genetic algorithm, the particle swarm optimization algorithm, the n-fold cross validation and the ABC algorithm were compared and analyzed.




Findings




The results show that the speed of the ABC algorithm is the fastest and the accuracy of the ABC algorithm is the highest too. The prediction performances of the GM (1, 1) model, the SVR model and the GMSVR model were compared, the results show that the GMSVR model has the best prediction ability, it can improve the FCG prediction accuracy of 7075 aluminum alloy greatly.




Originality/value




A new prediction model is proposed for FCG combined the non-equidistant grey model and the SVR model. Aiming at the problem of the model parameters are difficult to select, the GMSVR model parameter optimization method based on the ABC algorithm was presented. the results show that the GMSVR model has better prediction ability, which increase the FCG prediction accuracy of 7075 aluminum alloy greatly.

Fatigue crack growth prediction of 7075 aluminum alloy based on the GMSVR model optimized by the artificial bee colony algorithm

In this investigation, the high cycle fatigue life of 7075-T651 aluminum sheets under different stress levels were tested, and S-N curves were obtained by fitting the test data. For the problem that the traditional S-N curve fitting methods without considering the effect of fatigue limit on the S-N curve fitting accuracy, the traditional fitting methods were improved, the results show that the improved method can reduce the fitting error. The highcycle fatigue fracture morphology of 260MPa was observed by scanning electron microscopy (SEM). The observations reveal that the high cycle fatigue fracture morphology shows ductile fracture features. The main reason of the crack nucleation is that the second phase particles on the sub-surface, but the small dispersed precipitates can improve the fatigue performances. The morphology of the fatigue crack propagation is “stair” distribution. The fatigue crack transient beak region is characterized by the mixed-rupture characteristics of quasi-cleavage and dimples. Introduction With the advantages of small density, high strength, good machinability and heat treatment performances, the aluminum alloy of 7075 are widely used in the fields of aerospace industry, transportation, electronic communication [1-2]. These components suffered from the external alternating loads inevitably in service, eventually, the materials of components appear the irreversible fatigue failure, and either affect the normal operation of the equipment, or causes a serious disaster. Therefore, it is very significance to further investigate the fatigue performances and the fatigue mechanism of the 7075 aluminum alloy so that it can provide a guide for the structural design to reduce the fatigue accidents. In the past decade, many domestic and foreign scholars have done a lot of work on the fatigue performance investigation of aluminum alloy and have obtained a series of achievements. E. Cirik [3] investigated the effect of anodic oxidation on the fatigue performance of 7075-T6 alloy and fatigue tests with pre-corroded specimens showed that fatigue life of coated specimens was significantly affected by pre-corrosion, except for the specimen with the thickest coating layer. T. W. Zhao [4] tested the fatigue crack propagation rate of 7075-T651 aluminum alloy with the standard and non-standard compact specimens under the conditions of different stress ratio (R), overload, underload, high-low order, by improving the model of Wheeler [5], a good prediction results that the loading condition influence on the fatigue crack growth were obtained. R. S. Sachin [6] revealed the fretting fatigue behavior in 7075-T6 aluminum alloy use constant amplitude cyclic loading, using a Zeiss confocal microscope and a Hitachi Scanning Electron Microscope (SEM) to characterize the fretting fatigue damage, the results showed that fretting degradation depends on the microstructure of the material. J. Han [7] discovered the fatigue mechanism and fracture characteristics of the 7075-T651 aluminum alloy use the standard cylinder specimen, meanwhile, he obtained the fatigue life and S-N curve under different stress amplitudes. Despite comprehensive work, many questions remain to be answered, the fatigue mechanism of 7075-T651 aluminum alloy based on the low cycle fatigue (LCF) testing and use the cylinder International Conference on Material Science and Application (ICMSA 2015) © 2015. The authors Published by Atlantis Press 252 specimens usually, and there are less investigation of the fatigue performance of 7075-T651, especially, the high-cycle fatigue (HCF) performance should be studied more in-depth. In the current investigation, the fatigue performance often based on the medium thick sheet, but the fatigue performance of the aluminum alloy sheet was less studied. In this investigation, high-cycle fatigue experiments will be conducted using 7075-T651 aluminum alloy sheet under the conditions of different stress amplitude load, so that it can provide a fatigue performance guide for the application of 7075-T651 aluminum alloy.

/pdf/investigation-of-uniaxial-asymmetric-high-cycle-fatigue-23j77zz40n.pdf

Investigation of Uniaxial Asymmetric High-cycle Fatigue Failure Behavior of 7075-T651 Aluminum Alloy Sheet

Aiming at the difficulty of rolling bearing fault diagnosis in a strong noise environment, this paper proposes an enhanced integrated filter network. In the method, we firstly design an enhanced integrated filter, which includes the filter enhancement module and the expression enhancement module. The filter enhancement module can not only filter the high-frequency noise to extract useful features of medium and low-frequency signals but also maintain frequency and time resolution to some extent. On this basis, the expression enhancement module analyzes fault features intercepted by the upper network at multiple scales to get deep features. Then we introduce vector neurons to integrate scalar features into vector space, which mine the correlation between features. The feature vectors are transmitted by dynamic routing to establish the relationship between low-level capsules and high-level capsules. In order to verify the diagnostic performance of the model, CWRU and IMS bearing datasets are used for experimental verification. In the strong noise environment of SNR = −4 dB, the fault diagnosis precisions of the method on CWRU and IMS reach 94.85% and 92.45%, respectively. Compared with typical bearing fault diagnosis methods, the method has higher fault diagnosis precision and better generalization ability in a strong noise environment.

/pdf/a-rolling-bearing-fault-diagnosis-method-based-on-enhanced-2hk3nhtk.pdf

Jie Tao

Papers

Bearing Fault Diagnosis Based on Deep Belief Network and Multisensor Information Fusion

Fault Diagnosis of Rolling Bearing using Deep Belief Networks

Fatigue crack growth prediction of 7075 aluminum alloy based on the GMSVR model optimized by the artificial bee colony algorithm

Investigation of Uniaxial Asymmetric High-cycle Fatigue Failure Behavior of 7075-T651 Aluminum Alloy Sheet

A Rolling Bearing Fault Diagnosis Method Based on Enhanced Integrated Filter Network