Submit to Applied Sciences Review for Applied Sciences Propose a Special Issue

Journal Browser

► Journal Browser

Deep Learning Based Machine Fault Diagnosis and Prognosis

Print Special Issue Flyer
Special Issue Editors
Special Issue Information
Keywords
Published Papers

A special issue of Applied Sciences (ISSN 2076-3417).

Deadline for manuscript submissions: closed (31 May 2017) | Viewed by 43679

Share This Special Issue

Special Issue Editor

Prof. Dr. David He

E-Mail Website
Guest Editor

1. School of Mechanical Engineering and Automation, Northeastern University, Shenyang 110004, China
2. Department of Mechanical and Industrial Engineering (MC 251), University of Illinois at Chicago, Chicago, IL 60661, USA
Interests: equipment health monitoring and fault diagnosis; prognostics and health management (PHM); failure analysis; reliability and quality engineering; manufacturing systems; signal processing; acoustic emission
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

In the age of Internet of Things and Industrial 4.0, massive real-time data are collected from health monitoring systems for the purpose of fault diagnosis and prognosis. The health monitoring big data are characterized by large volume and diversity. Effectively mining features from such data, accurately diagnosing the faults and predicting the remaining useful life (RUL) of the equipment in use with new advanced methods become new issues in the field of prognostics and health management (PHM). Traditional data driven methods are based on shallow learning architectures and require manually establishing explicit model equations and much prior knowledge about signal processing techniques and expertise, and therefore are limited in the age of big data. In recent years, deep learning methods are becoming a popular approach for big data process and analysis. Deep learning has the ability to yield useful and important features from data that can ultimately be useful for improving predictive power. Deep learning represents an attractive option to process big data for fault diagnosis and prognosis as deep learning has the ability to automatically select features that otherwise require much skill, time, and experience. This Special Issues call for papers that address developing effective and efficient deep learning based fault diagnosis and prognosis methods.

Prof. Dr. David He
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

fault diagnosis
prognosis
deep learning
big data
PHM

Published Papers (7 papers)

Download All Papers

Research

3112 KiB

Open AccessArticle

Planetary Gearbox Fault diagnosis via Joint Amplitude and Frequency Demodulation Analysis Based on Variational Mode Decomposition

by Zhipeng Feng, Dong Zhang and Ming J. Zuo

Appl. Sci. 2017, 7(8), 775; https://doi.org/10.3390/app7080775 - 30 Jul 2017

Cited by 34 | Viewed by 5848

Abstract

Planetary gearbox vibration signals have strong modulation features due to the amplitude modulation and frequency modulation (AM-FM) effect of gear faults, as well as the amplitude modulation (AM) effect of time-varying vibration transfer paths, on gear meshing vibrations. This results in an involute sidebands structure in Fourier spectrum, possibly misleading fault diagnosis. The modulating frequency of both amplitude modulation (AM) and frequency modulation (FM) parts is closely related to the gear fault characteristic frequency. This inspires the idea of joint amplitude and frequency demodulation analysis, thus addressing the complex sidebands issue inherent in Fourier spectrum. Demodulation analysis requires mono-component signals for accurate estimation of instantaneous frequency, and proper selection of an AM-FM component sensitive to gear fault. To this end, we firstly decompose the complex signal into intrinsic mode functions (IMFs) via variational mode decomposition (VMD), by exploiting its capability in decomposing complex modulated signal into constituent AM-FM components. For effective application of VMD in complex planetary gearbox signal analysis, we propose a method to determine a key parameter in VMD, i.e. the number of IMFs to be separated. For accurate instantaneous frequency estimation, we decompose IMFs via empirical AM-FM decomposition, to remove the influence of AM on instantaneous frequency estimation. Then, we select the sensitive IMF that contains the main gear fault information for further demodulation analysis. In order to properly select the sensitive IMF, we propose a criterion based on the gear vibration characteristics and the VMD properties. Finally, we obtain the amplitude and frequency demodulated spectra by applying Fourier transform to the amplitude envelope and instantaneous frequency of the selected sensitive IMF. According to the characteristics exhibited in the demodulated spectra, we can detect planetary gearbox fault. The proposed method is illustrated via a numerical simulated planetary gearbox vibration signal, and is further validated using lab experimental vibration signals of a planetary gearbox. Faults on all the three types of gear (sun, planet and ring) are successfully identified. Full article

(This article belongs to the Special Issue Deep Learning Based Machine Fault Diagnosis and Prognosis)

► Show Figures

Figure 1

2014 KiB

Open AccessArticle

Remaining Useful Life Prediction of Hybrid Ceramic Bearings Using an Integrated Deep Learning and Particle Filter Approach

by Jason Deutsch, Miao He and David He

Appl. Sci. 2017, 7(7), 649; https://doi.org/10.3390/app7070649 - 23 Jun 2017

Cited by 68 | Viewed by 7030

Abstract

Bearings are one of the most critical components in many industrial machines. Predicting remaining useful life (RUL) of bearings has been an important task for condition-based maintenance of industrial machines. One critical challenge for performing such tasks in the era of the Internet of Things and Industrial 4.0, is to automatically process massive amounts of data and accurately predict the RUL of bearings. This paper addresses the limitations of traditional data-driven prognostics, and presents a new method that integrates a deep belief network and a particle filter for RUL prediction of hybrid ceramic bearings. Real data collected from hybrid ceramic bearing run-to-failure tests were used to test and validate the integrated method. The performance of the integrated method was also compared with deep belief network and particle filter-based approaches. The validation and comparison results showed that RUL prediction performance using the integrated method was promising. Full article

(This article belongs to the Special Issue Deep Learning Based Machine Fault Diagnosis and Prognosis)

► Show Figures

Figure 1

2749 KiB

Open AccessArticle

Detection of Eccentricity Faults in Five-Phase Ferrite-PM Assisted Synchronous Reluctance Machines

by Carlos López-Torres, Jordi-Roger Riba, Antonio Garcia and Luís Romeral

Appl. Sci. 2017, 7(6), 565; https://doi.org/10.3390/app7060565 - 31 May 2017

Cited by 9 | Viewed by 5147

Abstract

Air gap eccentricity faults in five-phase ferrite-assisted synchronous reluctance motors (fPMa-SynRMs) tend to distort the magnetic flux in the air gap, which in turn affects the spectral content of both the stator currents and the ZSVC (zero-sequence voltage component). However, there is a lack of research dealing with the topic of fault diagnosis in multi-phase PMa-SynRMs, and in particular, those focused on detecting eccentricity faults. An analysis of the spectral components of the line currents and the ZSVC allows the development of fault diagnosis algorithms to detect eccentricity faults. The effect of the operating conditions is also analyzed, since this paper shows that it has a non-negligible impact on the effectivity and sensitivity of the diagnosis based on an analysis of the stator currents and the ZSVC. To this end, different operating conditions are analyzed. The paper also evaluates the influence of the operating conditions on the harmonic content of the line currents and the ZSVC, and determines the most suitable operating conditions to enhance the sensitivity of the analyzed methods. Finally, fault indicators employed to detect eccentricity faults, which are based on the spectral content of the stator currents and the ZSVC, are derived and their performance is assessed. The approach presented in this work may be useful for developing fault diagnosis strategies based on the acquisition and subsequent analysis and interpretation of the spectral content of the line currents and the ZSVC. Full article

(This article belongs to the Special Issue Deep Learning Based Machine Fault Diagnosis and Prognosis)

► Show Figures

Figure 1

4238 KiB

Open AccessFeature PaperArticle

Detection of Pitting in Gears Using a Deep Sparse Autoencoder

by Yongzhi Qu, Miao He, Jason Deutsch and David He

Appl. Sci. 2017, 7(5), 515; https://doi.org/10.3390/app7050515 - 16 May 2017

Cited by 48 | Viewed by 7799

Abstract

In this paper; a new method for gear pitting fault detection is presented. The presented method is developed based on a deep sparse autoencoder. The method integrates dictionary learning in sparse coding into a stacked autoencoder network. Sparse coding with dictionary learning is viewed as an adaptive feature extraction method for machinery fault diagnosis. An autoencoder is an unsupervised machine learning technique. A stacked autoencoder network with multiple hidden layers is considered to be a deep learning network. The presented method uses a stacked autoencoder network to perform the dictionary learning in sparse coding and extract features from raw vibration data automatically. These features are then used to perform gear pitting fault detection. The presented method is validated with vibration data collected from gear tests with pitting faults in a gearbox test rig and compared with an existing deep learning-based approach. Full article

(This article belongs to the Special Issue Deep Learning Based Machine Fault Diagnosis and Prognosis)

► Show Figures

Figure 1

3459 KiB

Open AccessArticle

Numerical Control Machine Tool Fault Diagnosis Using Hybrid Stationary Subspace Analysis and Least Squares Support Vector Machine with a Single Sensor

by Chen Gao, Wei Xue, Yan Ren and Yuqing Zhou

Appl. Sci. 2017, 7(4), 346; https://doi.org/10.3390/app7040346 - 31 Mar 2017

Cited by 22 | Viewed by 4664

Abstract

Tool fault diagnosis in numerical control (NC) machines plays a significant role in ensuring manufacturing quality. However, current methods of tool fault diagnosis lack accuracy. Therefore, in the present paper, a fault diagnosis method was proposed based on stationary subspace analysis (SSA) and least squares support vector machine (LS-SVM) using only a single sensor. First, SSA was used to extract stationary and non-stationary sources from multi-dimensional signals without the need for independency and without prior information of the source signals, after the dimensionality of the vibration signal observed by a single sensor was expanded by phase space reconstruction technique. Subsequently, 10 dimensionless parameters in the time-frequency domain for non-stationary sources were calculated to generate samples to train the LS-SVM. Finally, the measured vibration signals from tools of an unknown state and their non-stationary sources were separated by SSA to serve as test samples for the trained SVM. The experimental validation demonstrated that the proposed method has better diagnosis accuracy than three previous methods based on LS-SVM alone, Principal component analysis and LS-SVM or on SSA and Linear discriminant analysis. Full article

(This article belongs to the Special Issue Deep Learning Based Machine Fault Diagnosis and Prognosis)

► Show Figures

Graphical abstract

5861 KiB

Open AccessArticle

Application of the DC Offset Cancellation Method and S Transform to Gearbox Fault Diagnosis

by Xinghui Zhang, Jianmin Zhao, Rusmir Bajrić and Liangliang Wang

Appl. Sci. 2017, 7(2), 207; https://doi.org/10.3390/app7020207 - 20 Feb 2017

Cited by 11 | Viewed by 5456

Abstract

In this paper, the direct current (DC) offset cancellation and S transform-based diagnosis method is verified using three case studies. For DC offset cancellation, correlated kurtosis (CK) is used instead of the cross-correlation coefficient in order to determine the optimal iteration number. Compared to the cross-correlation coefficient, CK enhances the DC offset cancellation ability enormously because of its excellent periodic impulse signal detection ability. Here, it has been proven experimentally that it can effectively diagnose the implanted bearing fault. However, the proposed method is less effective in the case of simultaneously present bearing and gear faults, especially for extremely weak bearing faults. In this circumstance, the iteration number of DC offset cancellation is determined directly by the high-speed shaft gear mesh frequency order. For the planetary gearbox, the application of the proposed method differs from the fixed-axis gearbox, because of its complex structure. For those small fault frequency parts, such as planet gear and ring gear, the DC offset cancellation’s ability is less effective than for the fixed-axis gearbox. In these studies, the S transform is used to display the time-frequency characteristics of the DC offset cancellation processed results; the performances are evaluated, and the discussions are given. The fault information can be more easily observed in the time-frequency contour than the frequency domain. Full article

(This article belongs to the Special Issue Deep Learning Based Machine Fault Diagnosis and Prognosis)

► Show Figures

Figure 1

3207 KiB

Open AccessArticle

Deep Fault Recognizer: An Integrated Model to Denoise and Extract Features for Fault Diagnosis in Rotating Machinery

by Xiaojie Guo, Changqing Shen and Liang Chen

Appl. Sci. 2017, 7(1), 41; https://doi.org/10.3390/app7010041 - 30 Dec 2016

Cited by 72 | Viewed by 6585

Abstract

Fault diagnosis in rotating machinery is significant to avoid serious accidents; thus, an accurate and timely diagnosis method is necessary. With the breakthrough in deep learning algorithm, some intelligent methods, such as deep belief network (DBN) and deep convolution neural network (DCNN), have been developed with satisfactory performances to conduct machinery fault diagnosis. However, only a few of these methods consider properly dealing with noises that exist in practical situations and the denoising methods are in need of extensive professional experiences. Accordingly, rethinking the fault diagnosis method based on deep architectures is essential. Hence, this study proposes an automatic denoising and feature extraction method that inherently considers spatial and temporal correlations. In this study, an integrated deep fault recognizer model based on the stacked denoising autoencoder (SDAE) is applied to both denoise random noises in the raw signals and represent fault features in fault pattern diagnosis for both bearing rolling fault and gearbox fault, and trained in a greedy layer-wise fashion. Finally, the experimental validation demonstrates that the proposed method has better diagnosis accuracy than DBN, particularly in the existing situation of noises with superiority of approximately 7% in fault diagnosis accuracy. Full article

(This article belongs to the Special Issue Deep Learning Based Machine Fault Diagnosis and Prognosis)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Deep Learning Based Machine Fault Diagnosis and Prognosis

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Published Papers (7 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI