A novel fusion diagnosis method for rotor system fault based on deep learning and multi-sourced heterogeneous monitoring data
Open accessThe DOI matches its Crossref record; 39 of the 39 references it cites were found in a registry or an index.
Abstract
Deep learning-based fault diagnosis has been acclaimed for its superiority in adaptively mining salient features. The monitoring data used as the input of deep learning typically includes only structured data (e.g. vibration signals, voltage signals, and acoustic emission signals), not unstructured data (e.g. infrared images), which provides another perspective on the mechanical health condition. To apply multi-sourced heterogeneous monitoring data fully, this paper presents a novel fusion diagnosis method integrating structured and unstructured data for rotor system faults. A novel multi-mode convolutional neural network (M-CNN) is first proposed to automatically learn fault-sensitive features from raw multisensory data composed of vibration signals and infrared images. M-CNN equipped with adjustable filter banks can identify data types and adaptively adopt the appropriate convolution mode. Then, t -distributed stochastic neighbor embedding (t-SNE) is introduced to fuse the deep features to further improve the quality of the learned features. Finally, the fused features are employed to conduct fault classification tasks. The effectiveness of the scheme is verified by fault diagnosis experiments in a rotor system, where it achieves a remarkable classification rate of 98.97%. Compared with similar methods, the proposed method exhibits outstanding performance, indicating the feasibility of using multi-sourced heterogeneous data for rotor system fault-diagnosis.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
(2008) · Journal of Machine Learning Research · Cited by 33,051
Greedy layer-wise training of deep networks
Bengio, Y., Lamblin, P., Popovici, D., & Larochelle, H. (2007) · Advances in Neural Information Processing Systems 19 · doi:10.7551/mitpress/7503.003.0024 · Cited by 4,841 · cited 1×
“The node number of the first hidden layer is set to be larger than the point number of input data, to ensure that more knowledge can be learned [38].”
BackgroundDeep learning and its applications to machine health monitoring
Zhao, R., Yan, R., Chen, Z., Mao, K., Wang, P., & Gao, R. X. (2019) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2018.05.050 · Cited by 2,894 · cited 2×
“Despite its well-known benefits, the application of deep learning to fault diagnosis faults is limited; presently, there is not a mature and systematic method to determine an appropriate model [8, 16].”
BackgroundMethodMultisensor data fusion: A review of the state-of-the-art
Khaleghi, B., Khamis, A., Karray, F. O., & Razavi, S. N. (2013) · Information Fusion · doi:10.1016/j.inffus.2011.08.001 · Cited by 2,074 · cited 1×
“Therefore, analyzing multivariate data together leads to vastly superior diagnostic results than monitoring a single aspect [19, 20].”
BackgroundDeep learning for smart manufacturing: Methods and applications
Wang, J., Ma, Y., Zhang, L., Gao, R. X., & Wu, D. (2018) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2018.01.003 · Cited by 1,740
Self-citationJia, F., Lei, Y., Lin, J., Zhou, X., & Lu, N. (2016) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2015.10.025 · Cited by 1,710 · cited 3×
“Accordingly, for bearings and gear-boxes (which are prominent components in a broad spectrum of machines), frequency spectra were entered into SAE [6], DBN [17], and CNN [18] for defect recognition.”
BackgroundMethodZhang, W., Li, C., Peng, G., Chen, Y., & Zhang, Z. (2018) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2017.06.022 · Cited by 1,421 · cited 2×
“Some studies [14, 23] have found that large filter kernels have advantages over small ones in capturing more fault information and suppressing high-frequency noise.”
BackgroundMethodInfrared thermography for condition monitoring – A review
Bagavathiappan, S., Lahiri, B. B., Saravanan, T., Philip, J., & Jayakumar, T. (2013) · Infrared Physics & Technology · doi:10.1016/j.infrared.2013.03.006 · Cited by 979
Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis
Guo, X., Chen, L., & Shen, C. (2016) · Measurement · doi:10.1016/j.measurement.2016.07.054 · Cited by 782 · cited 2×
“Moreover, complex hierarchical structures require a long training time to become a well-trained model [15, 36, 37], and this results in significantly increased computational costs and delays in real-time diagnosis of mechanical failures.”
BackgroundMethodLu, C., Wang, Z.-Y., Qin, W.-L., & Ma, J. (2017) · Signal Processing · doi:10.1016/j.sigpro.2016.07.028 · Cited by 775
A sparse auto-encoder-based deep neural network approach for induction motor faults classification
Sun, W., Shao, S., Zhao, R., Yan, R., Zhang, X., & Chen, X. (2016) · Measurement · doi:10.1016/j.measurement.2016.04.007 · Cited by 698 · cited 2×
“Previous research either sets them by empirical knowledge or determines them based on repeated experiments [10, 11, 13], which is particularly timeconsuming and may influence the performance of the diagnosis model.”
BackgroundMethodJing, L., Zhao, M., Li, P., & Xu, X. (2017) · Measurement · doi:10.1016/j.measurement.2017.07.017 · Cited by 698 · cited 1×
“Accordingly, for bearings and gear-boxes (which are prominent components in a broad spectrum of machines), frequency spectra were entered into SAE [6], DBN [17], and CNN [18] for defect recognition.”
MethodA novel deep autoencoder feature learning method for rotating machinery fault diagnosis
Shao, H., Jiang, H., Zhao, H., & Wang, F. (2017) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2017.03.034 · Cited by 684 · cited 3×
“Previous research either sets them by empirical knowledge or determines them based on repeated experiments [10, 11, 13], which is particularly time-consuming and may influence the performance of the diagnosis model.”
BackgroundMethodGan, M., Wang, C., & Zhu, C. (2016) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2015.11.014 · Cited by 550 · cited 1×
“Moreover, complex hierarchical structures require a long training time to become a well-trained model [15, 36, 37], and this results in significantly increased computational costs and delays in real-time diagnosis of mechanical failures.”
BackgroundRolling bearing fault diagnosis using an optimization deep belief network
Shao, H., Jiang, H., Zhang, X., & Niu, M. (2015) · Measurement Science and Technology · doi:10.1088/0957-0233/26/11/115002 · Cited by 517 · cited 1×
“By interpreting the external information, the state of mechanical internal components can be intelligently judged, which allows the use of condition-based maintenance (CBM) to improve reliability guarantees and avoid life-threatening catastrophes [2].”
BackgroundJia, F., Lei, Y., Guo, L., Lin, J., & Xing, S. (2018) · Neurocomputing · doi:10.1016/j.neucom.2017.07.032 · Cited by 517 · cited 1×
“Using raw vibration signals, SAE or its variants were demonstrated to integrate feature learning and fault diagnosis of rotary machinery [10, 11], planetary gearboxes [12], and induction motors [13] into one body; similar methods also applied to CNN for the diagnosis of rolling bearings [14, 15].”
MethodDing, X., & He, Q. (2017) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2017.2674738 · Cited by 489 · cited 1×
“Although some data reconstruction strategies [31 – 34] can increase data dimension, the cyclicity and continuity of vibration signals are destroyed to varying degrees.”
Gearbox fault diagnosis based on deep random forest fusion of acoustic and vibratory signals
Li, C., Sanchez, R.-V., Zurita, G., Cerrada, M., Cabrera, D., & Vásquez, R. E. (2016) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2016.02.007 · Cited by 470 · cited 1×
“This study [21] improved the diagnostic performance for gearboxes using acoustic emission and acceleration simultaneously.”
MethodVerstraete, D., Ferrada, A., Droguett, E. L., Meruane, V., & Modarres, M. (2017) · Shock and Vibration · doi:10.1155/2017/5067651 · Cited by 415 · cited 1×
“(a) Domain knowledge about signal processing and diagnostic expertise must be mastered to highlight fault characteristics in non-stationary and noisy signals; however, this kind of knowledge requires a long time to accumulate, and can introduce subjective factors [4, 5].”
BackgroundGearbox Fault Identification and Classification with Convolutional Neural Networks
Chen, Z., Li, C., & Sanchez, R.-V. (2015) · Shock and Vibration · doi:10.1155/2015/390134 · Cited by 391
Tran, V. T., AlThobiani, F., & Ball, A. (2014) · Expert Systems with Applications · doi:10.1016/j.eswa.2013.12.026 · Cited by 386 · cited 1×
“The combination of vibration, pressure, and current was adopted in a DBN-based diagnostic model of reciprocating compressor valves [22].”
MethodShao, H., Jiang, H., Zhang, H., Duan, W., Liang, T., & Wu, S. (2018) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2017.08.002 · Cited by 380
Jing, L., Wang, T., Zhao, M., & Wang, P. (2017) · Sensors · doi:10.3390/s17020414 · Cited by 379
Li, C., Sanchez, R.-V., Zurita, G., Cerrada, M., Cabrera, D., & Vásquez, R. E. (2015) · Neurocomputing · doi:10.1016/j.neucom.2015.06.008 · Cited by 295 · cited 1×
“Moreover, complex hierarchical structures require a long training time to become a well-trained model [15, 36, 37], and this results in significantly increased computational costs and delays in real-time diagnosis of mechanical failures.”
BackgroundAn adaptive deep convolutional neural network for rolling bearing fault diagnosis
Fuan, W., Hongkai, J., Haidong, S., Wenjing, D., & Shuaipeng, W. (2017) · Measurement Science and Technology · doi:10.1088/1361-6501/aa6e22 · Cited by 202 · cited 1×
“Advances in sensor technologies and machine learning algorithms have made data-driven-based intelligent fault diagnosis popular in machine maintenance frameworks [1].”
BackgroundIntelligent fault diagnosis of rotating machinery using infrared thermal image
Younus, A. M. D., & Yang, B.-S. (2012) · Expert Systems with Applications · doi:10.1016/j.eswa.2011.08.004 · Cited by 172
Virtualization and deep recognition for system fault classification
Wang, P., Ananya, Yan, R., & Gao, R. X. (2017) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2017.04.012 · Cited by 163
Thermal image based fault diagnosis for rotating machinery
Janssens, O., Schulz, R., Slavkovikj, V., Stockman, K., Loccufier, M., Rik Van De Walle, B., & Hoecke, S. V. (2015) · Infrared Physics & Technology · doi:10.1016/j.infrared.2015.09.004 · Cited by 149
Liu, J., Hu, Y., Wang, Y., Wu, B., Fan, J., & Hu, Z. (2018) · Measurement Science and Technology · doi:10.1088/1361-6501/aaaca6 · Cited by 129 · cited 1×
“(a) Domain knowledge about signal processing and diagnostic expertise must be mastered to highlight fault characteristics in non-stationary and noisy signals; however, this kind of knowledge requires a long time to accumulate, and can introduce subjective factors [4, 5].”
BackgroundTran, V. T., Yang, B., Gu, F., & Ball, A. D. (2013) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2013.02.001 · Cited by 127 · cited 1×
“Hence, various methods [26 – 29] were investigated to distinguish different health conditions using IR images, especially for faults that have no general solutions in vibration-based techniques [25].”
Serdio, F., Lughofer, E. D., Pichler, K., Buchegger, T., Pichler, M., & Efendic, H. (2014) · Information Fusion · doi:10.1016/j.inffus.2014.03.006 · Cited by 99 · cited 1×
“Therefore, analyzing multivariate data together leads to vastly superior diagnostic results than monitoring a single aspect [19, 20].”
BackgroundYang, Z.-X., Wang, X.-B., & Zhong, J.-H. (2016) · Energies · doi:10.3390/en9060379 · Cited by 84 · cited 1×
“Deep learning-based diagnosis methods have become popular, and free us from the burdensome signal processing and feature extraction that may empirically remove some ‘insignificant’ information that actually possesses undiscovered diagnostic value [9].”
BackgroundZhang, L., Gao, H., Wen, J., Li, S., & Liu, Q. (2017) · Microelectronics Reliability · doi:10.1016/j.microrel.2017.03.038 · Cited by 80 · cited 1×
“Accordingly, for bearings and gear-boxes (which are prominent components in a broad spectrum of machines), frequency spectra were entered into SAE [6], DBN [17], and CNN [18] for defect recognition.”
MethodSelf-citationString representations and distances in deep Convolutional Neural Networks for image classification
Barat, C., & Ducottet, C. (2016) · Pattern Recognition · doi:10.1016/j.patcog.2016.01.007 · Cited by 61
Segmented infrared image analysis for rotating machinery fault diagnosis
Duan, L., Yao, M., Wang, J., Bai, T., & Zhang, L. (2016) · Infrared Physics & Technology · doi:10.1016/j.infrared.2016.06.011 · Cited by 52
Self-citationNSCT-Based Infrared Image Enhancement Method for Rotating Machinery Fault Diagnosis
Bai, T., Zhang, L., Duan, L., & Wang, J. (2016) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2016.2579440 · Cited by 42
Self-citationYuan, H., Wang, X., Sun, X., & Ju, Z. (2017) · Measurement Science and Technology · doi:10.1088/1361-6501/aa6a07 · Cited by 27 · cited 1×
“Although the data-driven method directly transforms sensor data into insightful and actionable information without the requirement of an in-depth understanding of system physical behaviors, its performance relies strongly on the quality of the extracted features [3].”
MethodCalibration method for a vision guiding-based laser-tracking measurement system
Shao, M., Wei, Z., Hu, M., & Zhang, G. (2015) · Measurement Science and Technology · doi:10.1088/0957-0233/26/8/085009 · Cited by 14
Cited by
Applications of machine learning to machine fault diagnosis: A review and roadmap
Lei, Y., Yang, B., Jiang, X., Jia, F., Li, N., & Nandi, A. K. (2020) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2019.106587 · Cited by 2,997
A comprehensive review on convolutional neural network in machine fault diagnosis
Jiao, J., Zhao, M., Lin, J., & Liang, K. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.07.088 · Cited by 575
Fault Diagnosis Method Based on Principal Component Analysis and Broad Learning System
Zhao, H., Zheng, J., Xu, J., & Deng, W. (2019) · IEEE Access · doi:10.1109/access.2019.2929094 · Cited by 221
“[38] proposed a novel fusion diagnosis method based on multi-mode convolutional neural network and t-distributed stochastic neighbor embedding.”
MethodRole of artificial intelligence in rotor fault diagnosis: a comprehensive review
Nath, A. G., Udmale, S. S., & Singh, S. K. (2021) · Artificial Intelligence Review · doi:10.1007/s10462-020-09910-w · Cited by 172
Li, X., Cheng, J., Shao, H., Liu, K., & Cai, B. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2021.3125385 · Cited by 135
Metadata
- Authors
- Yuan, Z., Zhang, L., & Duan, L.
- Affiliations
- China University of Petroleum, Beijing (CN)
- Corresponding author
- Duan, Lixiang
- Venue
- Measurement Science and Technology
- Publisher
- IOP Publishing
- Published
- 2018-10-04
- Volume
- 29
- Issue
- 11
- Pages
- 115005
- ISSN
- 0957-0233 (ISSN-L)
- 1361-6501
- OpenAlex
- W2891841764
- Cited by
- 57
- References
- 39
- Citations by year
- 2026: 5 · 2025: 10 · 2024: 5 · 2023: 6 · 2022: 9 · 2021: 9 · 2020: 7 · 2019: 5 · 2018: 1
- FWCI
- 3.40
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- publishingsupport.iopscience.iop.org/iop-standard/v1 (vor)
- iopscience.iop.org/info/page/text-and-data-mining (tdm)
- Locations
- Measurement Science and Technology (publishedVersion)
- Funders
- National Key Research and Development Program of China (2017YFC0805803)
- National Natural Science Foundation of China (51674277)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Machine Fault Diagnosis Techniques
- Physical Sciences › Engineering › Control and Systems Engineering › Fault Detection and Control Systems
- Physical Sciences › Engineering › Civil and Structural Engineering › Structural Health Monitoring Techniques
- Keywords
- sensor fusion, vibration signals, infrared images, fault classification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.