Research Papers

Gearbox Fault Diagnosis Based on Selective Integrated Soft Competitive Learning Fuzzy Adaptive Resonance Theory

[+] Author and Article Information
Xiao-Jin Wan

School of Automotive Engineering,
Hubei Key Laboratory of Advanced Technology
for Automotive Components,
Wuhan 430070, China;
Hubei Collaborative Innovation Center for
Automotive Components Technology,
Wuhan University of Technology,
Wuhan 430070, China
e-mail: wxj_2001@163.com

Licheng Liu, Zhigang Xu

School of Automotive Engineering,
Hubei Key Laboratory of Advanced Technology
for Automotive Components,
Wuhan 430070, China;
Hubei Collaborative Innovation Center for
Automotive Components Technology,
Wuhan University of Technology,
Wuhan 430070, China

Zengbing Xu

College of Machinery and Automation,
Wuhan University of Science and Technology,
Wuhan 430070, China

1Corresponding author.

Contributed by the Computers and Information Division of ASME for publication in the JOURNAL OF COMPUTING AND INFORMATION SCIENCE IN ENGINEERING. Manuscript received June 20, 2018; final manuscript received October 11, 2018; published online November 19, 2018. Assoc. Editor: Conrad Tucker.

J. Comput. Inf. Sci. Eng 19(1), 011008 (Nov 19, 2018) (13 pages) Paper No: JCISE-18-1144; doi: 10.1115/1.4041776 History: Received June 20, 2018; Revised October 11, 2018

In this work, a soft competitive learning fuzzy adaptive resonance theory (SFART) diagnosis model based on multifeature domain selection for the single symptom domain and the single-target model is proposed. In order to solve the problem that the performance of traditional fuzzy ART (FART) is affected by the order of sample input, the similarity criterion of YU norm is introduced into the fuzzy ART network. In the meanwhile, the lateral inhibition theory is introduced to solve the wasteful problem of fuzzy ART mode node. By combining YU norm and lateral inhibition theory with fuzzy ART network, a soft competitive learning ART neural network diagnosis model that allows multiple mode nodes to learn simultaneously is designed. The feature parameters are extracted from the perspectives of time domain, frequency domain, time series model, wavelet analysis, and wavelet packet energy spectrum analysis, respectively. To further improve the diagnostic accuracy, the selective weighted majority voting method is integrated into the diagnosis model. Finally, the selected feature parameters are inputted to the integrated model to complete the fault classification and diagnosis. Finally, the proposed method is verified with a gearbox fault diagnosis test.

Copyright © 2019 by ASME
Your Session has timed out. Please sign back in to continue.


Leonhardt, S. , and Ayoubi, M. , 1997, “ Methods of Fault Diagnosis,” Control Eng. Pract., 5(5), pp. 683–692. [CrossRef]
Lu, C. , Wang, Z. , and Zhou, B. , 2017, “ Intelligent Fault Diagnosis of Rolling Bearing Using Hierarchical Convolutional Network Based Health State Classification,” Adv. Eng. Inf., 32, pp. 139–151. [CrossRef]
Narendra, K. G. , Sood, V. K. , Khorasani, K. , and Patel, R. , 1998, “ Application of a Radial Basis Function (RBF) Neural Network for Fault Diagnosis in a HVDC System,” IEEE Trans. Power Syst., 13(1), pp. 177–183. [CrossRef]
Li, Y. , Pont, M. J. , and Jones, N. B. , 2002, “ Improving the Performance of Radial Basis Function Classifiers in Condition Monitoring and Fault Diagnosis Applications Where Unknown Faults May Occur,” Pattern Recognit. Lett., 23(5), pp. 569–577. [CrossRef]
Wuxing, L. , Tse, P. W. , Guicai, Z. , and Tielin, S. , 2004, “ Classification of Gear Faults Using Cumulants and the Radial Basis Function Network,” Mech. Syst. Signal Process., 18(2), pp. 381–389. [CrossRef]
Scholkopf, B. , Sung, K. K. , Burges, C. J. C. , Girosi, F. , and Niyogi, P. , 1997, “ Comparing Support Vector Machines With Gaussian Kernels to Radial Basis Function Classifiers,” IEEE Trans. Signal Process., 45(11), pp. 2758–2765. [CrossRef]
Yang, H. T. , Liao, C. C. , and Chou, J. H. , 2001, “ Fuzzy Learning Vector Quantization Networks for Power Transformer Condition Assessment,” IEEE Trans. Dielectrics Electr. Insul., 8(1), pp. 143–149. [CrossRef]
Kayama, M. , Sugita, Y. , Morooka, Y. , and Fukuoka, S. , 1995, “ Distributed Diagnosis System Combining the Immune Network and Learning Vector Quantization,” IEEE 21st International Conference on Industrial Electronics, Control, and Instrumentation (IECON), Orlando, FL, Nov. 6–10, pp. 1531–1536.
Bassiuny, A. M. , Li, X. , and Du, R. , 2007, “ Fault Diagnosis of Stamping Process Based on Empirical Mode Decomposition and Learning Vector Quantization,” Int. J. Mach. Tools Manuf., 47(15), pp. 2298–2306. [CrossRef]
Widodo, A. , and Yang, B. S. , 2007, “ Support Vector Machine in Machine Condition Monitoring and Fault Diagnosis,” Mech. Syst. Signal Process., 21(6), pp. 2560–2574. [CrossRef]
Li, H. , and Zhang, Y. X. , 2009, “ An Algorithm of Soft Fault Diagnosis for Analog Circuit Based on the Optimized SVM by GA,” Ninth International Conference on Electronic Measurement and Instruments, Beijing, China, Aug. 16–19, pp. 4-1023–4-1027.
Cheng, J. , Yu, D. , Tang, J. , and Yang, Y. , 2013, “ Application of SVM and SVD Technique Based on EMD to the Fault Diagnosis of the Rotating Machinery,” Shock Vib., 16(1), pp. 89–98. [CrossRef]
Yuan, S. F. , and Chu, F. L. , 2006, “ Support Vector Machines-Based Fault Diagnosis for Turbo-Pump Rotor,” Mech. Syst. Signal Process., 20(4), pp. 939–952. [CrossRef]
Fei, S. , and Zhang, X. , 2009, “ Fault Diagnosis of Power Transformer Based on Support Vector Machine With Genetic Algorithm,” Expert Syst. Appl., 36(8), pp. 11352–11357. [CrossRef]
Sun, Y. , Zhang, S. , Miao, C. , and Li, J. , 2007, “ Improved BP Neural Network for Transformer Fault Diagnosis,” J. China Univ. Min. Technol., 17(1), pp. 138–142. [CrossRef]
Satish, B. , and Sarma, N. D. R. , 2005, “ A Fuzzy BP Approach for Diagnosis and Prognosis of Bearing Faults in Induction Motors,” IEEE Power Engineering Society General Meeting, San Francisco, CA, June 16, pp. 2291–2294.
Xiang, W. Q. , Zhang, H. , Wang, H. , and Xie, X. Z. , 2011, “ Application of BP Neural Network With LM Algorithm in Power Transformer Fault Diagnosis,” Power Syst. Prot. Control, 39(8), pp. 100–103.
Ma, D. , Liang, Y. , Zhao, X. , Guan, R. , and Shi, X. , 2013, “ Multi-BP Expert System for Fault Diagnosis of Power System,” Eng. Appl. Artif. Intell., 26(3), pp. 937–944. [CrossRef]
Wang, X. , and Liu, H. , 2018, “ Soft Sensor Based on Stacked Auto-Encoder Deep Neural Network for Air Preheater Rotor Deformation Prediction,” Adv. Eng. Inf., 36, pp. 112–119. [CrossRef]
Carpenter, G. A. , and Grossberg, S. , 1987, “ A Massively Parallel Architecture for a Self-Organizing Neural Pattern Recognition Machine,” Comput. Vision Graph. Image Process., 37(1), pp. 54–115. [CrossRef]
Carpenter, G. A. , and Grossberg, S. , 1988, “ ART 2: Self-Organization of Stable Category Recognition Codes for Analog Input Patterns,” Appl. Opt., 26(23), pp. 4919–4930. https://www.researchgate.net/publication/44649462_ART_2_Stable_Self-organization_of_Pattern_Recognition_Codes_for_Analog_Input_Patterns
Carpenter, G. A. , and Grossberg, S. , 1990, “ ART3: Hierarchical Search Using Chemical Transmitters in Self-Organizing Pattern Recognition Architectures,” Neural Networks, 3(2), pp. 129–152. [CrossRef]
Carpenter, G. A. , Grossberg, S. , and Reynolds, J. H. , 1991, “ ARTMAP: Supervised Real-Time Learning and Classification of Nonstationary Data by Self-Organizing Neural Network,” Neural Networks, 4(5), pp. 565–588. [CrossRef]
Carpenter, G. A. , Grossberg, S. , and Rosen, D. B. , 1991, “ Fuzzy ART: Fast Stable Learning and Categorization of Analog Patterns by an Adaptive Resonance System,” Neural Networks, 4(6), pp. 759–771. [CrossRef]
Carpenter, G. A. , Grossberg, S. , Markuzon, N. , Reynolds, J. H. , and Rosen, D. B. , 1992, “ Fuzzy ARTMAP: A Neural Network Architecture for Incremental Supervised Learning of Analog Multidimensional Maps,” IEEE Trans. Neural Networks, 3(5), pp. 698–713. [CrossRef]
Hartline, H. K. , 1974, Studies on Excitation and Inhibition in the Retina, Chapman and Hall, London.
Yandong, Y. , 1985, “ Triangular Norms and TNF-Sigma-Algebras,” Fuzzy Sets Syst., 16(3), pp. 251–264. [CrossRef]
Lowen, R. , 2012, Fuzzy Set Theory: Basic Concepts, Techniques and Bibliography, Springer Science and Business Media, Berlin.
Ripak, P. , 2018, “ Vibroacoustic Gear Signatures With Time-Frequency Spectrograms [EB/OL],” accessed Oct. 29, 2018, https://openei.org/datasets/dataset/gearbox-fault-diagnosis-data
Niu, G. , Han, T. , Yang, B.-S. , and Tan, A. C. C. , 2007, “ Multi-Agent Decision Fusion for Motor Fault Diagnosis,” Mech. Syst. Signal Process., 21(3), pp. 1285–1299. [CrossRef]
Dimitriadou, E. , Weingessel, A. , and Hornik, K. , 2001, “ Voting-Merging: An Ensemble Method for Clustering,” International Conference on Artificial Neural Networks, Vienna, Austria, Aug. 21–25, pp. 217–224.
Tang, Z. , and Yan, X. , 2007, “ Voting Algorithm of Fuzzy ARTMAP and Its Application to Fault Diagnosis,” International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), Haikou, China, Aug. 24–27, pp. 535–538.


Grahic Jump Location
Fig. 1

The fuzzy ART network

Grahic Jump Location
Fig. 2

The soft-ART network

Grahic Jump Location
Fig. 3

The topologies between comparison layer and lateral inhibition layer

Grahic Jump Location
Fig. 4

Fuzzy ART network diagnosis result

Grahic Jump Location
Fig. 5

SFART network diagnosis result

Grahic Jump Location
Fig. 6

Selective integration soft-ART diagnostic flowchart

Grahic Jump Location
Fig. 7

Association matrix

Grahic Jump Location
Fig. 8

Single model diagnosis result: (a) SFARTI diagnosis result, (b) SFARTII diagnosis result, (c) SFARTIII diagnosis result, (d) SFARTIV diagnosis result, and (e) SFARTV diagnosis result

Grahic Jump Location
Fig. 9

Integrated model diagnosis result: (a) integrated model I diagnosis result, (b) integrated model II diagnosis result, and (c) integrated model III diagnosis result

Grahic Jump Location
Fig. 10

Test site map: (a) mechanical fault simulation test bench, (b) eight-channel collector, (c) piezoelectric acceleration sensor, and (d) data collection system

Grahic Jump Location
Fig. 11

Defective gear: (a) worn gear, (b) cracked gear, (c) broken tooth gear, and (d)circumferential error gear

Grahic Jump Location
Fig. 12

Gear failure vibration signals

Grahic Jump Location
Fig. 13

Feature selection result: (a) time domain feature selection, (b) frequency domain feature selection, (c) AR model feature selection, (d) wavelet analysis feature selection, and (e) wavelet packet energy spectrum feature selection

Grahic Jump Location
Fig. 14

Single model diagnosis results: (a) SFARTI diagnostic results, (b) SFARTII diagnostic results, (c) SFARTIII diagnostic results, (d) SFARTIV diagnostic results, and (e) SFARTV diagnostic results

Grahic Jump Location
Fig. 15

Integrated model diagnostic results: (a) three models integrated, (b) four models integrated, and (c) five model integrated



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In