Model Parameter Transfer for Gear Fault Diagnosis under Varying Working Conditions

Chao Chen; Fei Shen; Jiawen Xu; Ruqiang Yan

doi:10.1186/s10033-020-00520-9

摘要

Gear fault diagnosis technologies have received rapid development and been effectively implemented in many engineering applications. However, the various working conditions would degrade the diagnostic performance and make gear fault diagnosis (GFD) more and more challenging. In this paper, a novel model parameter transfer (NMPT) is proposed to boost the performance of GFD under varying working conditions. Based on the previous transfer strategy that controls empirical risk of source domain, this method further integrates the superiorities of multi-task learning with the idea of transfer learning (TL) to acquire transferable knowledge by minimizing the discrepancies of separating hyperplanes between one specific working condition (target domain) and another (source domain), and then transferring both commonality and specialty parameters over tasks to make use of source domain samples to assist target GFD task when sufficient labeled samples from target domain are unavailable. For NMPT implementation, insufficient target domain features and abundant source domain features with supervised information are fed into NMPT model to train a robust classifier for target GFD task. Related experiments prove that NMPT is expected to be a valuable technology to boost practical GFD performance under various working conditions. The proposed methods provides a transfer learning-based framework to handle the problem of insufficient training samples in target task caused by variable operation conditions.

关键词： Gear fault diagnosis; Model parameter transfe; Varying working conditions; Least square support vector machine

本文引用格式

Chao Chen , Fei Shen , Jiawen Xu , Ruqiang Yan . Model Parameter Transfer for Gear Fault Diagnosis under Varying Working Conditions[J]. Chinese Journal of Mechanical Engineering, 2021 , 34(1) : 13 -13 . DOI: 10.1186/s10033-020-00520-9

Abstract

Gear fault diagnosis technologies have received rapid development and been effectively implemented in many engineering applications. However, the various working conditions would degrade the diagnostic performance and make gear fault diagnosis (GFD) more and more challenging. In this paper, a novel model parameter transfer (NMPT) is proposed to boost the performance of GFD under varying working conditions. Based on the previous transfer strategy that controls empirical risk of source domain, this method further integrates the superiorities of multi-task learning with the idea of transfer learning (TL) to acquire transferable knowledge by minimizing the discrepancies of separating hyperplanes between one specific working condition (target domain) and another (source domain), and then transferring both commonality and specialty parameters over tasks to make use of source domain samples to assist target GFD task when sufficient labeled samples from target domain are unavailable. For NMPT implementation, insufficient target domain features and abundant source domain features with supervised information are fed into NMPT model to train a robust classifier for target GFD task. Related experiments prove that NMPT is expected to be a valuable technology to boost practical GFD performance under various working conditions. The proposed methods provides a transfer learning-based framework to handle the problem of insufficient training samples in target task caused by variable operation conditions.

Key words： Gear fault diagnosis; Model parameter transfe; Varying working conditions; Least square support vector machine

参考文献

[1] H Wang, Y Zhang, Z Yang. A reliability allocation method of CNC lathes based on copula failure correlation model. Chinese Journal of Mechanical Engineering, 2018, 31: 111.
[2] Y Li, Y Wang, Y He, et al. Modeling method for flexible energy behaviors in CNC machining systems. Chinese Journal of Mechanical Engineering, 2018, 31: 6.
[3] Y Xiong, Y Cheng, M Xu, et al. Reliability assessment of heavy-duty computer numerical control machine tools based on multi-performance multi-sequence hidden Markov model. Quality Engineering, 2020, 32(3): 409-420.
[4] T Jin, Z Yang, D Wang, et al. Reliability modeling for hydraulic components of heavy duty machine tools in distribution of degradation amount for oil contamination profile. China Mechanical Engineering, 2020, 31(13): 1613-1620, 1628. (in Chinese)
[5] F Zhang, S Han, J Liu, et al. Research of heavy NC machine reliability evaluation method based on Bayes theory. Modern Manufacturing Engineering, 2015, 8: 122-1251.
[6] H Huang, Z Liu, J Mi, et al. Reliability modeling and analysis of heavy-duty CNC machine tool spindle under hybrid uncertainty. Scientia Sinica, 2018, 1: 42-53.
[7] J Wang, Z Wang, Q Liu, et al. Reliability prediction of rotary working table for CXK5463 turning and milling machining center based on analogous argumentation method of similar products. Machine Tool and Hydraulics, 2016, 44(21): 164-167.
[8] M Catelani, L Ciani, M Venzi. Failure modes, mechanisms and effect analysis on temperature redundant sensor stage. Reliability Engineering and System Safety, 2018, 180: 425-433.
[9] K W Yun. Failure modes and risk assessment of rotary compressor under extraordinary operating conditions. 6th International Compressor Engineering Conference, West Lafayette, IN, July 9–12, 2000: 1382.
[10] I Latachi, T Rachidi, M Karim, et al. Reusable and reliable flight-control software for a fail-safe and cost-efficient cubesat mission. Design and Implementation, 2020, 7(146): 146.
[11] I B Brahim, S A Addouche, A E Mhamedi, et al. Build a Bayesian network from FMECA in the production of automotive parts: diagnosis and prediction. IFAC-PapersOnLine, 2019, 52(13): 2572-2577.
[12] J K Mohanty, P R Dash, P K Pradhan. FMECA analysis and condition monitoring of critical equipments in super thermal power plant. International Journal of System Assurance Engineering and Management, 2020, 11: 583-599.
[13] A Certa, F Hopps, R Inghilleri, et al. A Dempster-Shafer theory-based approach to the failure mode, effects and criticality analysis (FMECA) under epistemic uncertainty: Application to the propulsion system of a fishing vessel. Reliability Engineering and System Safety, 2017(159): 69-79.
[14] A E Brom, I N Omelchenko, O V Belova. Lifecycle costs for energy equipment FMECA for gas turbine. Procedia Engineering Oil and Gas Engineering. Procedia Engineering, 2016, 152: 177-181.
[15] Y Wu, H Cao, C Gao, et al. Calculation of optimal segment number of N-segment n-connection mode for overhead lines. Smart Power, 2019, 47(12): 98-102.
[16] I Mzougui, S Carpitella, A Certa, et al. Assessing supply chain risks in the automotive industry through a modified MCDM-based FMECA. Processes, 2020, 8(5): 579.
[17] B Suo, L Zhao, Y Yan. A novel Dempster-Shafer theory-based approach with weighted average for failure mode and effects analysis under uncertainty. Journal of Loss Prevention in the Process Industries, 2020, 65: 104145.
[18] D Piumatti, J Sini, S Borlo, et al. Multilevel simulation methodology for FMECA study applied to a complex cyber-physical system. Electronics, 2020, 9(10): 1736.
[19] N M Thoppil, V Vasu, C S P Rao. Failure mode identification and prioritization using FMECA: A study on computer numerical control lathe for predictive maintenance. Journal of Failure Analysis and Prevention, 2019, 19(4): 1153-1157.
[20] A Jomde, V Bhojwani, S Kedia, et al. Failure modes effects and criticality analysis of the linear compressor. Materials Today: Proceedings, 2017, 4(9): 10184-10188.
[21] B Goo, J Lee, S Seo, et al. Design of reliability critical system using axiomatic design with FMECA. International Journal of Naval Architecture and Ocean Engineering, 2017, 11(1): 11-21.
[22] M Ghali, M Tlija, N Aifaoui, et al. A CAD method for tolerance allocation considering manufacturing difficulty based on FMECA tool. International Journal of Advanced Manufacturing Technology, 2017, 91: 2435-2446.
[23] R Deodath, J Jhingoorie, C Riverol. Direct methanol fuel cell system reliability analysis. International Journal of Hydrogen Energy, 2017, 42(16): 12032-12045.
[24] U Okoro, A Kolios, L Cui. Multi-criteria risk assessment approach for components risk ranking - The case study of an offshore wave energy converter. International Journal of Marine Energy, 2016, 17: 21-39.
[25] A Zhang, L Cui, P Zhang. Research on the maintenance strategy of CNC machine tool. Proceedings of 20th International Conference on Industrial Engineering and Engineering Management, Baotou, China, 2013: 583-589.
[26] D J Lawson. Failure mode, effect and criticality analysis. Electronic Systems Effectiveness and Life Cycle Costing, 1983, 3: 55-74. MathSciNet
[27] H Wang, L Sun, L Shi. Failure mode, effect and criticality analysis of cnc grinder. Applied Mechanics and Materials, 2012, 141: 284-288.
[28] Z Huang. Evaluating intelligent residential communities using multi-strategic weighting method in china. Energy and Buildings, 2014, 69(1): 144-153.
[29] M Ohki, S Hayashi, M Ohkita. Fast computational algorithm of a fuzzy reasoning using the product-sum-gravity method. Systems and Computers in Japan, 2000, 31(3): 40-48.
[30] J Rezaei. Best-worst multi-criteria decision-making method. Omega, 2015, 53: 49-57.
[31] J D Martínez-Morales, E R Palacios-Hernández, G A Velázquez-Carrillo. Artificial neural network based on genetic algorithm for emissions prediction of a SI gasoline engine. Journal of Mechanical Science and Technology, 2014, 28(6): 2417-2427.
[32] U Gorka, J C Francisco, J Z Juan, et al. Preventing chatter vibrations in heavy-duty turning operations in large horizontal lathes. Journal of Sound and Vibration, 2015, 340: 317-330.

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献