Data-driven and deep learning-based detection and diagnosis of incipient faults with application to electrical traction systems

被引:38
|
作者
Chen, Hongtian [1 ]
Jiang, Bin [1 ]
Zhang, Tianyi [1 ]
Lu, Ningyun [1 ]
机构
[1] Nanjing Univ Aeronaut & Astronaut, Coll Automat Engn, Nanjing 211106, Peoples R China
基金
中国国家自然科学基金;
关键词
Incipient faults; Fault detection and diagnosis (FDD); Electrical drive systems; Canonical correlation analysis (CCA); Convolutional neural network CNN);
D O I
10.1016/j.neucom.2018.07.103
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Incipient faults in electrical drive systems will evolve into faults or failures as time goes on. Successful detection and diagnosis of incipient faults can not only improve the safety and reliability but also provide optimal maintenance instructions for electrical drive systems. In this paper, an integration strategy of data-driven and deep learning-based method is proposed to deal with incipient faults. The salient advantages of the proposed method can be summarized as: (1) The moving average technique is firstly introduced into the canonical correlation analysis (CCA) framework, which makes the new residual signals more sensitive to incipient faults than the traditional CCA-based method; (2) Based on the defined residual signals, the new test statistics cooperating closely with Kullback-Leibler divergence (KLD) are proposed from the probability viewpoint, which can greatly improve the fault detectability; (3) It is of high computational efficiency because the estimation of probability density functions of residual signals is skilly avoided; (4) Based on the new developed test statistics, the fault matrices are defined and regarded as the input of convolutional neural network (CNN) whose feature extraction ability is highly improved compared with the traditional method, which helps to accurately diagnose of incipient faults; (5) The proposed method can be implemented without any priori knowledge on system information. Theoretical analysis and three sets of experiments on a practical electrical drive system demonstrate the effectiveness of the proposed method. (C) 2019 Elsevier B.V. All rights reserved.
引用
收藏
页码:429 / 437
页数:9
相关论文
共 50 条
  • [1] Data-driven Detection and Diagnosis of Incipient Faults in Electrical Drives of High-Speed Trains
    Chen, Hongtian
    Jiang, Bin
    Chen, Wen
    Yi, Hui
    [J]. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2019, 66 (06) : 4716 - 4725
  • [2] Data-Driven Approaches for Diagnosis of Incipient Faults in DC Motors
    Munikoti, Sai
    Das, Laya
    Natarajan, Balasubramaniam
    Srinivasan, Babji
    [J]. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2019, 15 (09) : 5299 - 5308
  • [3] Data-Driven Approaches for Diagnosis of Incipient Faults in Cutting Arms of the Roadheader
    Liu, Qiang
    Liu, Songyong
    Dai, Qianjin
    Yu, Xiao
    Teng, Daoxiang
    Wei, Ming
    [J]. SHOCK AND VIBRATION, 2021, 2021
  • [4] Small infrared target detection by data-driven proposal and deep learning-based classification
    Ryu, Junhwan
    Kim, Sungho
    [J]. INFRARED TECHNOLOGY AND APPLICATIONS XLIV, 2018, 10624
  • [5] GrAb: A Deep Learning-Based Data-Driven Analytics Scheme for Energy Theft Detection
    Tanwar, Sudeep
    Kumari, Aparna
    Vekaria, Darshan
    Raboaca, Maria Simona
    Alqahtani, Fayez
    Tolba, Amr
    Neagu, Bogdan-Constantin
    Sharma, Ravi
    [J]. SENSORS, 2022, 22 (11)
  • [6] Reinforcement learning-based prediction approach for distributed Dynamic Data-Driven Application Systems
    Lin, Szu-Yin
    [J]. INFORMATION TECHNOLOGY & MANAGEMENT, 2015, 16 (04): : 313 - 326
  • [7] Reinforcement learning-based prediction approach for distributed Dynamic Data-Driven Application Systems
    Szu-Yin Lin
    [J]. Information Technology and Management, 2015, 16 : 313 - 326
  • [8] Data-driven fault diagnosis for PEMFC systems of hybrid tram based on deep learning
    Zhang, Xuexia
    Zhou, Jingzhe
    Chen, Weirong
    [J]. INTERNATIONAL JOURNAL OF HYDROGEN ENERGY, 2020, 45 (24) : 13483 - 13495
  • [9] Data-Driven Deep Learning for OTFS Detection
    Yi Gong
    Qingyu Li
    Fanke Meng
    Xinru Li
    Zhan Xu
    [J]. China Communications, 2023, 20 (01) : 88 - 101
  • [10] A Novel Data-Driven Fault Diagnosis Method Based on Deep Learning
    Zhang, Yuyan
    Gao, Liang
    Li, Xinyu
    Li, Peigen
    [J]. DATA MINING AND BIG DATA, DMBD 2017, 2017, 10387 : 442 - 452