A modified neighborhood preserving embedding-based incipient fault detection with applications to small-scale cyber-physical systems

被引：21

作者：

Chen, Hongtian ^{[1
]}

Wu, Jianping ^{[1
]}

Jiang, Bin ^{[1
]}

Chen, Wen ^{[2
]}

机构：

[1] Nanjing Univ Aeronaut & Astronaut, Coll Automat Engn, 169 Shengtai West Rd, Nanjing 211106, Peoples R China

[2] Wayne State Univ, Div Engn Technol, Detroit, MI 48202 USA

来源：

ISA TRANSACTIONS | 2020年 / 104卷

基金：

中国国家自然科学基金;

关键词：

Detection of incipient faults; Modified neighborhood preserving embedding; Industrial cyber-physical systems (ICPSs); Electrical drive systems; DRIVEN; DESIGN; ATTACK;

D O I：

10.1016/j.isatra.2019.08.022

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Industrial cyber-physical systems (ICPSs) are backbones of the Industrial 4.0 where control, physical entities, and monitoring are intensively interacted. Aiming to improve safety of a small-scale ICPS whose physical entity is an electrical drive system, this paper will develop a new detection strategy for incipient faults in neighborhood preserving embedding (NPE) framework that can provide stable solutions. The proposed modified NPE can not only extract local information effectively on data manifold of the ICPS but also solve the singularity problem caused by generalized eigenvalue decomposition skills. Additional advantages of this design for ICPSs include the enhanced fault detectability, inherent scalability, and accelerated computation efficiency. The proposed method is firstly evaluated by mathematical deviations and then is evaluated by its application to a small-scale ICPS. Three sets of experimental results show the efficacy of the proposed method in dealing with online detection of incipient faults in the ICPS. (C) 2019 ISA. Published by Elsevier Ltd. All rights reserved.

引用

页码：175 / 183

页数：9

共 31 条

[1] Probability-Relevant Incipient Fault Detection and Diagnosis Methodology With Applications to Electric Drive Systems [J].

Chen, Hongtian ;

Jiang, Bin ;

Ding, Steven X. ;

Lu, Ningyun ;

Chen, Wen .

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY, 2019, 27 (06) :2766-2773

[2] A Review of Fault Detection and Diagnosis for the Traction System in High-Speed Trains [J].

Chen, Hongtian ;

Jiang, Bin .

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2020, 21 (02) :450-465

[3] Deep PCA Based Real-Time Incipient Fault Detection and Diagnosis Methodology for Electrical Drive in High-Speed Trains [J].

Chen, Hongtian ;

Jiang, Bin ;

Lu, Ningyun ;

Mao, Zehui .

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, 2018, 67 (06) :4819-4830

[4] An improved incipient fault detection method based on Kullback-Leibler divergence [J].

Chen, Hongtian ;

Jiang, Bin ;

Lu, Ningyun .

ISA TRANSACTIONS, 2018, 79 :127-136

[5] Fault Detection for Non-Gaussian Processes Using Generalized Canonical Correlation Analysis and Randomized Algorithms [J].

Chen, Zhiwen ;

Ding, Steven X. ;

Peng, Tao ;

Yang, Chunhua ;

Gui, Weihua .

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2018, 65 (02) :1559-1567

[6]

Colombo AW, 2017, P IEEE, V104, P899

[7] Data-driven realizations of kernel and image representations and their application to fault detection and control system design [J].

Ding, Steven X. ;

Yang, Ying ;

Zhang, Yong ;

Li, Linlin .

AUTOMATICA, 2014, 50 (10) :2615-2623

[8]

Ding SX, 2014, ADV IND CONTROL, P1, DOI 10.1007/978-1-4471-6410-4

[9]

Gao Z., 2015, IEEE T IND ELECTRON, V62, P3768, DOI DOI 10.1109/TIE.2015.2419013

[10] A new fault diagnosis approach for analog circuits based on spectrum image and feature weighted kernel Fisher discriminant analysis [J].

He, Wei ;

He, Yigang ;

Zhang, Chaolong .

REVIEW OF SCIENTIFIC INSTRUMENTS, 2018, 89 (07)

← 1 2 3 4 →