Criteria for optimizing kernel methods in fault monitoring process: A survey

被引：10
作者：
Bernal-de-Lazaro, Jose M. [1 ]
-Corona, Carlos Cruz [2 ]
Silva-Neto, Antonio J. [3 ]
Llanes-Santiago, Orestes [1 ]
机构：
[1] Univ Tecnol Habana Jose Antonio Echeverria, CUJAE, Dept Automat & Comp, Havana, Cuba
[2] Univ Granada, Dept Comp Sci & Artificial Intelligence, Granada, Spain
[3] Univ Estado Rio De Janeiro, IPRJ UERJ, Dept Mech Engn, Rio de Janeiro, RJ, Brazil
来源：
ISA TRANSACTIONS | 2022年 / 127卷
关键词：
Kernel methods; Kernel functions; Fault detection; Data preprocessing; Kernel parameter; Objective functions; INDEPENDENT COMPONENT ANALYSIS; SUPPORT VECTOR MACHINE; FISHER DISCRIMINANT-ANALYSIS; PARTICLE SWARM OPTIMIZATION; CANONICAL VARIATE ANALYSIS; FEATURE-EXTRACTION; NONLINEAR PROCESSES; DATA-DRIVEN; DIAGNOSIS; KPCA;
D O I：
10.1016/j.isatra.2021.08.040
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
Nowadays, how to select the kernel function and their parameters for ensuring high-performance indicators in fault diagnosis applications remains as two open research issues. This paper provides a comprehensive literature survey of kernel-preprocessing methods in condition monitoring tasks, with emphasis on the procedures for selecting their parameters. Accordingly, twenty kernel optimization criteria and sixteen kernel functions are analyzed. A kernel evaluation framework is further provided for helping in the selection and adjustment of kernel functions. The proposal is validated via a KPCA-based monitoring scheme and two well-known benchmark processes. (C) 2021 ISA. Published by Elsevier Ltd. All rights reserved.
引用
收藏
页码：259 / 272
页数：14
相关论文
共 131 条
[131]   Non-linear multivariate and multiscale monitoring and signal denoising strategy using Kernel Principal Component Analysis combined with Ensemble Empirical Mode Decomposition method [J].
Zvokelj, Matej ;
Zupan, Samo ;
Prebil, Ivan .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2011, 25 (07) :2631-2653
567891011121314