Novel Bootstrap-Based Discriminant NPE Integrated With Orthogonal LPP for Fault Diagnosis
- 25 January 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-9
Abstract
It has become a challenge to identify the discriminant information and the local geometric feature from the complex process data for improving fault diagnosis accuracy. Facing this challenge, this article proposes a novel fault diagnosis methodology based on discriminant neighborhood preserving embedding integrated with orthogonal locality preserving projections (DNPEOLPP). In DNPEOLPP, a novel local discriminant weight is constructed to identify the local geometric feature for dimensionality reduction (DR) so that the label of data is fully utilized. What is more, the cosine distance considering the spatial features of data is introduced into the process of neighbor graph construction to extract representative features. The bootstrap based DNPEOLPP (B-DNPEOLPP) is eventually developed to further address the issue of singular matrix caused by the small number of classes when applying DNPEOLPP for DR. Additionally, the Akaike information criterion (AIC) is utilized to identify the suitable order of DR, and AdaBoost.M2 is utilized to classify the type of fault. Finally, the Tennessee Eastman process (TEP) and the Grid-connected PV System (GPVS) are used to validate the performance of B-DNPEOLPP. Through case study analysis, it is demonstrated that B-DNPEOLPP can effectively identify different faults and obtain superior accuracy in fault diagnosis.Keywords
Funding Information
- National Natural Science Foundation of China (61973024, 61973022, 62073022)
- Fundamental Research Funds for the Central Universities (JD2225)
This publication has 30 references indexed in Scilit:
- Data-Driven Detection of Hot Spots in Photovoltaic Energy SystemsIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2019
- Statistical process monitoring based on nonlocal and multiple neighborhoods preserving embedding modelJournal of Process Control, 2018
- Multimode Process Monitoring and Fault Detection: A Sparse Modeling and Dictionary Learning MethodIEEE Transactions on Industrial Electronics, 2017
- Nonlinear Process Fault Diagnosis Based on Serial Principal Component AnalysisIEEE Transactions on Neural Networks and Learning Systems, 2016
- Semisupervised Kernel Learning for FDA Model and its Application for Fault Classification in Industrial ProcessesIEEE Transactions on Industrial Informatics, 2016
- Nonlocal structure constrained neighborhood preserving embedding model and its application for fault detectionChemometrics and Intelligent Laboratory Systems, 2015
- Generalized orthogonal locality preserving projections for nonlinear fault detection and diagnosisChemometrics and Intelligent Laboratory Systems, 2009
- Multiple-Fault Diagnosis of the Tennessee Eastman Process Based on System Decomposition and Dynamic PLSIndustrial & Engineering Chemistry Research, 2004
- Nonlinear Dimensionality Reduction by Locally Linear EmbeddingScience, 2000
- A Global Geometric Framework for Nonlinear Dimensionality ReductionScience, 2000