An incipient fault detection and self-learning identification method based on robust SVDD and RBM-PNN. (January 2020)

• Record Type: Journal Article
• Authors: Zhang, Chuanfang; Peng, Kaixiang; Dong, Jie
• Is Part Of: Journal of process control. Volume 85(2020)
• Page Start: 173
• Page End: 183
• Publication Date: 2020-01
• Subjects: Support vector data description -- Restricted Boltzmann Machine -- Probabilistic neural network -- Fault detection -- Fault identification; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2019.12.002
• Languages: English
• ISSNs: 0959-1524

Abstract

Highlights: Commonly used machine learning methods, including SVDD, RBM and PNN, are investigated. Faulty samples are adopted for robust SVDD modeling and the computation of sphere radius is modified, which is used for fault detection. Considering the superiorities in feature extraction and classification, RBM is combined with PNN for fault identification. Application in Tennessee Eastman process is carried out to show the theoretical results. Abstract: Incipient faults have low amplitudes and can be easily covered by system disturbances and noises. As timely incipient fault detection is the key to guarantee operation safety and suppress fault deterioration. In this paper, a novel incipient fault detection method based on robust support vector data description (RSVDD) is proposed. On the basis of traditional SVDD, both normal samples and faulty samples are introduced for RSVDD modeling and the computation of sphere radius is improved. Furthermore, a novel self-learning method based on restricted Boltzmann machine (RBM) and probabilistic neural network (PNN) is proposed for fault identification. The benefits of the proposed RSVDD and RBM-PNN scheme are illustrated by Tennessee Eastman benchmark.