Open AccessResearch Status and Trend of Fault Diagnosis Based on Deep Belief Network
Author(s) -
Bo Xiong,
Bo Tao,
Gongfa Li
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1302/2/022082
Subject(s) - deep belief network , deep learning , artificial intelligence , computer science , fault (geology) , field (mathematics) , feature (linguistics) , machine learning , complex network , feature extraction , scale (ratio) , data science , mathematics , linguistics , philosophy , physics , quantum mechanics , seismology , world wide web , pure mathematics , geology
Modern industrial systems are moving towards large-scale, complex, and high-speed. It is difficult to solve a series of technical problems that rely on traditional fault feature extraction methods. Since the concept of deep learning has been proposed, it has shown obvious advantages in many aspects. It includes feature extraction and pattern recognition. Therefore, many scholars have conducted deep learning to solve the problems of complex industrial system fault diagnosis. The deep belief network is the typical deep learning technologies in the field of control. This paper mainly introduces the deep belief network and describes its main ideas and methods. Finally, the paper summarizes the problems faced by the current deep belief network in the area of fault diagnosis and the future research direction.