Deep stacking l1-norm center configuration convex hull and its application in fault diagnosis of rolling bearing. (January 2020)

• Record Type: Journal Article
• Authors: Cheng, Zhengyang; Wang, Rongji
• Is Part Of: Mechanism and machine theory. Volume 143(2020)
• Publication Date: 2020-01
• Subjects: Deep stacking l1-norm center configuration convex hull -- Center configuration -- Rolling bearing -- Fault diagnosis; Machine theory -- Periodicals; Machinery -- Periodicals; Machines -- Périodiques; Génie mécanique -- Périodiques; Machine theory; Machinery; Periodicals; 621.81
• DOI: 10.1016/j.mechmachtheory.2019.103648
• Languages: English
• ISSNs: 0094-114X

Abstract

Abstract: Maximum margin classifier with flexible convex hulls (MMC-FCH) is an adaptive pattern recognition method based on convex hull vector and shrinkage factor, which can effectively identify different fault states. However, MMC-FCH is a shallow learning algorithm that cannot effectively diagnose complex signals. Meanwhile, MMC-FCH is essentially a binary classifier. For multi-classification, MMC-FCH can only perform multiple binary classifications. To overcome the shortcomings of MMC-FCH, we propose a deep stacking center configuration convex hull (DSC 3 H), which combines the convex hull with the idea of stacking-based representation learning (SRL). In DSC 3 H, the output of all previous modules combine with the original signal as the input of the next module to fully learn the information in the original signal. At the same time, the concept of center configuration is used to construct the multi-classification objective function of center configuration convex hull (C 3 H). However, redundant information and noise information may still exist in the proposed DSC 3 H method. Therefore, we further propose a deep stacking l 1 -norm center configuration convex hull (DS l 1 C 3 H) method, which makes the prediction model more robust and sparser under the constraint of l 1 -norm distance. The experiments of rolling bearings show that the proposed DS l 1 C 3 H method has better classification performance.