Abstract
Dictionary-based fault diagnosis methods, focusing on storing feature patterns of known faults, have been widely used for electromechanical systems. The state of component degradation caused soft faults, however, are continuously changeable. Thus, conventional dictionaries cannot be applied for diagnosis of soft faults with multi-degradation levels. To address this issue, this article develops a new type of dictionary by combining the unit residual signal vector (URV) and the linear discriminant analysis (LDA) for feature transformation, which is referred to as URV-LDA dictionary. The unit residual signal vector keeps the fault feature growth trends but eliminates the degradation severity influence. The linear discriminant analysis is then implemented to find the best projection directions for classification. Specifically, two dictionaries named as the URV-MLDA binary-value dictionary and the URV-SLDA unique-value dictionary are proposed. To validate the efficiency of two developed dictionaries, an electromagnetic relay is carried out and two conventional methods are compared. The comparison results show the developed dictionaries can better solve the soft faults issues with significant increases on diagnostic accuracy.
Highlights
In modern industries, electromechanical systems have become more integrated and complex [1]
This article develops a new type of dictionary — the unit residual vector linear discriminant analysis (URV-LDA) dictionary, which is constructed with the unit residual signal vector (URV) and the linear discriminant analysis (LDA)
Compared with conventional linear or nonlinear transformation used for fault dictionaries, the developed URV-LDA transformation method first introduces the URV to perform the fault feature growth trends instead of the original feature distribution
Summary
Electromechanical systems have become more integrated and complex [1]. Inspired by the above studies, this article proposes a new type of dictionary— the unit residual vector linear discriminant analysis (URV-LDA) dictionary, to drawback the soft fault issue of conventional dictionary-based diagnosis methods. The multi-value dictionary is suitable for cases in which each feature under different fault states only exists several countable values or intervals It is widely used for analog circuit fault diagnosis, which is known as the ‘‘integercoded dictionary’’. DIAGNOSIS USING DICTIONARY WITH URV-LDA REPRESENTATION As mentioned in the Introduction, the soft faults and the feature intervals overlapping issues are two critical problems for dictionary-based fault diagnosis In this part, the developed new type of dictionary is described, which applies the linear discriminant analysis and the unit residual signal vector for dictionary construction to solve the overlapping issue and the soft fault issue, respectively
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
