Converter‐level reliability prediction and analysis in switched reluctance motor drive. Issue 3 (5th February 2020)

• Record Type: Journal Article
• Authors: Xu, Shuai; Chen, Hao; Dong, Feng
• Is Part Of: IET electric power applications. Volume 14:Issue 3(2020)
• Page Start: 500
• Page End: 511
• Publication Date: 2020-02-05
• Subjects: fault tolerance -- failure analysis -- reliability -- reluctance motor drives -- bridge circuits -- power convertors -- thermal stresses -- Markov processes -- stress measurement; switched reluctance motor drive -- asymmetric half‐bridge power converter -- reliability estimation procedure -- electrothermal model -- failure rate model -- Markov model -- reliability analysis -- k‐out‐of‐n:G model -- converter‐level reliability prediction -- fault‐tolerance strategies -- thermal stress measurement; Electric power -- Periodicals; Electric power systems -- Periodicals; 621.305
• DOI: 10.1049/iet-epa.2019.0315
• Languages: English
• ISSNs: 1751-8660

Abstract

Abstract : Reliability has been a dominant factor in the performance of power converters. In this study, an enhanced k ‐out‐of‐ n :G model is presented to predict and analyse the reliability of typical asymmetric half‐bridge power converter in switched reluctance motor drive. First, the reliability estimation procedure is introduced including electrothermal model, failure rate model, and reliability model, during which the electrothermal model is critical to accelerate the process of reliability prediction. Next, the enhanced k ‐out‐of‐ n :G model can be built, and corresponding solving process is derived. Meanwhile, the reliability comparison is quantitatively illustrated between conventional k ‐out‐of‐ n :G models, generally accepted Markov model, and enhanced k ‐out‐of‐ n :G model, indicating the result with the proposed model is consistent with Markov model. Moreover, the modelling and solving processes are simpler. Then, the reliability is optimised in respect to turn‐on angle, turn‐off angle, and fault‐tolerance strategies with the enhanced k ‐out‐of‐ n :G model, providing a guideline to select the most effective control parameters and strategies for reliability improvement. Finally, the fault experiments and thermal stress measurement are carried out to indirectly validate the reliability analysis.