Real‐time open‐switch fault diagnosis in automotive permanent magnet synchronous motor drives based on Kalman filter. Issue 12 (1st September 2020)

• Record Type: Journal Article
• Authors: Naseri, Farshid; Schaltz, Erik; Lu, Kaiyuan; Farjah, Ebrahim
• Is Part Of: IET power electronics. Volume 13:Issue 12(2020)
• Page Start: 2450
• Page End: 2460
• Publication Date: 2020-09-01
• Subjects: synchronous motors -- fault diagnosis -- permanent magnet motors -- control engineering computing -- machine control -- synchronous motor drives -- stators -- Kalman filters -- closed loop systems; three‐phase motor currents -- Kalman filter -- estimated stator currents -- normalised residual signals -- OSFs -- faulty switches -- current sensor faults -- KF‐based method -- fast detection time -- load variations -- open‐loop PMSM drives -- drive control unit -- hardware‐in‐the‐loop experiments -- real‐time feasibility -- open‐switch fault diagnosis -- automotive permanent magnet synchronous motor drives -- effective fault diagnosis method -- single switch faults -- multiple open‐switch faults -- three‐phase permanent magnet synchronous motor drives -- real‐time estimation -- power 1.5 kW; Power electronics -- Periodicals; 621.31705
• DOI: 10.1049/iet-pel.2019.1498
• Languages: English
• ISSNs: 1755-4535

Abstract

Abstract : This study proposes an effective diagnosis method for single and multiple open‐switch faults (OSFs) in three‐phase permanent magnet synchronous motor (PMSM) drives. The proposed method is based on the real‐time estimation of the three‐phase motor currents using three Kalman filters (KFs). Three residual signals are defined as the differences between the three‐phase measured and estimated stator currents. The averaged normalised residual signals are used as diagnostic criteria for the detection of OSFs. The proposed method accurately detects the OSFs, localises the faulty switches and effectively discriminates between the OSFs and current sensor faults. The main superiorities of the proposed KF‐based method are its fast detection time and high robustness to measurement noises/errors and load variations. Moreover, the proposed method is useful for both closed‐loop and open‐loop PMSM drives. The proposed method can be embedded as a subroutine in the drive control unit without any hardware extension. The effectiveness of the proposed method is confirmed through extensive simulations and several hardware‐in‐the‐loop experiments on a testbed with a 1.5 kW PMSM and dSPACE1104 control board. In addition, the real‐time feasibility of the proposed method using low‐cost microprocessor technology is guaranteed by testing it on TMS320F28335 from Texas Instruments © .