High-resistance connection diagnosis in five-phase pmsms based on the method of magnetic field pendulous oscillation and symmetrical components

An on-line approach for diagnosing high-resistance connection (HRC) faults in five-phase permanent magnet synchronous motor (PMSM) drives is presented in this paper. The development of this approach is based on a so-called magnetic field pendulous oscillation (MFPO) technique and symmetrical compone...

Full description

Saved in:
Bibliographic Details
Main Authors: Hao, Chen, He, Jiangbiao, Guan, Xing, Demerdash, Nabeel A. O., El-Refaie, Ayman M., Lee, Christopher Ho Tin
Other Authors: School of Electrical and Electronic Engineering
Format: Article
Language:English
Published: 2021
Subjects:
Online Access:https://hdl.handle.net/10356/153657
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:An on-line approach for diagnosing high-resistance connection (HRC) faults in five-phase permanent magnet synchronous motor (PMSM) drives is presented in this paper. The development of this approach is based on a so-called magnetic field pendulous oscillation (MFPO) technique and symmetrical components method. Under HRC fault condition, a 'swing-like MFPO phenomenon is observed compared to the healthy condition. Furthermore, with the extracted current features in symmetrical components domain, different HRC fault types are successfully identified and distinguished. These fault types include (i) single-phase faults, e.g., HRC fault in phase-A, (ii) two-phase non-adjacent faults, e.g., HRC fault in phase-A&C, and (iii) two-phase adjacent faults, e.g., HRC fault in phase-A&B. Meanwhile, the localization of the faulty phase/phases is also accomplished, and the fault severity is estimated. In this approach, only sensing of the phase currents is needed. Hence, the implementation cost is very low, since the sensory data of the currents are typically already available in the closed-loop vector-controlled drives for control purpose and no additional sensors or related signal conditioning circuits are required. The effectiveness of the presented diagnostic approach is verified by simulations and experimental results.