Open AccessSimilarity Analysis-Based Diagnosis Method for Open-Circuit Faults of Inverters in PMSM Drive Systems
Author(s) -
Tao Chen,
Yonglong Pan
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/571/1/012025
Subject(s) - fault (geology) , similarity (geometry) , fault detection and isolation , measure (data warehouse) , computer science , control theory (sociology) , voltage , similarity measure , engineering , data mining , artificial intelligence , control (management) , electrical engineering , seismology , actuator , image (mathematics) , geology
A fast, real-time and similarity analysis-based diagnosis method for open-circuit faults in three-phase voltage-source inverters is proposed. The proposed method is based on the mixed logical dynamic (MLD) model of the PMSM drive system to simulate three-phase currents under different faulty conditions, and the Pearson correlation coefficients (PCCs) are used to measure the similarity between the simulated fault types and the actual one, then the fault detection and isolation can be achieved according to the characteristics of PCCs under different faulty conditions. The proposed method only needs to use the current sensors existing in the closed-loop system to measure the currents for fault diagnosis, and can detect the fault within one current fundamental wave period. Simulation results verify the effectiveness of the proposed method.