Open-Circuit Fault Diagnosis of Six-Phase Permanent Magnet Synchronous Motor Drive System Based on Empirical Mode Decomposition Energy Entropy

Abstract

This paper proposes a method to diagnose the open-circuit faulty phases and faulty points of the six-phase permanent magnet synchronous motor (PMSM) drive circuit. The current sensor is used to obtain the six-phase current signal, and the least mean square error (LMS) adaptive filtering algorithm is used to filter out the vibration and noise. Empirical Mode Decomposition (EMD) is performed on the filtered current signals, and the EMD energy entropy of each phase current signal is calculated. The change of energy entropy can simplify the double-bridge arm open-circuit fault to the single-bridge arm opencircuit fault, which reduces the number of fault characteristics. After the faulty phase is judged by the change of energy entropy, the fault diagnosis system can diagnose the specific faulty point according to the normalized average value of each phase current signal. Finally, the current data of normal state and fault state are used to train the support vector machine (SVM) and classify the fault state. Each type of fault can be accurately diagnosed by substituting experimental data into the SVM. Experimental results prove that the proposed method can accurately diagnose the open-circuit faults of the six-phase PMSM drive system. INDEX TERMS Six-phase permanent magnet synchronous motor, open-circuit fault, empirical mode decomposition, energy entropy, support vector machine.

Torque Improvement Effect of Reverse-Phase Superimposition for Fault-Tolerance in Permanent Magnet Synchronous Motor

“…This means that the amplitude required to be unified and the phase difference must be reversed 180°. In this case, the combination of current amplitude and phase is the same for all methods, as shown in (11). For 7-phase motors, the phase current is energized using Eqn.…”

“…Typical failures are open winding failures in motors [11][12] and short-circuit failures in inverter modules [13][14]. This paper focuses on open failures for motor windings.…”

A Novel, Simple, and Flexible Fault-Tolerant Control Algorithm for Multiphase Electrical Machine Operation Under Open Circuit Faults

“…Some of them are focused on the theories of Artificial Neural Networks (ANNs) [19] and Fuzzy Logic (FZ) [20][21][22], where the development of a Knowledge Base allows the implementation of the rules in the Expert System, which also provides the data processing for the fault diagnosis. Other recent studies have proposed techniques based on the energy entropy change in the whole drive system [23], where the diagnosis is achieved by computing the empirical mode decomposition energy entropy of the currents flowing through the m-phase machine, or on the theory of symmetrical components, in which the fault-tolerant condition can be identified by analyzing the system degradation in terms of possible unbalanced conditions of the electrical quantities involved in the machine after the OCF or SCF [24][25]. Further methods consider the adoption of different winding layout arrangements [26], or the dynamic model of the machine that adopts either state or parameter estimation [27][28].…”

Quantitative Analysis of Broken Rotor Bars in Cage Motor Based on Energy Characteristics of Vibration Signals

“…On the other hand, the wavelet bases are explicitly chosen in the wavelet transform process and cannot be adaptively adjusted during the decomposition process according to the signal characteristics [20]. Unlike the wavelet transform, empirical mode decomposition is an adaptive processing technique that can be applied to the analysis of complex signals based on the inherent characteristics of the signal [21]. Although it avoids the choice of decomposition layers and wavelet bases and has multiresolution analysis capability, while there are sudden changes or disturbances in the signal, part of the timescale will be lost, leading to a severe modal mixing phenomenon [22].…”