Speed and Position Estimation for 5-ph PMSM Using SOGI Based on SMO Considering Short-Circuit Fault

Abstract

To cancel harmonics and get fundamental back EMFs for precise rotor position and speed estimations under short-circuit (SC) fault, this article introduces a new sensorless fault-tolerant control method for 5-phase PMSM using a sigmoid sliding mode observer (SSMO) with a second-order generalized integrator (SOGI) to replace the low-pass filter (LPF). For tuning SOGI, two different frequency estimators are proposed. First, a closed-loop synchronous reference frame phase-locked loop (SRF-PLL) is fed back to SOGI for frequency adaptation based on two orthogonally estimated filtered-back EMFs. Second, an independent open-loop frequency estimator (OLFE), based on two-estimated adversely distorted back EMFs, is proposed to ensure simple implementation. The speed and rotor position estimations are eventually accomplished from the quadrature PLL. Both SOGI-SRF-PLL and SOGI-OLFE-based SSMO achieve satisfactory estimation performance under SC fault conditions, eliminate chattering issues, have good tracking errors, and have strong robustness against different disturbances. The results of simulation and practical experiments proved the effectiveness of these methods in all conditions.

Keywords

• Multi-phase PMSM
• single-phase short-circuit fault
• sensorless fault-tolerant control FTC
• sliding mode observer SMO
• second-order generalized integrator SOGI