A Novel High Frequency Signal Injection Strategy for Self-Sensing Control of Electric AC Machine Drives

Abstract

In this paper, an improved high frequency (HF) injection method coupled with a robust observer is proposed for Interior Permanent Magnet Synchronous Motor (IPMSM), to estimate position, speed and acceleration without mechanical sensors. The contribution is twofold. First, a square-wave HF signal injection method with a new demodulation process is adopted to improve the performances of the self-sensing control. Second, an extended sliding mode observer combined with the HF method is proposed to get over the knowledge of machine parameters (inductances, inertia, frictions,...) and to improve the estimation error in transient modes, which are the drawbacks of existing tracking algorithms (phase-locked loop (PLLs), mechanical system observers, inverse of tangent). Simulation and experimental results are provided to show the effectiveness and the robustness of the proposed strategy in the framework of electric propulsion used in automotive applications.