An Improved Active Flux Sliding Mode Observer Based on Fuzzy SOGI for IPMSM Sensorless Drives

An improved active flux sliding mode observer (IAFSMO) and a fuzzy second-order generalized integrator (FSOGI) are proposed in this paper to address the issues of sliding mode chattering and high-order harmonics inherent in the traditional sliding mode observer (SMO) utilized for sensorless control of interior permanent magnet synchronous motor (IPMSM). Firstly, to suppress chattering, a sinusoidal saturation function is designed in this paper, with its boundary layer capable of adaptive adjustment based on speed. This enhancement not only enhances the estimation accuracy of the active flux but also diminishes the estimation errors of rotor position and speed. Secondly, to mitigate the impact of high-order harmonics on the estimation accuracy of rotor position, an FSOGI is proposed in this paper. Fuzzy rules are employed in this method to adaptively adjust the system gain, thereby achieving optimal filtering and dynamic performance. Ultimately, the rotor position information is precisely extracted from the active flux by utilizing the phase-locked loop (PLL), enabling sensorless control of IPMSM. Experimental results validate the effectiveness and feasibility of the proposed method.