Sensorless Control of PMSM Based on A Novel Adaptive Gain Super Twisting Algorithm-Sliding Mode Observer

Jin Gao,Qifan Xu,Mingzhe Zhang,Qianxiang Liu
DOI: https://doi.org/10.1109/EEET61723.2023.00023
2023-12-01
Abstract:In the sensorless control system of Permanent Magnet Synchronous Motors (PMSM), the traditional first-order sliding mode observer (SMO) exhibits significant observation errors and system oscillations across varying speed ranges. Leveraging the super twisting algorithm, a design was formulated that places the switching function in the higher-order derivatives of the sliding variable, ensuring system stability and addressing oscillations.To resolve the limitations of fixed gain SMO, which inaccurately accommodate different speeds and loads, a novel adaptive gain that varies with speed and load changes was developed. This adaptive gain aims to enhance the accuracy of motor speed and rotor position observation. Stability was verified through the global bounded conditions of the super twisting algorithm. Simulation comparisons demonstrated that the novel adaptive gain super twisting algorithm-sliding mode observer (STA-SMO) significantly improves observation accuracy and reduces system oscillations.
Computer Science,Engineering
What problem does this paper attempt to address?