Discrete Sliding-Mode-Based MRAS for Speed-Sensorless Induction Motor Drives in the High-Speed Range

By combining the high robustness of sliding-mode observer (SMO) with the adaptability of model reference adaptive system (MRAS), the sliding-mode-based MRAS can improve the speed estimation performance of speed-sensorless induction motor (IM) drives. However, the analysis of this article shows that there exists a mismatch problem between the fixed-value of SMO gains and the variable stability boundary, which changes with IM operation speed, resulting in speed estimation performance degradation or even instability in high-speed range. To cope with this problem, a discrete SMO is proposed to expand the effective operation range of sliding-mode-based MRAS. First, a discrete terminal SMO with modified observer structure is designed for rotor flux estimation. Then, a variable gain is designed to ensure the SMO stability in high-speed range, and avoid the chattering caused by gain value overestimation. Compared with existing sliding-mode-based MRAS, the studied method can improve the stability and accuracy of speed estimation in high-speed range. Finally, the experimental results from a 3.7 kW IM test bench can confirm the validness of studied speed estimation scheme.