Proactive Low-Frequency Ride-Through Method and its Ripple Reduction for Sensorless Induction Motor Drives under Load Variations

In recent years, the sensorless induction motor drives (SIMD) technique has seen significant advances in industrial applications and electrical transmission systems. Compared to sensor-based control strategies, the SIMD technique has low cost, high reliability, and increased resistance to harsh environments. Conventional approaches of SIMD, such as model reference adaptive system, extended Kalman filter, sliding mode observer, and full-order flux observer, nonetheless, fail to estimate the rotor speed stably in the low-speed regenerating mode. Therefore, a proactive low-frequency ride-through (PLFRT) method and its ripple reduction technique for sensorless induction motor drives under load variation are proposed in this paper. Firstly, based on the determinant analysis of the error coefficient matrix, the instability of the IM in low-speed regenerating mode is analyzed. The influence of load variation on IM is revealed. With constant magnetizing current under changing load conditions, two problems exist: speed unobservability and poor system stability. Thus, the PLFRT method is introduced. At a fixed moment, the electromagnetic torque keeps constant. The change of the magnetizing current means the synchronous change of the torque current and the slip speed. Therefore, The PLFRT method can modify the magnetizing current reference and then the synchronous speed will be prevented from the unstable region. When the synchronous speed reaches the preset limit, the PLFRT method will be triggered, and the current reference will be modified. Then speed observability is ensured during the crossing process. The adaptive increase of the magnetizing current reference will be triggered automatically and repeatedly until the IM is in the appropriate range to cross the low-frequency region. The adaptive variation interval of magnetizing current and its correction step length are derived based on the stator current limit. The influence of synchronous speed limit is studied to enhance the torque output capacity of the sensorless IM control system. The stator current limit with the implemented PLFRT method is discussed, and the value range of the synchronous speed limit is also given. The suggested selection of the limit value is presented. In the real induction motor system, an inevitable phase delay exists between the magnetizing current and torque current. Hence, the currents cannot change simultaneously, leading to a mismatch between the electromagnetic torque and the load. Such problems occur with ripples of rotor speed, estimated flux, and great torque overcurrent. A torque current phase compensation method is developed to ensure real-time matching of the electromagnetic torque and the load. In the discrete system, the amplitude of the torque current is added in every beat to meet the theoretical value to suppress the ripples of speed, flux, and current during zero-frequency ride-through. The effectiveness of the proposed method is validated on a 2.2 kW induction motor experimental setup. Performance with conventional and the PLFRT method under variable regenerating load is presented. Although the ripples exist, comparative experiments demonstrate that the PLFRT method can cross the zero-stator-frequency line successfully. Then, the torque current compensation method is applied in the experiments. The ripples and the overcurrent of the torque current are decreased. Furthermore, the estimated flux trajectory of the SIMD system with the PLFRT method enabled keeps convergent and observable during the crossing process. © 2023 Chinese Machine Press. All rights reserved.