Data-Driven Online Induction Motor Parameter Adjustment Method for Maximum Torque Output
Xing Qi,Qian Zhang,Lassi Aarniovuori,Wenping Cao
DOI: https://doi.org/10.1109/tie.2023.3303620
IF: 7.7
2023-01-01
IEEE Transactions on Industrial Electronics
Abstract:Electric vehicle motors are expected to both run stably and output maximum power. However, most existing online motor parameter estimation methods are model-based, which aim to estimate parameter values for stable running but not necessarily for maximum power output. In order to address this issue, this article proposes a data-driven rotor resistance adjustment method, aiming to ensure that induction motors output maximized torque over a broad range of operating conditions. The method differs from traditional model-based approaches, as it converges to the motor's optimal states through analyzing operating data, rather than relying on mathematical models. The proposed method consists of two stages: an offline data-mining stage and an online parameter adjustment stage. With the computationally intensive part performed in the offline stage, the timeliness of the online operation can be ensured. Experimental results show that the method can complete the parameter adjustment task within 1–5 s during online operation, and result in 10%–15% higher torque production than classical model-based methods under a wide range of operation conditions. Therefore, the proposed method is well-suited for electric vehicle motor control. Furthermore, comparative studies indicate that the proposed method's performance remains unaffected by model errors, showing better robustness than traditional model-driven methods.
automation & control systems,engineering, electrical & electronic,instruments & instrumentation