The Optimal "Speed-Torque" Control of Asynchronous Motors in the Field-Weakening Region Based on ADRC and ELM

In this paper, the Active Disturbance Rejection Control (ADRC) is introduced to improve the speed loop of the control system to speed up the response speed of "speed-torque" and reduce overshoot. The Extreme Learning Machine (ELM) is introduced to improve the current loop to solve the problem of poor accuracy of the d-q axis stator current (i sd , i sq ) caused by the influence of temperature on the motor time constant (T r ). Thus, the optimal "speed-torque" control of asynchronous motor for electric vehicle in fieldweakening region is realized. First, the analytical model of the maximum electromagnetic torque is established, and the effects of i sd and i sq on the torque are analyzed. Second, to improve the response speed of "speed-torque" and reduce its overshoot, ADRC-PI is designed. Third, the maximum torque vector control system based on ADRC-PI is built to verify the effectiveness of the system by simulation. Then, an AVL (AVL List GmbH) experimental platform based on ADRC-PI control system is built, and the data samples under the optimal "speed-torque" are obtained by constant temperature experiment at 85 °C. ELM prediction model with slip, i sd , i sq , stator flux as input and d-q axis stator voltage as output is established. Finally, the ELM predictive model is embedded in the vector control system to output the optimal "speed-torque". Experiments verify the effectiveness of the proposed method.