Multi-condition dynamic model control strategy of the direct drive motor of electric vehicles based on PIO–LightGBM algorithm.

Permanent magnet synchronous motor (PMSM) is widely used in new energy vehicles. At present, to make electric vehicles have a wider speed range, the motor can reach the rated speed above through the field-weakening control. However, when the traditional field-weakening control strategy is above the rated speed, the dynamic response ability of the vehicle declines. Problems such as torque oscillation and current jump occur. To solve these problems, based on a pigeon-inspired optimization (PIO) algorithm and optimized light gradient boosting machine (LightGBM), the dynamic response capability of the permanent magnet synchronous motor is improved. The robust adaptability of the control system to disturbances and parameter changes is also further improved. By collecting experimental data, the importance of relevant variables is analyzed, and the variable with the largest weight is selected as the input of the model. PIO is used to optimize LightGBM, and the loss function is optimized. Finally, the regression model is established. Simulation and experimental results show that the method is effective. [ABSTRACT FROM AUTHOR]