Abstract: Parameter optimization of the controller is a key step in improving system performance. In order to solve the problems of slow dynamic response and weak robustness in the traditional manually tuned speed loop controller parameters of permanent magnet synchronous motors, an Improved Crested Porcupine Optimizer (ICPO) algorithm is proposed and applied to optimize the PID controller parameters of the speed loop of permanent magnet synchronous motors (PMSM) to improve system performance. In response to the problems of low convergence accuracy and susceptibility to local optima in the Crested Porcupine Optimizer (CPO) algorithm, the ICPO algorithm first initializes the population with a set of optimal points to improve the algorithm’s traversal ability; Secondly, introducing the sine cosine perturbation mechanism to perform additional perturbation operations on individuals whose fitness has not been optimized, in order to expand the search range and enhance the algorithm’s ability to escape local optima; Finally, the Cauchy Gaussian mixture mutation strategy is used to update the position of the individual best to further improve convergence accuracy and global optimal probability. The simulation results show that the ICPO algorithm has stronger global optimal solution search ability and computational accuracy. A PMSM control model is built in MATLAB/SIMULINK. Through simulation experiments, it is found that the motor control system optimized by the ICPO algorithm has better dynamic performance and stability, and the system performance has been effectively improved.