PMSM robust current control with adaptive tuning of axis decoupling

Current control of a permanent magnet synchronous motor is designed for optimal robust performance under both uncertainties in the stator parameters, and variation of operating speed. To avoid the conservatism inherent in many robust control approaches, and to admit a flexible choice of optimization criteria, the design is carried out with the help of a genetic algorithm. The controller is composed of two SISO PI-controllers and a static feedback decoupling network that minimizes the interaction between the d- and g-axes uniformly across the parameter uncertainty range, and across the speed range. Remaining interaction is eliminated by a simple on-line adaptive cancellation. This results in a practically perfect decoupling in the stationary case, while leaving the stability and performance level guaranteed by the robust design intact. It is shown that the stability and parameter convergence of the proposed adaptive cancellation mechanism is robust with respect to stator parameter uncertainties for a relative broad range of reasonable controller parameters. The adaptive mechanism is tested both in simulations and on the actual motor.