Active Flux Based Predictive Control of Interior Permanent Magnet Synchronous Machine

In this paper, the predictive control of Interior Permanent Magnet Synchronous Machine (IPMSM) based on the Active Flux (AF) concept is developed. The speed tracking is based on a linear PI controller in the external loop, while the fast variables as the electromagnetic torque, or currents, and magnetic flux are compensated by a predictive controller. The set points of the current and torque are generated by a lookup table searching and a spline interpolation process with Maximum Torque Per Ampere (MTPA) algorithm. The cost function has been improved by adding the AF term. In this case, it is searched the magnetic flux that ensures the high torque developed by the IPMSM. The numerical simulations performed for different cost functions present different particularities. The cost function form influences the output response of the speed and torque. It can be concluded that the cost function may be selected for reducing the computational effort based on statistical analysis.