Multimode Ultralocal-Model-Based Predictive Current Control of a Variable Reluctance Flux Controllable Permanent Magnet Motor

To solve the parameter mismatch caused by the flux variation effect in variable reluctance flux controllable permanent magnet (VRFCPM) motor drive systems, a multimode ultralocal model (MM-ULM) predictive current control method is proposed and investigated in this article. According to the driving cycle and flux regulation principle of the VRFCPM motor, a MM-ULM with variable control parameters is constructed. The contribution lies in the extrication of the variable motor parameters into two motor control parameters with different operating conditions, which are named as the voltage regulation coefficient and system bandwidth. Then, the characterization model of the motor and control parameters for different operating conditions are presented, where the two control parameters are adjusted according to the variable motor parameters. Furthermore, the extended state observer is designed to estimate the system disturbances and utilized to determine the system disturbance control parameters. The parameter mismatch and performance degradation caused by the variable flux of the VRFCPM motor are avoided, and consequently the steady-state and robustness performance are enhanced. Finally, comparative experiments are conducted to validate the effectiveness of the proposed control strategy.