A Bi-stable permanent magnet actuator (BPMA) shares the same magnetic circuit as the breaking and closing coils, and the magnetic flux generated by any coil passes through the breaking and closing air gaps. The permanent magnet automatically distributes the permanent magnetic flux according to the dynamic reluctance of the air gaps. The electromagnetic flux and permanent magnetic flux in the upper and lower air gaps always cause the moving iron core to be coupled by two opposite magnetic forces. As the motion of the moving iron core and the change of coil current, the magnetic circuit quickly saturates, and the electromagnetic flux and permanent magnetic flux interact, exacerbating the complexity of nonlinear coupling in the breaking and closing air gaps. To flexibly control the action characteristics of permanent magnet switches, it is necessary to simultaneously control the air gap flux and the magnetic force pointing to the breaking and closing positions. Therefore, this paper proposes an air gap flux decoupling control method based on finite control set-model predictive control (FCS-MPC). Decoupling control can be achieved by rapidly weakening the magnetic force pointing to the non-excited coil and rapidly increasing the magnetic force pointing to the excited coil.

Firstly, according to the operating principle of BPMA, the vector magnetic force acting on the moving iron core depends on the “magnetic flux squared difference” of the breaking and closing air gaps. Therefore, only controlling this vector magnetic flux square difference in real-time can dynamically control BPMA. Secondly, a predictive model of the breaking and closing air gap magnetic flux is designed through discretization of the voltage balance equation, which can predict the magnetic flux at the next moment based on the voltage and current values collected at the current moment. Thirdly, the breaking and closing air gap magnetic flux and the mechanism drive circuit are regarded as a whole. A set of switching states is constructed through the excitation intensity analysis under different switching states. Predictive magnetic flux is obtained by traversing all switching state combinations. Finally, a decoupling control cost function is designed, the predictive magnetic flux under different switching combinations is input into the cost function, and the optimal control is selected for the next control period. In rolling optimization over multiple control periods, the breaking and closing air gap magnetic flux quickly approaches their respective reference values, achieving decoupling control.

A co-simulation platform for intelligent control is designed based on LabVIEW and Multisim, and hardware testing circuits are constructed. The simulation and experimental waveforms show that this proposed scheme can effectively control the breaking and closing air gap flux. As a result, the non-excited air gap flux to zero is quickly reduced, approaching the set reference value of the excited air gap flux and effectively weakening the coupling between the air gaps. Compared with the traditional current closed-loop control scheme, the proposed control scheme reduces the energy loss during the entire action process and improves the response and action time of the core action.