An Improved Model Predictive Control Approach for Wireless Power Transfer System of Electric Vehicles

In the context of implementing dynamic wireless power transfer (DWPT) in electric vehicles (EVs), the mutual inductance of the magnetic coupler is continuously changing. Therefore, a controller with enough dynamic response is required to ensure the utilization of the energy transfer capacity and stability. In this paper, a model predictive control (MPC) approach with variable step-size and optimization-point-number is proposed to balance the dynamic response speed, steady-state error and computational burden of the system. The dynamics of the wireless power transfer (WPT) system is modeled and the boundary conditions for the zero-voltage-switching (ZVS) implementation are derived. The rules for the selection of step-size and optimization-point-number are established and it is worth noting that the variance of the output power is introduced as a basis for judgment, which greatly accelerates the convergence rate near the steady state. Finally, the simulation actively verifies the feasibility of the proposed controller.