Comparative Studies of Front-End Model Predictive Control for Direct Inductive Power Transfer Systems

Abstract

In this article, two transmitter-side model predictive control (MPC), namely, a model predictive-based phase shift control (MPPC) and a model predictive-based frequency control (MPFC), are proposed for direct inductive power transfer (IPT) systems. Both MPC strategies are empirically verified to have better dynamic responses than the proportional-integral (PI)-based counterparts using inexpensive digital controllers, but the MPFC is proven to be better than the MPPC for direct IPT systems if variable frequency operations are allowed. Both simulation and experimental results confirm that the MPPC will suffer from notching effects and high-order harmonics caused by the dead time of transmitter-side inverters, while the MPFC is immune from the dead time effects. The results also validate that the MPFC with a parallelly proposed initial frequency selection algorithm can cover the shortcomings of the MPPC in achieving soft switching for IPT systems over wide load conditions. Besides, the transmitter-side inverter with MPFC is demonstrated to have lower radiation noise than the same inverter with MPPC in practice.