The receiving coil of dynamic wireless charging (DWC) systems of electric vehicles will be offset relative to the transmitting coil, which will cause a decrease in mutual inductance and thus a decrease in output power. The conventional solution is to add a buck converter on the energy receiving side to compensate for the power drop. Indeed, the receiving-side buck converter is a nonlinear structure, and the corresponding nonlinear control strategy study is required. In addition, the rapid and large-range changes in the mutual inductance of the DWC system will also increase the control difficulty. To solve these problems, we propose a feedback linearization control (FLC) algorithm for the buck converter on the receiving side of the DWC system. We carry out detailed modeling of the receiving-side buck converter, linearize the relationship between input and output, and analyze the stability of the controlled system. The results show that the adjustment speed of the proposed FLC is fast, and it is suitable for the DWC system. Moreover, when the mutual inductance changes rapidly in a wide range, the controlled output of the receiving-side buck converter can still track the reference value stably, demonstrating the effectiveness and superiority of the proposed FLC algorithm.

Dynamic wireless power transfer (DWPT) of electric vehicles (EVs) is considered as an economic and effective solution on traffic electrification. However, the coupling coefficient between the transmitting and receiving coils changes dynamically with the movement of EV, resulting in a sharp drop in energy transfer power and efficiency. In this case, the double closed loop controller with current sharing of interleaved DC/DC converter is proposed to against coupling coefficient variation. And two layers double D coil structure is presented to make the distribution of magnetic density and temperature more uniform. Moreover, Simulation studies are carried out in PLECS to verify the proposed strategy in DWPT system. Results show that the proposed algorithm can not only significantly improve the energy transfer power and efficiency, but also ensure the current sharing of interleaved DC/DC converter, so as to enhance the robustness and stability of DWPT system.

Taking the high real-time performance, good stability and high reliability of DC-DC converter in electric vehicles as the research target, a constrained model predictive control (MPC) based on spider monkey optimization algorithm was proposed. Taking the Buck mode for an example, the equivalent circuit model is obtained under different switching states, and the state space model is established, then the prediction model is established through discretization. Define the appropriate cost function according to the control requirements and add the constraints of the control variables. The spider monkey optimization algorithm(SMO) is introduced to solve optimization problem of the constrained MPC in order to get the optimal solution more quickly and accurately. the simulation was carried out by MATLAB/Simulink, and the simulation results of PI control, MPC based on Particle Swarm Optimization(PSO-MPC) and SMO-MPC was compared. The simulation show that the converter using MPC has better control effect, and SMO is feasible and effective.

In order to solve the problem of device impact and switching speed in dynamic wireless charging device for electric vehicles (EVs), first, a new SiC MOSFET model was established based on manufacturer's model and datasheet. Combined with this model, the hard-switching and soft-switching processes of SiC MOSFETs are compared and analyzed. Although soft-switching can solve the problems of changing switching speed and reducing device impact, it needs to modify the hardware circuit, which is not practical in the changeable environment of dynamic wireless charging of EV s. Based on this question, a new multi-grade voltage drive circuit is designed to meet the various requirements of the drive circuit in the wireless charging process of EVs. Finally, the drive circuit is simulated in Pspice, and its feasibility is explored according to its waveforms.

The DC-DC circuit is widely discussed in Electric Vehicle Charging Systems (EVCS). In this paper, a high-power, high-voltage four-phase interleaved parallel Buck converter topology structure is designed, which is suitable for both contact and contactless charging systems. A small-signal mathematical model of a four-phase interleaved parallel Buck converter is established by constructing the state variables of the converter; a Proportion Integration (PI) controller is designed by analyzing the Bode diagram to compensate the system and ensure the control performance of the closed-loop system; a control strategy based on PI control and maximum current sharing control is proposed. The simulation results demonstrate that, the output ripple of this new type of converter cancels each other, the currents of each phase are balanced, and a large output power can be achieved.

Coupling system has great influence on the transmission performance of wireless charging system. Improving the mutual inductance between coupling coils can effectively improve the transmission efficiency. Using magnetic core is an effective way to enhance mutual inductance of coupling coils. The introduction of magnetic core also leads to core loss. Based on the DD coil structure and considering the magnetic core weight and coupling coefficient, a method for ferrite core design and optimization is proposed, which can reduce the core loss between transmitter and receiver and improve the average utilization ratio of the magnetic core. MAXWELL simulated and analyzed the coupling coil before and after optimization, and discussed its feasibility.

An improved deadbeat plus plug-in repetitive control scheme is proposed for grid-connected three-phase four-leg inverters to achieve fast and accurate feed-in grid current tracking. Discrete-time mathematical model of the three-phase four-leg grid-connected inverter is developed. An improved deadbeat (DB) plus plug-in repetitive control (RC) scheme is proposed to reject periodic disturbances while keeping good dynamic response. Simulation results are provided to validate the effectiveness and advantages of the proposed method.

High computational burden in solving quadratic programming problem is a major obstacle when we apply model predictive control to industrial process. Recurrent neural networks offer a new quadratic programming optimization approach due to its parallel computational performance. In this paper, we present a new architecture of solving model predictive control (MPC) problem based on one layer recurrent neural network. We give algorithm of model predictive control based on recurrent neural network and prove convergence property of one layer recurrent neural network at each sample step. Two examples demonstrate the effectiveness and efficient of the proposed recurrent neural network based MPC. Simulation results show that this approach can utilize fast converge property and the parallel computation ability of recurrent neural network and apply to real-time industrial process control.