Electric vehicles are beneficial to the environment owing to its nonproduction of emissions and excessive noise; however, they have their own limitations with respect to charging difficulties and mileage anxiety. In order to address these problems, dynamic wireless power transfer (DWPT) technology has been developed. In this article, we replace a conventional open-loop method without dc-dc converter with the proposed dc-dc converter in the energy receiver in order to improve output power. Moreover, to make sure that the DWPT system is reliable when the coupling coefficient changes rapidly over a wide range of values, we propose model predictive control (MPC) to ensure optimal dynamic-tracking performance. Considering the fact that sampling delay will cause an error in the controller output, the system is modeled accurately in order to improve the MPC performance. In addition, the MPC and double closed-loop proportional-integral-differential control are compared with simulations and experiments, the results of which demonstrate the effectiveness of MPC-based DWPT.

In order to improve the quality of the power injected into a grid, this paper presents a model predictive control strategy for three-phase four-leg grid-tied inverters. For the convenience of optimization, the discrete-time model of the inverter in which duty ratios are modeled as continuous control variables is investigated. A current tracking error oriented cost function is employed as a criterion to optimize duty ratios of the inverters. In order to eliminate the effects of sampling delay, a model predictive control with delay compensation method (MPC-DC) is proposed. Because there is a large amount of calculations in implementing predictive control algorithm, a double-CPU, namely FPGA plus DSP controller, is employed to implement parallel calculation, so as to reduce the computation time. Simulation and experimental results demonstrate the effectiveness of the proposed method.

Electric vehicles (EVs) can behave as distributed energy storage devices for providing on-demand smart grid support service, that is an emerging Vehicle-to-Grid (V2G) technology. A high-performance and easy-implementation current control strategy for V2G Three-phase four-leg inverter with LCL filter is proposed. It consists of a deadbeat (DB) controller and a paralleled repetitive controller (RC). The DB controller is based on weighted average inductor current (WAIC) scheme, which simplifies the third-order LCL filter to be an equivalent 1st order L filter. The stability of the DB controlled inverter with the unmodelled system time delay is analyzed. DB controller is of very fast response and easy implementation, but is not immune to system time delay and various uncertainties. To overcome the disadvantages, a plug-in RC is added to reinforce the DB controller to remove harmonic distortion from the feed-in current in the presence of parameter uncertainties. A lab prototype of 10kW grid-connected three-phase four-leg inverter has been built up to validate the proposed current control strategy. The simulations and experiments are provided to demonstrate the validity of the proposed control strategy.

Durability is considered as one of the main technical obstacles to the large-scale commercialization of proton exchange membrane fuel cells (PEMFCs), which can be effectively improved through degradation prediction techniques. This article proposes a stacked echo state network (ESN) based on the genetic algorithm (GA) to predict the future degradation trend of PEMFCs. By alternately using the projection layer and the encoding layer, the proposed method can make full use of the temporal kernel property of the ESN to encode the multiscale and multilevel dynamics of the stack voltage, thereby obtaining more robust generalization performance and higher accuracy than the existing methods. Specifically, a stack voltage time series of PEMFCs is projected into the high-dimensional echo state space of the reservoir. Then, an autoencoder projects the echo state representation into the low-dimensional feature space. After that, the GA is utilized to optimize the hyperparameters of the developed model. Based on two open-source datasets of PEMFCs with different accelerated test conditions, this article systematically tested the proposed degradation prediction methods based on different model structures. Test results demonstrate that the proposed method is superior to traditional prediction methods in terms of accuracy and generalization performance.