In the last decade, the market for Electric Vehicles (EV) has expanded at an unprecedented rate. Concurrently, there has been a surge in the demand for public Direct Current (DC) EV charging stations. To accommodate a diverse array of EVs, these charging stations are required to support a broad operational voltage and power spectrum. However, this necessity introduces a trade-off between the operational range and the efficiency and utilization of the installed power within the charging infrastructure. The core of the issue lies in the fact that the charging stations, designed to cater to a wide range of operational conditions, often do not operate at their peak efficiency outside their designated optimal performance region.
To address this challenge, this thesis investigates the advantageous DC/DC power electronics solutions that can maintain high efficiency across a wider operational range while optimizing the utilization of installed power. The research topics include the suitable DC/DC converters for the back-end power modules of the charging system, the multi-objective design process of the power modules, and the advanced operation of the power modules.@en

Traditional methods such as Steinmetz's equation (SE) and its improved variant (iGSE) have demonstrated limited precision in estimating power loss for magnetic materials. The introduction of Neural Network technology for assessing magnetic component power loss has significantly enhanced accuracy. Yet, an efficient method to incorporate detailed flux density information—which critically impacts accuracy—remains elusive. Our study introduces an innovative approach that merges Fast Fourier Transform (FFT) with a Feedforward Neural Network (FNN), aiming to overcome this challenge. To optimize the model further and strike a refined balance between complexity and accuracy, Multi-Objective Optimization (MOO) is employed to identify the ideal combination of hyperparameters, such as layer count, neuron number, activation functions, optimizers, and batch size. This optimized Neural Network outperforms traditionally intuitive models in both accuracy and size. Leveraging the optimized base model for known materials, transfer learning is applied to new materials with limited data, effectively addressing data scarcity. The proposed approach substantially enhances model training efficiency, achieves remarkable accuracy, and sets an example for Artificial Intelligence applications in loss and electrical characteristic predictions with challenges of model size, accuracy goals, and limited data.@en

This article presents an analysis, multiobjective design, and benchmark of three modified 3 phase shift full-bridge (PSFB) converters that are well-suited for electric vehicle (EV) battery charging applications, covering both typical battery voltage classes (400 and 800 V). These three modified PSFB converters, denoted as the t-PSFB, r-PSFB, and i-PSFB converters, have the ability to reconfigure and provide better efficiency performance in the wide voltage range necessary for public EV battery charging applications. In this article, the characteristics and design considerations of these reconfigurable PSFB converters are discussed in detail. A multiobjective converter design process is proposed to optimize the average efficiency, normalized cost, and power density of the magnetic components and heat sinks. This design process employs the correlations between the cost and performance indexes of the key components derived based on open and accessible components' data to estimate the design objectives. In this way, the design process is not constrained by certain component choices, making it easier to identify the most advantageous design. A benchmark study is conducted among the reconfigurable PSFB topologies and the conventional PSFB circuit using the proposed multiobjective design process. To validate the analysis, a close-to-Pareto-front 11-kW, 45-kHz r-PSFB converter prototype with 640-840-V input voltage and 250-1000-V output voltage ranges is developed and tested.

@en

A variable switching frequency modulation for the Dual Active Bridge (DAB) converter is proposed in this paper. With this variable switching frequency modulation, the DAB converter can be operated in the ZVS-beneficial operational modes without the necessity to transition to others, thus a larger ZVS range for the DAB converter can be achieved. This modulation has potential of providing higher power efficiency and better EMI performance for the DAB converter in wide voltage range applications such as Electric Vehicle (EV) charging. A DAB converter with the variable frequency modulation method is simulated, and its effectiveness on the ZVS performance is demonstrated.

@en

This paper proposes two modulation schemes for Dual Active Bridge (DAB) converters, with the aim of maximizing Zero Voltage Switching (ZVS) operation over a wide operational range. The first is a ZVS-optimized constant frequency modulation scheme, constructed based on the boundary conditions of ZVS operation. This scheme maximizes the number of ZVS events across a broad operational range and is easy to implement. Additionally, a variable frequency modulation scheme is proposed, enabling continuous full ZVS operation for the DAB converter at full power and eliminating the loss of ZVS due to transitioning between modulation regions. This functionality extends the full ZVS range, yielding improved Electromagnetic Interference (EMI) performance and overall power efficiency. The synergy of the proposed modulation schemes is particularly well-suited for applications like off-board Electric Vehicle (EV) charging. Experimental validation, conducted on an 11-kW DAB converter prototype with an output voltage range of 250V to 950V, demonstrates the efficacy of the proposed schemes in achieving ZVS and boosting converter efficiency.

@en

This article analyzes, designs, and tests a reconfigurable phase shift full-bridge (r-PSFB) isolated dc/dc converter well suited for a wide voltage operating range. By controlling the auxiliary switches, a series or parallel connection can be realized on the secondary side of the converter. As a result, the r-PSFB converter can operate in an extremely wide voltage range without compromising the system efficiency. In this article, the characteristics of the r-PSFB converter and its design considerations are discussed in detail. An 11-kW r-PSFB converter prototype with 640-840-V input voltage and 250-1000-V output voltage ranges is developed and tested to validate the analysis and efficiency of the designed converter. A comparative study against a conventional PSFB converter is conducted for benchmark purposes to prove the advantages of the studied r-PSFB converter.

@en

This paper compares the influences of two different charging strategies on the performance of the Electric Vehicle (EV) charging system, including the EV battery, the Battery Management System (BMS), and the power electronic converter. Firstly, the two types of EV charging strategies are explained. Herein, the characteristics, the influences on the EV battery, and the respective requirements for the hardware and communication of the BMS and power electronic converter are summarized. Secondly, a simulation case study is conducted to investigate the influences of the two charging strategies on the performance of a typical power electronic converters employed as a back-end converter of a fast EV charger. A modular 50kW DC-DC EV charger constructed with five parallel-connected 10kW Phase-Shift Full-Bridge (PSFB) DC-DC converter is modelled with two different charging strategies, and two comparison metrics are proposed for a benchmark: the charging-cycle efficiency and the installation utilization rate. The results show that these two comparison metrics are influenced by the charging strategies of the EV.@en

This paper verifies the potential of utilizing the Dual Active Bridge (DAB) converter as a 10 kW Power Electronics Building Block (PEBB) for the charging of Electric Vehicles (EVs) where a wide output voltage range is required. Following the tendency of the new EV charging standards from CCS and CHAdeMO, the commonly employed isolated DC-DC converter should be able to supply the rated power for different vehicles which employ battery technologies which could range from 300 V up to 1000 V. Rectangular (or phase-shift), triangular and trapezoidal modulation strategies are investigated, of which the last two techniques are used for designing the PEBB. Suitable SiC-based semiconductors are chosen, the passive components are designed and the power losses are derived for the whole operating range. It is found that a high power efficiency can be achieved using the designed system.@en