在当前能源短缺和环境恶化的背景下，可再生能源的利用需求日益增长。基于高性 能电力电子器件与分布式协调控制技术的能源互联网（Energy Internet, EI）架构，已成为 学术界在智能电网领域的研究热点。本文针对能源互联网中的双有源桥变换器（Dual Active Bridge，DAB）存在电流应力较大和动态性能欠佳等问题进行了深入研究。 论文阐述了传统单移相调制下 DAB 变换器的工作特性，指出输入输出电压不匹配 时存在电流应力大的问题。在此基础上，详细分析了拓展移相调制下 DAB 变换器的功 率传输、电流应力和软开关特性。对比传统的单移相调制和拓展移相调制策略，在输入 输出电压不匹配时，采用拓展移相调制能有效拓宽功率调节范围，并可以减小电流应力， 提高变换器效率。 为了降低输入输出电压不匹配时的电流应力，以峰值电流表达式为目标函数，采用 拉格朗日乘数法构建数学模型，根据内外移相比的大小，将拓展移相调制分为两种模式， 求解两种不同运行模式的最优移相比。对比不同运行模式优化后的电流应力，确保在可 行域内峰值电流最小。通过分析拓展移相两种模式下传输功率和电压传输比的关系，得 到不同功率范围内开关管的 ZVS 条件，实现了两种模式下所有开关管的 ZVS。最后，采 用动态矩阵控制（Dynamic Matrix Control）算法，通过确定控制量使预测值不断接近给 定值，减小变换器电压波动，加快其动态响应速度。 为了验证优化策略的有效性，利用 Matlab/Simulink 软件构建仿真模型。并搭建了 300W 的 DAB 变换器实验平台，通过该实验平台对单移相调制和拓展移相调制策略分别 进行了电流应力、软开关特性及动态特性测试，与单移相调制策略相比，在输入输出电 压不匹配的情况下，拓展移相调制下的双有源桥变换器电流应力降低了 12.1%，并且可 以实现全功率范围内的软开关；动态矩阵控制算法有效改善了变换器的动态响应性能。

Against the backdrop of current energy shortages and environmental degradation, the demand for renewable energy utilization is increasing day by day. Energy Internet (EI) architecture based on high-performance power electronic devices and distributed coordinated control technology has become a research hotspot in the field of smart grid in academia. This paper focuses on the problems of high current stress and poor dynamic performance of dual active bridge (DAB) converters in the energy internet. The paper elaborates on the working characteristics of DAB converters under traditional single phase shift modulation, and points out the problem of high current stress when the input and output voltages do not match. On this basis, the power transmission, current stress, and soft switching characteristics of DAB converters under extended phase-shift modulation were analyzed in detail. Compared with traditional single phase shift modulation and extended phase shift modulation strategies, using extended phase shift modulation can effectively broaden the power regulation range and optimize the peak current to improve the efficiency of the converter when the input and output voltages do not match. In order to reduce the current stress when the input and output voltages do not match, a mathematical model is constructed using the Lagrange multiplier method with the peak current expression as the objective function. Based on the magnitude of the internal and external shift ratio, the extended phase-shift modulation is divided into two modes, and the optimal shift ratio for the two different operating modes is solved. Compare the optimized current stress of different operating modes to ensure the minimum peak current within the feasible range. By analyzing the relationship between transmission power and voltage transmission ratio in two extended phase-shift modes, the ZVS conditions of switch tubes in different power ranges were obtained, and the ZVS of all switch tubes in both modes was achieved. Finally, the Dynamic Matrix Control algorithm is adopted to continuously approach the given value by determining the control quantity, reducing the voltage fluctuation of the converter and accelerating its dynamic response speed. To verify the effectiveness of the optimization strategy, a simulation model was constructed using Matlab/Simulink software. And a 300W DAB converter experimental platform was built, through which current stress, soft switching characteristics, and dynamic characteristics were tested for single phase shift modulation and extended phase shift modulation strategies, respectively. Compared with the single phase shift modulation strategy, the current stress of the dual active bridge converter under extended phase shift modulation was significantly reduced in the case of input and output voltage mismatch, and soft switching could be achieved in the full power range; The dynamic matrix control algorithm effectively improves the dynamic response performance of the converter.