基于氢的光伏电池储能系统中用于混合动力电动汽车充电站的电网同步与孤岛控制

This article proposes a widely linear complex-valued estimated-input-based least mean square (WLC-EILMS) control strategy for a grid-supportive photovoltaic (PV) battery system, aimed at supporting hydrogen production and hybrid EV charging stations. The charging system enables seamless transitions between grid-connected and islanded modes, ensuring continuous operation during grid disturbances. The proposed strategy effectively suppresses harmonic currents from nonlinear loads, maximizes PV power extraction through a feed-forward loop, and enables automatic grid resynchronization. In grid-connected mode, it ensures accurate current estimation under distorted conditions, while in islanded mode, it maintains stable power supply to both the electrolyzer and local loads. Comparative analysis confirms the superior dynamic performance of WLC-EILMS, with a steady-state error of 0.537 A, rise time of 3.799 ms, weight convergence in 66.680 ms, and recovery response of 58.765 ms. The method also provides load balancing, harmonic mitigation, reactive power compensation, and power factor correction, resulting in improved power quality. Simulation and experimental results confirm the robustness and effectiveness of the proposed system, achieving THD less than 5% in accordance with IEEE 519-2022 standards and demonstrating its suitability for real-time smart grid interactive hybrid EV charging applications.