Evaluating Disturbance Ride-Through Capability of Fast Electric Vehicle Charging Stations Using DVshave

Abstract

Reliable power quality is crucial for electric vehicle charging stations (EVCS), but persistent voltage quality issues in distribution networks pose a significant challenge. This study proposes DVShave, a novel system integrating a dynamic voltage restorer (DVR) with peak-shaving functionality to significantly enhance EVCS resilience. Designed, modelled and tested in MATLAB/Simulink and validated through OPAL-RT real-time studies, DVShave features a DVR supplied by a 700 V energy storage system (ESS) and controlled by an artificial neural network (ANN) using a synchronous reference frame strategy. The system's performance was evaluated under severe symmetrical voltage dips (30%, 60% and 80%), swells, unsymmetrical faults and non-linear voltage conditions. DVShave effectively prevents charging interruptions and maintains stable charging rates during these grid disturbances, notably reducing supply voltage total harmonic distortion from 19.60% to 4.84%. The ANN-based controller demonstrated a small but notable improvement in harmonic distortion elimination compared to PI-based DVRs. Concurrently, its integrated peak-shaving feature leverages the DVR's ESS, using a rule-based control technique to lower the peak-to-average ratio from 1.9 to 1.49 daily over 4.5 h for a charging station with 50 EV chargers.