With rising electric vehicles (EV) penetration, industrial parks and microgrids face increased load volatility and peak stress. This study aims to develop a coordinated EV charging/discharging strategy that simultaneously reduces load fluctuations, ensures state of charge (SOC) targets and protects battery lifetime.
The authors propose a linear programming model predictive control (LP-MPC) that, at every time step, linearizes SOC dynamics, charging/discharging constraints and battery-degradation cost to solve a multi-objective optimization for peak shaving, SOC target fulfillment and user benefit maximization. The method is evaluated in large-scale EV coordination simulations and compared with Greedy and predictive Greedy benchmarks.
LP-MPC markedly lowers load variance, raises SOC completion rates and substantially mitigates cumulative battery degradation compared with the benchmark strategies, demonstrating improved trade-offs among grid stability, charging performance and battery health.
The paper introduces a practical LP-MPC formulation that explicitly embeds lifetime-aware cost into real-time multi-objective EV scheduling; this provides a computationally tractable, scalable approach for large-scale EV coordination in industrial parks and microgrids.
