Optimal Energy Management of Electric Vehicle Charging and Discharging in Coupled Urban Transportation and Power Networks

ABSTRACT

This paper proposes a two‐stage day‐ahead stochastic energy management framework for coordinated electric vehicle (EV) charging and discharging in coupled urban transportation networks and electric power distribution systems (UTN–EPDS). Stage 1 integrates probabilistic trip chains, Dijkstra‐based routing, and road‐, traffic‐, and temperature‐dependent energy models to determine EV location, state of charge, parking availability, and vehicle‐to‐grid (V2G) flexibility. Stage 2 formulates a stochastic mixed‐integer nonlinear programming model that jointly coordinates EV users (EVUs), plug‐in EV parking‐lot operators (PEV‐PLOs), and the electric power distribution system operator (EPDSO), while considering battery degradation, parking‐lot limits, nonlinear AC power flow, and voltage and thermal constraints. Differential evolution is used to obtain feasible near‐optimal schedules. Case studies on a 69‐bus EPDS with 15‐min resolution show that 80%–85% of the 100‐EV fleet remains parked even during the busiest travel period. Coordinated charging reduces EVU net cost by 11.20% and eliminates line‐loading violations. Bidirectional V2G increases PEV‐PLO profit by 187.36%, reduces EPDSO operating cost and power‐loss cost by 1.37% and 4.16%, respectively, and improves the combined system objective by 1.68%. At 300 EVs, voltage and thermal violations remain zero.