Spatiotemporal coordination of electric vehicle traffic and energy flows in coupled power-transportation networks with multiple energy replenishment and vehicle-to-grid strategies

Abstract

With the growing integration of electric vehicles (EVs) and diverse energy replenishment infrastructures, the interdependence between the power distribution network (PDN) and the transportation network (TN) has become increasingly prominent. In this context, efficient spatiotemporal coordination of EV users’ routing decisions in TN and energy replenishment and Vehicle-to-Grid (V2G) decisions in PDN by virtue of locational dynamic pricing while taking into account the respective network constraints, constitutes a major task in transportation electrification. This paper proposes a bi-level optimization model to achieve coordinated operation of the coupled power-transportation networks (CPTN). To resolve the challenges faced by static and semi-dynamic TN modeling, a novel linear program based dynamic traffic assignment model is proposed in the upper level (UL) to capture the spatiotemporal evolution of EV traffic flows on roads, fast charging and battery-swapping stations, considering EVs’ diversified energy replenishment and V2G strategies for the first time. Subsequently, an energy flow model is constructed for accurate spatiotemporal mapping of EV traffic flows to aggregate EV SoC level. The UL problem is subject to a lower level (LL) problem which employs a second-order cone program to describe PDN operating and endogenously form the distribution locational marginal prices, accounting for the spatiotemporal distribution of EV net charging demand. To analytically solve bi-level problem, the second-order cone constraints of the LL are linearized with polyhedral approximation, equivalent Karush–Kuhn–Tucker optimality conditions of the LL are then derived to convert the bi-level problem to a mathematical program with equilibrium constraints. Case studies on a small-scale and a large-scale test system with real-world data demonstrate the effectiveness of the proposed method in accurately characterizing the optimal spatiotemporal distribution of EV flows and charging demand in CPTN, while highlighting the benefits for EV users in terms of traveling cost reduction and V2G economic feasibility, and for CPTN in terms of peak demand reduction, absorption of renewable energy generation and network congestion management.