Traffic-Aware Grid Planning for Dynamic Wireless Electric Vehicle Charging
Dipanjan Ghose, S Sivaranjani, Junjie Qin
TL;DR
This paper addresses the challenge of planning power infrastructure to support Dynamic Wireless EV Charging (DWC) along roadways by coupling macroscopic traffic dynamics with grid optimization. It introduces a traffic-aware framework that uses Cellular Transmission Model (CTM) traffic dynamics to estimate spatiotemporal DWC energy demand and integrates this into a SOCP-relaxed AC-OPF for microgrid sizing and operation, coupled with solar and energy-storage (ES) siting. Case studies on a 14-mile I-210W corridor show that traffic-aware planning significantly reduces capital and operating costs compared with worst-case planning, while maintaining reliability under diverse traffic conditions. The framework demonstrates practical benefits for deploying DWC infrastructure and highlights opportunities for extending the approach to demand response and market participation in grid services.
Abstract
Dynamic Wireless Electric Vehicle Charging (DWC) on electrified roadways is an emerging technology that can significantly reduce battery sizes, eliminate charging downtime, and alleviate range anxiety, specially for long-haul transportation and fleet operations of electric vehicles (EVs). However, these systems introduce new challenges for power system planning due to their short-duration and high-power demands which can strain the grid if not properly managed. As the energy demands from DWC depend on vehicle speed, density, dwell time in charging zones, and load profiles along road segments, there is a need for integrated planning of such systems, jointly considering both traffic behavior and EV energy consumption. In this paper, we propose a traffic-aware grid planning framework for DWC. We leverage a macroscopic Cell Transmission Model of traffic flow to estimate real-time, spatiotemporal EV charging demand from DWC corridors. The demand model is then integrated into an AC Optimal Power Flow based formulation to optimally size a microgrid that supports DWC under varying traffic conditions while minimizing the cost of operation. Our framework explicitly models how spatiotemporal traffic patterns affect the utilization of grid resources to obtain system designs that achieve lower costs and are easier to operationalize as compared to planning models that rely on worst-case traffic data. We demonstrate the framework on data from a 14-mile segment of the I-210W highway in California, USA, evaluating multiple traffic scenarios like free-flow, severe congestion, accidents of varying severity, and natural disasters like forest fires. Our results demonstrate that traffic-aware grid planning significantly reduces infrastructure costs as compared to worst-scenario based modeling, while ensuring reliability of service in terms of meeting charging demands under diverse traffic conditions.