Modeling Electric Vehicle Charging Load Dynamics: A Spatial-Temporal Approach Integrating Trip Chains and Dynamic User Equilibrium

Abstract

The spatial-temporal distribution (STD) of electric vehicle (EV) charging load is significantly influenced by EV users’ route choice and charging behaviors. In a congested traffic network that accommodates gasoline vehicles and EVs, users’ route choice mutually influences each other, resulting in a dynamic user equilibrium (DUE). Due to the limited battery capacity and insufficient charging facilities, fast charging during trips and slow charging at trip destinations may be co-existed for EV users to complete their daily trip-chain demand. This paper introduces a trip-chain-based DUE (TC-DUE) model to comprehensively analyze EV users’ route choice, fast and slow charging behaviors throughout their entire trip chains, and to derive the STD of EV fast and slow charging load. In the TC-DUE model, each user makes route choice for each trip in the trip chain to maximize his trip chain utility. Route choice serves as the foundation for employing a traffic-charge microsimulation (TCM) to simulate each user’s driving, parking, fast, and slow charging behaviors. The STD of fast and slow charging load, and the assessment of traffic congestion are derived by the TCM. An iterative solution procedure via route choice and TCM is adopted for the TC-DUE model. The iterations stop until a TC-DUE is reached where no user can improve his trip chain utility by charging his route choice unilaterally. Finally, we apply the TC-DUE model to the road network of Berlin, Germany, comprising 11,345 nodes and 24,335 links. The results indicate that the faster routes and earlier charging behaviors are attained when considering the traffic congestion caused by the participation of other users, which leads to a more balanced spatial distribution of fast charging load, an earlier peak hour and a higher peak value for both fast and slow charging load.