Coordinating dynamic traffic-power systems under decentralized and centralized decision environments

Abstract

This paper proposes a dynamic traffic-electric power network model to investigate the interactions between the power distribution network (PDN) and the electric road network (ERN), whose operations are linked via the local marginal electricity price and the electric vehicles (EVs) charging demand. For the ERN, a novel formulation based on the link transmission model is proposed to: (1) accommodate the critical features of EVs and fast charging stations (FCSs), such as EVs with different driving ranges, initial states of charge of EVs, number of chargers and their charging power in a FCS; (2) explicitly model the charging process of EVs; (3) solve the optimal dynamic traffic assignment problem considering the mix of EVs and gasoline vehicles. For the economic operation of the PDN, an alternating current optimal power flow model is solved to minimize the electricity expenditure. Moreover, we propose mathematical algorithms to describe the interdependent and interactive schemas between the two networks by modeling the decentralized and centralized decision-making environments. The proposed modeling framework is capable of capturing the dynamic interactions that are not possible in classical traffic models. The illustrative traffic-power system shows that decentralized decision-making always results in losses of operational cost and renewable integration, compared to centralized decision-making; however, these losses can be greatly mitigated by having ERN and PDN operators share information about the planned EV charging demand and the projected locational marginal electricity price.