A general corridor model for designing plug-in electric vehicle charging infrastructure to support intercity travel. (July 2016)

• Record Type: Journal Article
• Authors: Ghamami, Mehrnaz; Zockaie, Ali; Nie, Yu (Marco)
• Is Part Of: Transportation research. Volume 68(2016)
• Page Start: 389
• Page End: 402
• Publication Date: 2016-07
• Subjects: Plug-in electric vehicle -- Corridor model -- Intercity travel -- Metaheuristic -- Simulated annealing; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2016.04.016
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: We propose to optimally configure plug-in electric vehicle (PEV) charging infrastructure using a general corridor model. The model allows realistic travel demand patterns and considers charging delay induced by congestion at charging stations. We formulate the model as a mixed integer program with nonlinear constraint and solved it with a specialized metaheuristic. Numerical experiments show that the metaheuristic produces satisfactory solutions, and is efficient for larger problems. Abstract: This paper proposes to optimally configure plug-in electric vehicle (PEV) charging infrastructure for supporting long-distance intercity travel using a general corridor model that aims to minimize a total system cost inclusive of infrastructure investment, battery cost and user cost. Compared to the previous work, the proposed model not only allows realistic patterns of origin–destination demands, but also considers flow-dependent charging delay induced by congestion at charging stations. With these extensions, the model is better suited to performing a sketchy design of charging infrastructure along highway corridors. The proposed model is formulated as a mixed integer program with nonlinear constraints and solved by a specialized metaheuristic algorithm based on Simulated Annealing. Our numerical experiments show that the metaheuristic produces satisfactory solutions in comparison with benchmark solutions obtained by a mainstream commercial solver, but is more computationally tractable for larger problems. Noteworthy findings from numerical results are: (1) ignoring queuing delay inducted by charging congestion could lead to suboptimal configuration of charging infrastructure, and its effect is expected to be more significant when the market share of PEVs rises; (2) in the absence of the battery cost, it is important to consider the trade-off between the costs of charging delay and the infrastructure; and (3) building long-range PEVs with the current generation of battery technology may not be cost effective from the societal point of view.