An agent-based modeling approach for public charging demand estimation and charging station location optimization at urban scale. (April 2023)

• Record Type: Journal Article
• Authors: Yi, Zhiyan; Chen, Bingkun; Liu, Xiaoyue Cathy; Wei, Ran; Chen, Jianli; Chen, Zhuo
• Is Part Of: Computers, environment and urban systems. Volume 101(2023)
• Publication Date: 2023-04
• Subjects: Electric vehicles -- Agent-based simulation -- Charging demand modeling -- Charging infrastructure -- Maximal coverage location problem; City planning -- Data processing -- Periodicals; Regional planning -- Data processing -- Periodicals; 303.4834
• DOI: 10.1016/j.compenvurbsys.2023.101949
• Languages: English
• ISSNs: 0198-9715

Abstract

Abstract: As the market penetration of electric vehicles (EVs) increases, the surge of charging demand could potentially overload the power grid and disrupt infrastructure planning. Hence, an efficient deployment strategy of electrical vehicle supply equipment (EVSE) is much needed. This study attempts to address the EVSE problem from a microscopic perspective by formulating the problem in two steps: public charging demand simulation and charging station location optimization. Specifically, we apply agent-based modeling approach to produce high-resolution daily driving profiles within an urban-scale context using MATSim. Subsequently, we perform EV assignment based on socioeconomic attributes to determine EV adopters. Energy consumption model and public charging rule are specified for generating synthetic public charging demand and such demand is validated against real-world public charging records to guarantee the robustness of simulation results. In the second step, we apply a location approach – capacitated maximal coverage location problem (CMCLP) model – to reallocate existing charging stations with the objective of maximizing the coverage of total charging demands generated from the previous step under the budget and load capacity constraints. The entire framework is capable of modeling the spatiotemporal distribution of public charging demand in a bottom-up fashion, and provide practical support for future public EVSE installation. Highlights: We applied agent-based modeling to produce high-resolution daily driving profiles. We utilized a location approach to reallocate existing charging stations. The synthetic charging demand is validated against real-world charging records. The results can provide practical guidance for future charging station deployment.