Analyzing real-world driving and charging data from a large-scale electric vehicle (EV) fleet is crucial for supporting decision-making, particularly in deploying charging infrastructure and developing EV-focused policies. Today, EVs come equipped with various data collection devices, capable of capturing real-time driving and charging information, thus providing a valuable data source for further analysis and applications. To this end, our study analyzes an 11-month usage data from a large sample of EVs, including 3,777 battery EVs (BEVs) and 5,973 plug-in hybrid EVs (PHEVs) in Shanghai, China. Based on their daily travel intensity, the EVs are clustered into two groups used primarily for private and commercial purposes. Comprehensive analyses are then conducted for each group of EVs.

Nevertheless, the sensitive nature of these data, driven by privacy concerns, poses substantial challenges when it comes to sharing it across various institutions. In response, we develop deep generative models to generate realistic and diverse EV usage data that mirrors the distribution of the real-world dataset, enabling more effective and privacy-compliant utilization of EV data for various purposes, including education and research. Based on the generated data, we develop a modeling framework to quantify the potential of controlling EV charging behavior to reduce the carbon emissions in Shanghai.

Because of the high complexity of the EV users’ charging behavior and traffic flow dynamics, it is complicated to model the electrified urban transportation system accurately. With this background, we develop a simulation platform for investigating the EV users’ behavior and the resulting aggregate system performance. We enforce the platform in Shanghai, with the map extracted from Open Street Map and the OD information from the data center in Shanghai. We can serve it as a test bed for further investigation.

• Li, Z., Xu, Z., Chen, Z.*, Chen, G., Xie, C., and Zhong, M., 2023. An Empirical Analysis of Electric Vehicles’ Charging Patterns. Transportation Research Part D, 117, 103651.
• Li, Z., Bian, Z., Chen, Z.*, Ozbay, K., & Zhong, M. 2024. Synthesis of Electric Vehicle Charging Data: A Real-World Data-Driven Approach. Communications in Transportation Research, 4, 100128.