Researchers from the University of Melbourne, including Yongqi Zhang, Dong Ngoduy, Li Duan, Mingchang Zhu, and Zhuo Chen, have developed a new approach to managing electric vehicle (EV) charging demand that could help charging station operators (CSOs) better handle the complexities of EV charging. The study was published in the journal “Transportation Research Part C: Emerging Technologies.”
The rapid increase in electric vehicles is creating significant challenges for charging station operators. Traditional methods of managing demand often oversimplify user behavior and can’t scale effectively. The researchers propose a new framework that uses a bi-level Stackelberg game to optimize pricing and manage demand more effectively. This approach considers the complex and varied behavior of EV users, including factors like price sensitivity, battery aging, risk attitudes, and network travel costs. Unlike previous models, it avoids network equilibrium constraints, making it more scalable and using queuing theory to represent congestion effects.
To solve the large-scale optimization problem, the researchers used a rolling-horizon approach that combines the Dynamic Probabilistic Sensitivity Analysis-guided Cross-Entropy Method (PSA-CEM) with the Method of Successive Averages (MSA). They validated their framework with a real-world case study in Clayton, Melbourne, using data from 22 charging stations. The results showed that their mechanism significantly reduced queuing penalties and improved user utility compared to fixed and time-of-use pricing.
This research provides a robust and scalable tool for strategic EV charging management. It offers a practical solution for charging station operators to balance realism with computational efficiency, ultimately helping to optimize the EV charging process and improve user satisfaction. The study was published in the journal “Transportation Research Part C: Emerging Technologies.”
This article is based on research available at arXiv.