| Author: | Zhao, Zhonghao |
| Title: | Intelligent planning and operation for electric vehicle charging networks |
| Advisors: | Lee, K. M. Carman (ISE) |
| Degree: | Ph.D. |
| Year: | 2024 |
| Subject: | Electric vehicles -- Power supply Battery charging stations (Electric vehicles) City planning -- Design Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Industrial and Systems Engineering |
| Pages: | xviii, 245 pages : color illustrations |
| Language: | English |
| Abstract: | As the world increasingly embraces the imperative of transitioning to low-carbon transportation, electric vehicles (EVs) have emerged as a promising alternative to conventional internal combustion engine (ICE) vehicles and have received significant attention from researchers, policymakers, and the general public. The widespread adoption of EVs is an important strategy in mitigating climate change, reducing greenhouse gas (GHG) emissions, and achieving sustainable transportation. However, the successful transition towards a predominantly EV-driven transportation system depends heavily on the availability of a reliable and comprehensive EV charging network (EVCN). Moreover, the distinctive characteristics of EVs, such as limited driving range, long charging times, and range anxiety, impose challenges for the planning and operation of EVCNs. Thus, a well-designed and efficiently operated EVCN is crucial in bolstering the market share of EVs and further accelerating the transportation electrification process. This thesis investigates a series of research issues related to the planning and operation of EVCNs, including quality of service (QoS) evaluation, public charging station (PCS) location, pricing-based EV charging scheduling, and solution method development. Some practical factors, such as waiting time, range anxiety, cost control, non-linear charging profile, grid stability, are taken into consideration to better reflect the real charging environment. The studies included in this thesis will empower decision-makers with the necessary insights to develop intelligent solutions for EVCN planning and operation with the large-scale proliferation of EVs. In order to understand the needs of EV drivers, this study first investigates the QoS evaluation problem from the EV users’ standpoint. In Chapter 3, a performance metric that explicitly considers the uncertainties associated with charging demand is designed to assess the QoS for each PCS and the overall EVCN based on fuzzy queuing theory and extended universal generating function. The evaluation results can flexibly serve as the optimization objective or constraints for EVCN-related planning and operational problems. Subsequently, we focus on the planning stage of the EVCN. In Chapter 4, the emphasis lies in the budget-limited capacity planning problem of PCSs, taking into account the uncertainties in the charging process. The primary optimization objective is to maximize the QoS considering the waiting time in the queue and the blocking rate of the power grid, while ensuring that the total investment cost does not exceed the available budgetary limits. Chapter 5 investigates the optimal PCS deployment problem considering EV drivers’ range anxiety and the power grid stability, aiming to optimize the location and capacity of PCSs in the EVCN to ensure widespread accessibility and address the concerns of EV drivers and grid operators. By reformulating the deployment problem as a Markov decision process (MDP), a deep reinforcement learning (DRL)-based solution method is developed to improve the generalization capability and scalability. Finally, this study aims to address the EV charging scheduling problem for EVCNs in the operational stage. A dynamic pricing-based method is proposed in Chapter 7 to shift the charging demand from peak hours to off-peak hours so as to reduce the load at PCSs and flatten the demand over multiple time periods. Moreover, a hierarchical two-level charging scheduling scheme with the consideration of an online booking system and a pricing-based control system is developed in Chapter 7 to further improve the scheduling effect while ensuring EV users’ satisfaction. An online DRL-based approach is presented to intelligently find the optimal scheduling solution. In this thesis, a comprehensive set of numerical experiments and simulations is conducted. The results demonstrate that the proposed methods and solutions models are effective in addressing the EVCN planning and operation problems. The developed DRL-based methods outperform a wide range of baseline methods in terms of the solution quality and computational efficiency. The outcomes obtained from this thesis facilitate decision-makers in formulating effective EVCN planning and operational strategies, while also providing managerial insights and policy recommendations. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12969