隨著全球對可永續交通轉型以及電動車（EV）普及化的關注增加，高效、實時且穩健的充電基礎設施的需求變得日益突出。本論文提出一種創新方法以應對日益增長的電動車需求與現有充電基礎設施之間的不均衡情況：即移動充電站（MCSs）的概念。此研究主要描繪了一種全新的在線演算法，以動態地配置這些移動充電站，目標在於顯著降低電動車主等待充電的時間。

本研究的核心是提出一種移動充電站的二階段配置與管理（2PM-MARL）框架，該框架應用多智能體強化學習進行動態的充電需求與供應平衡。我們詳細分析並證明了移動充電站配置問題的NP-hard性質，並將其與無容量設施位置問題（UFLP）相關聯，從而強調了計算的挑戰性，並凸顯出對像2PM-MARL框架這樣的智慧型實時解決方案的需求。

透過使用真實世界的充電數據進行廣泛的實驗與分析，我們驗證了2PM-MARL框架的效能。實驗結果顯示出在開始充電前的等待時間已經顯著降低，這顯示了2PM-MARL模型的實用性和效率。本論文為電動車充電基礎設施的管理提供了一個有效的實時解決方案，期望推進了電動車的普及化和永續城市交通系統的進步。

As global interest shifts toward sustainable transportation with the proliferation of electric vehicles (EVs), the critical demand for efficient, real-time, and robust charging infrastructure becomes increasingly pronounced. This thesis introduces a groundbreaking approach to address the imbalance between the surging EV demand and the existing charging infrastructure: the concept of Mobile Charging Stations (MCSs). The research outlines a novel online algorithm for the dynamic placement of these MCSs to significantly reduce waiting time for charging experienced by EV owners.

At the core of this research is the Mobile Charging Station Two-stage Placement and Management (2PM-MARL) framework, applying multi-agent reinforcement learning for a dynamic balancing of charging demand and supply. The complexity of the problem is elaborated by demonstrating the NP-hard nature of the MCS placement issue through a relation to the Uncapacitated Facility Location Problem (UFLP), underscoring the computational challenges and emphasizing the need for intelligent real-time solutions such as the proposed 2PM-MARL framework.

The proposed framework is validated through comprehensive experiments and analysis using real-world charging session data. The results exhibit significant reductions in the waiting time before charging starts, suggesting the potential practicality and efficiency of the 2PM-MARL model. By proposing an effective real-time solution, this research contributes substantially to the management of EV charging infrastructure, encouraging widespread EV adoption, and advancing sustainable urban transportation systems.

[1] R. Irle, “Global ev sales for 2022,” 2023, [Online]. Available: https://www.ev-volumes.com/.

[2] E. Parliament and the Council of the European Union, “Directive 2014/94/eu of the european parliament and of the council of 22 october 2014 on the deployment of alternative fuels infrastructure text with eea relevance,” Official Journal of the European Union, p. 23, 2014, https://eurlex.europa.eu/eli/dir/2014/94/oj.

[3] “Charging stations by state,” 2021, https://evadoption.com/ev-charging-stations-statistics/charging-stations-by-state/.

[4] A. Moradipari and M. Alizadeh, “Pricing and routing mechanisms for differentiated services in an electric vehicle public charging station network,” IEEE Transactions on Smart Grid, vol. 11, pp. 1489–1499, 2019.

[5] S. Hussain, Y.-S. Kim, S. Thakur, and J. G. Breslin, “Optimization of waiting time for electric vehicles using a fuzzy inference system,” IEEE Transactions on Intelligent Transportation Systems, vol. 23, pp. 15 396–15 407, 2022.

[6] L. P.-Y. Ting, P.-H. Wu, H.-Y. Chung, and K.-T. Chuang, “An incentive dispatch algorithm for utilization-perfect ev charging management,” in Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2022.

[7] N. Sadeghianpourhamami, J. Deleu, and C. Develder, “Definition and evaluation of model-free coordination of electrical vehicle charging with reinforcement learning,” IEEE Transactions on Smart Grid, vol. 11, pp. 203–214, 2018.

[8] Z. Sadreddini, S. Guner, and O. Erdin ̧c, “Design of a decision-based multicriteria reservation system for the ev parking lot,” IEEE Transactions on Transportation Electrification, vol. 7, pp. 2429–2438, 2021.

[9] K.-B. Lee, M. A. Ahmed, D.-K. Kang, and Y.-C. Kim, “Deep reinforcement learning based optimal route and charging station selection,” Energies, 2020.

[10] M. J. Eagon, S. Fakhimi, G. Lyu, A. Yang, B. Lin, and W. F. Northrop, “Model-based framework to optimize charger station deployment for battery electric vehicles,” 2022 IEEE Intelligent Vehicles Symposium (IV), pp. 1639–1648, 2022.

[11] L. von Wahl, N. Tempelmeier, A. Sao, and E. Demidova, “Reinforcement learning-based placement of charging stations in urban road networks,” Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022.

[12] Q. Liu, Y. Zeng, L. Chen, and X. Zheng, “Social-aware optimal electric vehicle charger deployment on road network,” Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, 2019.

[13] M. Campa ̃na, E. Inga, and J. C ́ardenas, “Optimal sizing of electric vehicle charging stations considering urban traffic flow for smart cities,” Energies, 2021.

[14] S. O. Bae, I. Jang, S. Gros, B. Kulcs ́ar, and J. Hellgren, “A game approach for charging station placement based on user preferences and crowdedness,” IEEE Transactions on Intelligent Transportation Systems, vol. 23, pp. 3654–3669, 2022.

[15] A. A. Shalaby, M. F. Shaaban, M. Mokhtar, H. H. Zeineldin, and E. F. El-Saadany, “A dynamic optimal battery swapping mechanism for electric vehicles using an lstm-based rolling horizon approach,” IEEE Transactions on Intelligent Transportation Systems, vol. 23, pp. 15 218–15 232, 2022.

[16] M. Aljaidi, N. Aslam, G. Samara, S. Almatarneh, K. E. Al-Qawasmi, and A. AlQammaz,“Ev charging station placement and sizing techniques: Survey, challenges and directions for future work,” 2022 International Arab Conference on Information Technology (ACIT), pp. 1–6, 2022.

[17] K. Lin, R. Zhao, Z. Xu, and J. Zhou, “Efficient large-scale fleet management via multi-agent deep reinforcement learning,” Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2018.

[18] B. Zheng, Q. Hu, L. Ming, J. Hu, L. Chen, K. Zheng, and C. S. Jensen, “Soup: Spatial-temporal demand forecasting and competitive supply in transportation,” IEEE Transactions on Knowledge and Data Engineering, vol. 35, pp. 2034–2047, 2023.

[19] X. Tang, F. Zhang, Z. Qin, Y. Wang, D. Shi, B. Song, Y. Tong, H. Zhu, and J. Ye, “Value function is all you need: A unified learning framework for ride hailing platforms,” Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, 2021.

[20] J. Sun, H. Jin, Z. Yang, L. Su, and X. Wang, “Optimizing long-term efficiency and fairness in ride-hailing via joint order dispatching and driver repositioning,” Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022.

[21] J. Zhong, Y. Qian, X. Wang, Y. Zhao, P. Wu, C. Chen, and M. Cai, “Mobile charging platform improves distribution system resilience and electric vehicles charging service,” 2023 IEEE 6th International Conference on Industrial Cyber-Physical Systems (ICPS), pp. 1–7, 2023.

[22] M. E. Kabir, I. Sorkhoh, B. Moussa, and C. M. Assi, “Routing and scheduling of mobile ev chargers for vehicle to vehicle (v2v) energy transfer,” 2020 IEEE Power & Energy Society General Meeting (PESGM), pp. 1–5, 2020.

[23] L. Yan, H. Shen, L. Kang, J. Zhao, and C. Xu, “Reinforcement learning based scheduling for cooperative ev-to-ev dynamic wireless charging,” 2020 IEEE 17th International Conference on Mobile Ad Hoc and Sensor Systems (MASS), pp. 401–409, 2020.

[24] S. Afshar and V. R. Disfani, “Optimal scheduling of electric vehicles in the presence of mobile charging stations,” 2022 IEEE Power & Energy Society General Meeting (PESGM), pp. 1–5, 2022.

[25] M. A. Beyazıt and A. Ta ̧scıkaraoˇglu, “Optimal management of mobile charging stations in urban areas in a distribution network,” 2022 International Conference on Smart Energy Systems and Technologies (SEST), pp. 1–6, 2022.

[26] L. Liu, Z. Xi, K. Zhu, R. Wang, and E. Hossain, “Mobile charging station placements in internet of electric vehicles: A federated learning approach,” IEEE Transactions on Intelligent Transportation Systems, vol. 23, pp. 24 561–24 577, 2022.

[27] A. K. Aktar, A. Ta ̧scıkaraoˇglu, and J. P. S. Catal ̃ao, “Optimal charging and discharging operation of mobile charging stations,” 2022 International Conference on Smart Energy Systems and Technologies (SEST), pp. 1–6, 2022.

[28] G. Cornu ́ejols, G. L. Nemhauser, and L. A. Wolsey, “The uncapacitated facility locationproblem,” 1990. [Online]. Available: https://api.semanticscholar.org/CorpusID:8880493

[29] P. Sunehag, G. Lever, A. Gruslys, W. M. Czarnecki, V. F. Zambaldi, M. Jaderberg,M. Lanctot, N. Sonnerat, J. Z. Leibo, K. Tuyls, and T. Graepel, “Value-decompositionnetworks for cooperative multi-agent learning,” ArXiv, vol. abs/1706.05296, 2017.[Online]. Available: https://api.semanticscholar.org/CorpusID:25026734