Reinforcement Learning for Multi-EV Charging: A Multi-Objective Approach to Renewable Integration and Vehicle-to-Grid
| dc.rights.license | CC-BY-NC-ND | |
| dc.contributor.advisor | Mitici, M.A. | |
| dc.contributor.author | Vakili, Aran | |
| dc.date.accessioned | 2025-09-04T23:01:53Z | |
| dc.date.available | 2025-09-04T23:01:53Z | |
| dc.date.issued | 2025 | |
| dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/50348 | |
| dc.description.sponsorship | Utrecht University | |
| dc.language.iso | EN | |
| dc.subject | This thesis develops a reinforcement learning framework to coordinate multi-EV charging in community energy systems with solar generation. It integrates vehicle-to-grid, battery degradation, and fairness objectives. Using algorithms like A2C, PPO, and DDQN, it shows that actor–critic methods with curriculum and Bayesian optimization achieve superior coordination, while explainability techniques provide insights into policy decisions. | |
| dc.title | Reinforcement Learning for Multi-EV Charging: A Multi-Objective Approach to Renewable Integration and Vehicle-to-Grid | |
| dc.type.content | Master Thesis | |
| dc.rights.accessrights | Open Access | |
| dc.subject.courseuu | Computing Science | |
| dc.thesis.id | 53681 |
