Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/178D6E1D-968F-4411-A943-E438CC64E26C | ||
| Year | 2024 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Maria Mamantaki, "Design of an electric vehicle smart charging system", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2024 https://doi.org/10.26233/heallink.tuc.100597 | ||
| Appears in Collections |
Real-time smart charging and control of plug-in electric vehicles (PEVs) are vital for optimizing their integration into the power grid. This technology facilitates dynamic adjustments of charging power in response to grid conditions, energy prices, and the availability of renewable energy sources. By effectively managing demand and supply in real-time, smart charging mitigates grid overloads, reduces peak demand, and enhances the utilization of renewable energy, thereby fostering a more sustainable and efficient energy ecosystem. Furthermore, it enables vehicle-to-grid (V2G) capabilities, allowing PEVs to supply power back to the grid during peak periods, which further stabilizes the energy network. This approach not only improves the reliability and resilience of the grid but also provides economic advantages to both consumers and utilities by lowering energy costs and deferring the need for additional infrastructure investments.In this work, a real-time smart charging method for electric vehicles is developed with minimal need for the forecast of significant quantities. For this purpose, expert systems were used, namely a fuzzy logic system with inputs the flexibility of the electric vehicle to adjust its power and the electricity price; and output the charging active power of the electric vehicle. The optimal parameters of the fuzzy logic system, such as the centers and the ranges of the membership functions, are obtained using the Particle Swarm Optimization (PSO) algorithm. Data obtained from smart electric vehicle charging methods using classical optimization techniques, in particular using Matlab's fmincon function, were used to train the fuzzy logic system. Several simulation scenarios were carried out and since no knowledge of forecast features is required, the results were satisfactory and thus the training of the proposed fuzzy logic system for real-time smart charging of electric vehicles was successful. The fact that the proposed method is independent of the forecasting of variable quantities like electricity price enhances its applicability to real-world systems.
Copyright © DIAS 2013
