Towards Local Consumption of Renewable Energy Using Electric Vehicles
Cauz, Marine
Promotor(s) : Ernst, Damien
Date of defense : 27-Jun-2016/28-Jun-2016 • Permalink : http://hdl.handle.net/2268.2/1553
Details
Title : | Towards Local Consumption of Renewable Energy Using Electric Vehicles |
Translated title : | [fr] Vers une consommation locale d'énergie renouvelable au moyen de voitures électriques |
Author : | Cauz, Marine |
Date of defense : | 27-Jun-2016/28-Jun-2016 |
Advisor(s) : | Ernst, Damien |
Committee's member(s) : | Van Cutsem, Thierry
Louveaux, Quentin Wehenkel, Louis |
Language : | English |
Number of pages : | 90 |
Discipline(s) : | Engineering, computing & technology > Electrical & electronics engineering |
Institution(s) : | Université de Liège, Liège, Belgique |
Degree: | Master en ingénieur civil électricien, à finalité approfondie |
Faculty: | Master thesis of the Faculté des Sciences appliquées |
Abstract
[en] This master thesis is dedicated to the study of strategies promoting the local consumption of electricity produced from renewable resources using electric vehicles (EVs). In this context, EVs are considered as mobile and flexible batteries offering the opportunity to reduce the curtailment of electricity produced from renewable resources. Given the knowledge of charging stations and EVs starting point and destinations, minimizing the curtailment of electricity can be formalized into a dispatching problem, i.e. finding an allocation strategy between EVs and charging stations. We propose two approaches to tackle this problem: the first approach, qualified as the priority based strategy, is a strategy based on a set of priority rules, exploiting some predetermined principles based on human behaviour. This first approach also serves as a reference strategy. The second approach is built upon genetic optimization, where dispatching strategies are encoded using genes, and the search towards (local) optima is done using genetic operators. In addition to the dispatching strategies, a simulation platform has been developed, showing close-to-reality characteristics, with geographical parameters corresponding to Belgium. The master thesis proposes an incremental presentation of this platform in terms of complexity augmenting features and realistic assumptions from one chapter to the other. Our two approaches are finally compared using the simulation platform. Simulation results show that genetic optimization performs better than the priority rules-based approach, but this comes with a computational cost that may grow significantly with the number of EVs and charging stations. We finally propose a discussion about the enhancement of the genetic programming approach, and propose several research directions based on the incorporation of prior knowledge into the genetic optimization process.
File(s)
Document(s)
Cite this master thesis
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.