Optimisation sous contrainte d'un générateur thermoélectrique pour la récupération de chaleur par différents algorithmes heuristiques

Authors: Allyson-Cyr, Mathieu
Advisor: Gosselin, Louis
Abstract: This study presents a model of a thermoelectric generator placed directly on the surface of a heat source. One unique feature of this model is that the heat source is subject to fixed heat flux and surface temperature that the system must respect. The main objective is to develop this model in this particular context with the possibility to be adapted to any heat source and the option to add a cooling system. The model has been developed entirely on the software Matlab. Then, a genetic algorithm is used to perform an optimisation in order to find the design with the maximal power output and minimal number of thermoelectric modules. With the cooling system included, the total surface of exchange and pumping power is also considered. A preliminary analysis is conducted to analyse the impact of the heat flux and surface temperature constraint on such system. Thereafter, a multi-objective optimisation is performed to find the optimal design considering multiple optimisation objectives. Finally, different heuristic algorithms are compared for solving the thermoelectric model proposed. The performance is discussed using different performance criteria to show the pros and cons of each heuristic algorithm when solving a complex optimisation design problem.
Document Type: Mémoire de maîtrise
Issue Date: 2019
Open Access Date: 6 March 2019
Permalink: http://hdl.handle.net/20.500.11794/33956
Grantor: Université Laval
Collection:Thèses et mémoires

Files in this item:
Description SizeFormat 
34867.pdf3.87 MBAdobe PDFThumbnail
All documents in CorpusUL are protected by Copyright Act of Canada.