Procediments heurístics de disseny de sistemes d'electrificació rural amb energies renovables

Electrification hybrid systems (wind-photovoltaic systems) are a suitable option to supply electricity independently in isolated communities. To design these systems, there are recent mathematical models that provide the location and type of each of the electrification components and the design of t...

Descripción completa

Detalles Bibliográficos
Autor: Triadó Aymerich, Joan
Tipo de recurso: tesis doctoral
Fecha de publicación:2016
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:catalán
OAI Identifier:oai:upcommons.upc.edu:2117/107046
Acceso en línea:https://hdl.handle.net/2117/107046
https://dx.doi.org/10.5821/dissertation-2117-107046
Access Level:acceso abierto
Palabra clave:Àrees temàtiques de la UPC::Economia i organització d'empreses
Descripción
Sumario:Electrification hybrid systems (wind-photovoltaic systems) are a suitable option to supply electricity independently in isolated communities. To design these systems, there are recent mathematical models that provide the location and type of each of the electrification components and the design of the possible micro-distribution networks. When the amount of consumption points to electrify increases, solving the mathematical models require computational times that become infeasible in practice. For these cases, three heuristic methods based on mixed integer linear programming (MILP) are presented in this thesis: Relax and Fix heuristics, heuristics based on Corridor Method and Increasing Radius heuristics. In all procedures first a relaxed MILP is solved to obtain a base solution and then it is used as a starting point to find a feasible solution by searching in a more reduced search space. For each type of heuristics several options to relax and to reduce the solution space were developed and tested. Finally, extensive computational experiments based on real projects were carried out and results show that the best heuristic to apply varies according to the size of the instances to be solved.