Diseño de un modelo de toma de decisiones para mantenimiento basado en el monitoreo del deterioro multiestado para un sistema de aerogeneración instalado en Costa Rica
Loría-García, Ana Laura
MetadatosMostrar el registro completo del ítem
Nowadays, maintenance is seen as a strategic management issue, after being considered for years just as a technical issue. This new role implies giving solutions to ever more complex problems and a more efficient asset management. In recent years, statistical and probabilistic models have been developed for evaluating system reliability based on the components’ reliability, the design and the assembly of them in the system. These models present a high level of application to the more exigent worldwide industries, for example the renewable energy industry. The increasing trend in renewable ways of generation has created new operational challenges, particularly in the wind industry. The uncertainty related to the wind is a very important topic to take in account for the successful integration of wind energy to the existing electric systems. The distinctive feature of the maintenance making decision model based on the multistate monitoring for a wind generation system installed in Costa Rica is the utilization of artificial neural networks (ANN), which are mathematical models inspired in the behavior of the human neurons. This project offers a wide literature review with key concepts for the definition of the model, wind turbines theory with essential characteristics of their components, the outlook for an eventual implementation in which the more representative wind generation national system is evaluated, a complete description of the selected model, the required configuration for the execution of this new maintenance strategy in a wind farm from Costa Rica and economic aspects regarding a possible implementation of this model.
Proyecto de Graduación (Licenciatura en Ingeniería en Mantenimiento Industrial) Instituto Tecnológico de Costa Rica, Escuela de Ingeniería en Mantenimiento Industrial, 2016.