Eventos Anais de eventos
ENCIT 2018
Brazilian Congress of Thermal Sciences and Engineering
Fault Detection and Diagnosis in a Refrigeration System Using Thermoeconomic Methodology and Artificial Intelligence
Submission Author:
Tiago Mendes , MG , Brazil
Co-Authors:
Tiago Mendes, Euler Horta, Osvaldo Jose Venturini, Marcelo Pirani
Presenter: Tiago Mendes
doi://10.26678/ABCM.ENCIT2018.CIT18-0100
Abstract
The refrigeration systems are great consumers of electrical power in the commercial and industrial sector. Nowadays fault detection and diagnosis techniques, mathematical models and computer modelling have been used to improve the performance and predict behavior of refrigeration systems. Considering this scenario, the aim of this work is to apply a thermoeconomic methodology (the Fuel-Impact Formula, Malfunction and Dysfunction) for diagnosis. In addition, an artificial intelligence classifier was used to fault detection in the individual components based on measuring quantities of the system. It was analyzed the influence of each component malfunction in efficiency and power consumption of the system. Through this analysis, is possible to obtain a prognosis of the system behavior, as well as to perform a complete analysis of the whole refrigeration system and of each component.
Keywords
Thermoeconomic methodology, Refrigeration system, Computer modelling, Artificial Intelligence
