Доступ предоставлен для: Guest
Портал Begell Электронная Бибилиотека e-Книги Журналы Справочники и Сборники статей Коллекции
Computational Thermal Sciences: An International Journal
ESCI SJR: 0.244 SNIP: 0.434 CiteScore™: 0.7

ISSN Печать: 1940-2503
ISSN Онлайн: 1940-2554

Computational Thermal Sciences: An International Journal

DOI: 10.1615/ComputThermalScien.2019025467
Forthcoming Article

EXPERIMENTAL MEASUREMENTS AND COMPUTATIONAL SIMULATION (CFD) APPLIED FOR THERMAL ANALYSIS OF AN UNINTERRUPTIBLE POWER SUPPLY

Ricardo José Pontes Lima
Federal University of Ceará
Lucas Luppi
Federal University of Ceará
Paulo Rocha
Federal University ofCeará
Maria Eugenia Vieira da Silva
Federal University of Ceará
Cícero Marcos Tavares Cruz
Federal University of Ceará
Stéphano Mendonça
Federal University of Ceará

Краткое описание

The temperature field generated by the components of electronic devices has great influence on their cooling system design. Computational Fluid Dynamics simulation may be used to predict the airflow velocities and temperature without building thermal prototypes. It can greatly reduce the cost of engineering physical models and data acquisition equipment. However, the power density of actual systems is pushing the limits of the electronics cooling technology. In this sense, it is expected high degree of accuracy of simulations applied to predict the airflow velocities and temperatures, therefore obtaining a shorter development cycle by reducing or even eliminating the need for building prototypes. This study aims to demonstrate the reliability of the temperature field simulation of an Uninterruptible Power Supply. The simulation focused on the transformer, because it dissipates the major heat flux, where approximately 80 W of power is generated on the laminated core. In order to ensure that the results are not mesh dependent, a study of the convergence was performed and presented. The findings demonstrate good agreement between measured and expected temperatures for the different component parts. The overall mean error between the simulation data and the experimental results is about 4%.