Modeling the outlet temperature in heat exchangers: Case study

• Main Authors: Barbulescu Alina, Barbes Lucica
• Format: Article
• Language: English
• Published: VINCA Institute of Nuclear Sciences 2021-01-01
• Series: Thermal Science
• Subjects: heat transfer; neural networks; reynolds number; outlet temperature
• Online Access: http://www.doiserbia.nb.rs/img/doi/0354-9836/2021/0354-98361900449B.pdf

This article presents the results of the study of the heat transfer in a heat ex-changer where the working fluid is the crude oil prepared for desalination, and the thermic agent is the re-circulating heavy gasoline fraction. Firstly, the Reynolds numbers have been computed using the temperatures and flow rates of the fluids as input variables. Then, general regression neural network and multi-layer perceptron were used for the outlet temperatures estimation using the inlet temperatures and the Reynolds numbers as input variables. The best models on the training dataset were obtained utilizing a multilayer perceptron with one hidden layer, while the best performance on the validation dataset was obtained using a multilayer perceptron network with two hidden layers.