A new method to predict temperature distribution on a tube at constant heat flux

Abstract

Surface temperature distribution on a tube is one of the main factors affecting the calculation of the heat transfer coefficient calculation. When an electric heater heats the tube, a magnetic flux is generated that affects the thermocouples readings; therefore, an efficient fitting technique is needed to represent these readings. This work proposes an interpolated spline method to mathematically represent experimental data of a thermal distribution on a tube with heat flux. Linear regression was compared with a double linear interpolation process with an optimization algorithm and cubic spline curve method on the proposed problem. The results show that the interpolated experimental data can highly improve the regression of the spline curve. Consequently, an interpolated spline curve gives better surface temperature distribution and better estimation for the average temperature. The interpolated points on spline segments are chosen by an optimization algorithm, which is particle swarm optimization, in a way that provides more minor errors.

Multi objective optimization for house roof using artificial neural network model

“…The multi-layered perceptron of Artificial Neural Networks (ANN) can be likened to a multi-mapping black box analysis function (Gardner and Dorling, 1998). Artificial Neural Networks (ANN) have been effectively utilised in a variety of applications due to their flexible characteristics and capacity for self-learning, including predicting indoor temperature (Pandey et al, 2012), heat radiation modelling (Tausendschön and Radl, 2021), optimizing the energy efficiency of residential buildings (Gao, 2022), and predicting temperature distribution (Askar et al, 2021). This work, as all calculation procedures involve some assumptions, the accuracy of any method must ultimately be established by comparing the results of calculations with experimental results and simulations performed by Ansys.…”

Multi-objective optimization and simulation for multiple models of walls to estimate heat gain using Artificial Neural Network model

“…Artificial Neural Networks multi-layered perceptron can be compared to a multi-mapping black box analysis function (Gardner and Dorling, 1998). Due to their adaptability and capacity for self-learning, Artificial Neural Networks (ANN) have been successfully used in a wide range of applications, including predicting indoor temperature (Pandey et al, 2012), heat radiation modeling (Tausendschön and Radl, 2021), optimizing the energy efficiency of residential buildings (Gao, 2022), and predicting temperature distribution (Askar et al, 2021). For that reason, ANN needs data to build it so HAP software can estimate the heat gain through building components.…”