Radial basis Neural Network Model to Prediction of Thermal Resistance and Heat Transfer Coefficient of Oscillating Heat Pipe Using Graphene and Acetone-Based Nanofluids

The primary objective of this study was to assess the operational efficiency of an oscillating heat pipe featuring an inner diameter of 1.7 mm and an outer diameter of 3 mm. This OHP was filled with an acetone -based fluid infused with graphene nanoparticles. The research aimed to analyze the effects of altering filler ratio and heat inputs on temperature difference, heat transfer coefficient, and thermal resistance in an oscillating heat pipe, with a specific focus on filler ratios ranging from 50% to 80% and heat inputs between 20W and 40W. The results reveal that there is a maximum in heat transfer coefficient of 220.48W/m 2 degrees C and 224.1 W/m 2 degrees C for acetone and graphene respectively. There is a decrease in thermal resistance 0f 1.441 degrees C/W and 1.421 degrees C/W for acetone and graphene for optimal combinations (40W with 80% filler ratio). Finally, the experimental data of 1800 data sets were used to develop the Artificial Neural network model using Radial basis function by considering three input parameters viz, fill ratio (50% to 80%), heat load (25W to 40W) and time with an output of temperature difference, heat transfer coefficient and thermal resistance. The developed ANN using the radial basis function (RBF) was able to predict the experimental parameters of temperature difference, heat transfer coefficient and thermal resistance with 97.70% and 97.12% accuracy for graphene and acetone respectively. Based on the obtained results MSE values for graphene and acetone are 1.015 and 1.064 respectively. (c) 2024 Jordan Journal of Mechanical and Industrial Engineering. All rights reserved