Prediction of heat transfer due to presence of copper–water nanofluid using resilient-propagation neural network

Heat transfer due to laminar natural convection of copper–water nanofluid in a differentially heated square cavity has been predicted by Artificial Neural Network (ANN). The nanofluid has been considered as non-Newtonian. The ANN has been trained by a resilient-propagation (RPROP) algorithm. The required input and output data to train the ANN has been taken from the results of numerical simulation that was performed simultaneously where the transport equations has been solved numerically using finite volume approach incorporating SIMPLER algorithm. Results from simulation and resilient-propagation (RPROP) based ANN have been compared. It has been observed that the ANN predicts the heat transfer correctly within the given range of training data. It is further observed that resilient-propagation (RPROP) based ANN is an efficient tool to predict the heat transfer than simulation, which takes much longer time to compute.