Prediction of Boiling Heat Transfer Coefficients in Pool Boiling of Liquids Using Artificial Neural Network

This paper reports the prediction of pool boiling heat transfer coefficients using artificial neural network for three single components (distilled water, ethanol and cyclohexane) and a multicomponent system at atmospheric pressure from the literature. The predictability of the network was extremely good if the training data were chosen appropriately. In comparison of performance analysis of ANN, the relative error (RE) was studied and maximum error was found to be very low. The training was faster initially then it slowed down asymptotically. The prediction of ANN results was very close to the actual experimental values with a mean absolute relative error less than 1.5 %. The modified form of Newton-Raphson optimization technique was employed to minimize the error. For training the networks, the goal was fixed based on SSE and errors built in the updating the weight and biases. For input and hidden layers, tanh sigmoidal function and linear function for the output layer was taken.