Performance prediction of a cooling tower using artificial neural network

Abstract This paper describes an application of artificial neural networks (ANNs) to predict the performance of a cooling tower under a broad range of operating conditions. In order to gather data for training and testing the proposed ANN model, an experimental counter flow cooling tower was operated at steady state conditions while varying the dry bulb temperature and relative humidity of the air entering the tower and the temperature of the incoming hot water along with the flow rates of the air and water streams. Utilizing some of the experimental data for training, an ANN model based on a standard back propagation algorithm was developed. The model was used for predicting various performance parameters of the system, namely the heat rejection rate at the tower, the rate of water evaporated into the air stream, the temperature of the outgoing water stream and the dry bulb temperature and relative humidity of the outgoing air stream. The performances of the ANN predictions were tested using experimental data not employed in the training process. The predictions usually agreed well with the experimental values with correlation coefficients in the range of 0.975–0.994, mean relative errors in the range of 0.89–4.64% and very low root mean square errors. Furthermore, the ANN yielded agreeable results when it was used for predicting the system performance outside the range of the experiments. The results show that the ANN approach can be applied successfully and can provide high accuracy and reliability for predicting the performance of cooling towers.