Development of an Artificial Neural Network Model for the Prediction of the Performance of a Silica-Gel Desiccant Wheel

Abstract

This work presents mathematical equations derived from Artificial Neural Networks (ANNs) for the estimation of dry bulb temperature and specific humidity at the outlet of a desiccant wheel to predict useful data for designers and engineers. The neural network model comprises five inputs and two output neurons that define the outlet conditions (dry bulb temperature and specific humidity) of a desiccant wheel. The results obtained by the ANN model are compared with the actual data by using input variables. The results show that the mean absolute percentage errors for dry bulb temperature and specific humidity are found to be 0.80% and 1.56% respectively; and the correlation coefficient (R) values obtained are approximately 0.986 for both output variables. The root mean square errors, which is another significant point in this study, are found to be 0.54% and 0.18% for dry bulb temperature and specific humidity respectively.