Contribution of Neural Networks for Modeling Trihalomethanes Occurrence in Drinking Water

Abstract

The presence of chlorination by-products such as trihalomethanes (THMs) in drinking water has become an issue of particular concern for utility managers. Modeling THM occurrence in water may be a valuable tool for decision makers in dealing with these potentially hazardous by-products. This paper presents the application of artificial neural networks (ANNs) to model THM occurrence in drinking water. ANNs are compared with other modeling approaches, logistic regression and multivariate regression, to classify water utilities according to their susceptibility to generate high levels of THMs and to predict concentrations of formed THMs with variable water quality and chlorination conditions, respectively. In general, for both applications, ANN models gave similar or better results than other modeling techniques.