Use of Artificial Neural Networks for Modelling Chlorine Residuals in Water Distribution Systems

Drinking water contaminated by microorganisms can be a major risk to public health. Disinfection is used to destroy microorganisms that are potentially dangerous to humans. In order to prevent bacterial regrowth, it is also desirable to maintain a disinfectant residual in the water distribution system. The most commonly used disinfectant is chlorine. If the dosing rate of chlorine is too low, there may be insufficient residual at the end of the distribution system, resulting in bacterial regrowth. On the other hand, the addition of too much chlorine can lead to customer complaints about taste and odour, corrosion of the pipe network and the formation of potentially carcinogenic by-products. Consequently, in order to determine the optimal chlorine dosing rate, it is necessary to be able to predict chlorine decay in the network. In this paper, two data-driven techniques, namely linear regression models and multi layer perceptron artificial neural networks, are used to predict chlorine concentrations at two key locations in the Hope Valley water distribution system, which is located to the north of Adelaide, South Australia. A 5-year data set containing routinely measured parameters is used for model development and validation. The results obtained indicate that both techniques are relatively successful in predicting chlorine concentrations in the distribution system. This is despite the fact that there is no hydraulic model of the system and that only data that were collected on a routine basis are used for model development. Overall, the performance of the multi layer perceptron is slightly better than that of the regression model, suggesting the presence of some non-linearities in the underlying physical processes governing chlorine decay.