Chlorcast©: a methodology for developing decision-making tools for chlorine disinfection control

Managers of drinking water supply systems are increasingly preoccupied with changes that may occur in the quality of water from the time it leaves the treatment plant until it reaches the consumerʹs tap. Modeling the quality of water in a distribution system and more particularly the evolution of residual chlorine may constitute an interesting means for more efficient management. This article presents the Chlorcast© methodology which provides the guidelines on building a decision-making tool, based on the use of artificial neural networks, for chlorination control in the final disinfecting phase. Two examples will demonstrate the ability of this method to represent the dynamics of the evolution of residual chlorine by making it possible to forecast residual chlorine in the City of Sainte-Foyʹs drinking water tank and distribution system.