Described is the development of a full-scale artificial neural network (ANN) model for the removal of natural organic matter (NOM) by enhanced coagulation at the Rossdale Water Treatment Plant (WTP) in Edmonton, Alberta, Canada. Few attempts have been made to develop a full-scale model of the enhanced coagulation process due to extreme variability in the process parameters and the complex nonlinear relationships between them. When applied to previously unseen data, the model predicted effluent colour with a high degree of accuracy. The model will be incorporated into real-time process control at the WTP following a period of online testing.
- © IWA Publishing 1999