Intelligent based auto-coagulation control applied to a water treatment works

An intelligent based algorithm, in the form of a fuzzy neural network is applied to the control of coagulant dosing at a 400 Ml/day water treatment works. Coagulant dosing rate is related to raw water parameters such as colour, turbidity, temperature, etc. Process data is used to train the fuzzy neural network to predict coagulant dosing rate. The proposed scheme is evaluated in computer simulation studies using real process data before application to the real plant. A comparison is also made with a multiple regression modelling approach.