Research on prediction model of optimal coagulant dosage in water purifying plant based on nerual network

Coagulant dosing process is an important part in water treatment plan, it directly affects the water quality and operating costs of production. It is very difficult to set up its mathematical model accurately basing on its reactive mechanism at present. Factors that affect the coagulation effect are analyzed in this paper, then a BP neural network prediction model of coagulant dosage is established. A improved BP algorithm - LM algorithm is used to train the neural network, it can improve the data's convergent speed. Experimental results show that the prediction accuracy of the BP neural network model is very high. The online predictive control of coagulant dosage can be made basing on this model, so it can optimize the coagulant dosage.