“…24,25 A number of models have been applied to predict the turbidity of the effluent [26][27][28][29][30][31] and the requisite coagulant dosage. 24,26,[32][33][34] A hybrid of k-means clustering and an adaptive neuro-fuzzy inference system was applied to predict the turbidity of water and the optimal coagulant dosage, and yielded values of R 2 > 0.8. 24 Andrews and Griffiths, 26 and Narges et al 32 developed a seasonal ANN model and an artificial fuzzy neural network based on subtractive clustering, respectively, to predict the optimal coagulant dosage needed to achieve the desired turbidity of the effluent in a sedimentation tank, with values of R 2 ranging from 0.78 to 0.89 and from 0.63 to 0.79, respectively.…”