Development of adaptive neuro-fuzzy inference system model for predict trihalomethane formation potential in distribution network simulation test

Predictive capability of THM models for drinking water treatment and distribution

“…The ML based approaches show promise by demonstrating lower error compared to their multiple linear regression based model counterparts. 51,55,56 Additionally, Zhang et al , 2023 demonstrated that conducting a stepwise multiple linear regression for selection of significant input variables prior to ML model training allowed for more efficient training and implementation of ML model. 51 Efficient implementation of ML models is particularly useful for real-time prediction of THM.…”

“…There have been a growing amount of research exploring the use of ML based techniques for predicting THMs. [51][52][53][54][55][56][57] Most of the research uses some type of artificial neural network (ANN) based model to develop non-linear relationships between water quality variables and THM concentration. The ML based approaches show promise by demonstrating lower error compared to their multiple linear regression based model counterparts.…”