Prediction of settled water turbidity and optimal coagulant dosage in drinking water treatment plant using a hybrid model of k-means clustering and adaptive neuro-fuzzy inference system

Abstract

Coagulation is an important process in drinking water treatment to attain acceptable treated water quality. However, the determination of coagulant dosage is still a challenging task for operators, because coagulation is nonlinear and complicated process. Feedback control to achieve the desired treated water quality is difficult due to lengthy process time. In this research, a hybrid of k-means clustering and adaptive neuro-fuzzy inference system (kmeans-ANFIS) is proposed for the settled water turbidity prediction and the optimal coagulant dosage determination using full-scale historical data. To build a well-adaptive model to different process states from influent water, raw water quality data are classified into four clusters according to its properties by a k-means clustering technique. The sub-models are developed individually on the basis of each clustered data set. Results reveal that the sub-models constructed by a hybrid k-means-ANFIS perform better than not only a single ANFIS model, but also seasonal models by artificial neural network (ANN). The finally completed model consisting of sub-models shows more accurate and consistent prediction ability than a single model of ANFIS and a single model of ANN based on all five evaluation indices. Therefore, the hybrid model of kmeans-ANFIS can be employed as a robust tool for managing both treated water quality and production costs simultaneously.

Suspended Sediment Modeling Using a Heuristic Regression Method Hybridized with Kmeans Clustering

“…K-means has been successfully used in recent literature to improve machine learning models' prediction accuracy due to its robust nature in estimation [40][41][42][43][44]. The application of hybrid the MARS-KMeans method is rarely found in the literature for prediction [45].…”

Optimisation and Prediction of the Coagulant Dose for the Elimination of Organic Micropollutants Based on Turbidity

“…22 Another study was performed by a hybrid of k-signifies an adaptive neuro-fuzzy inference system (k-means-ANFIS) for the turbidity of settled water prediction and optimal determination of the coagulant dose using historical data at large scale. 23 To construct a well adaptive model to different states of inflow water process, raw water quality data was classified into four groups according to its properties by a k-means clustering technique. The sub-models were developed individually based on each clustered data set.…”

Novel Ensemble Forecasting of Streamflow Using Locally Weighted Learning Algorithm

“…However, these models are mostly built upon the assumption that the process follows a normal distribution, however the streamflow process is generally non-linear and stochastic in nature [13]. Machine learning (ML) models, which have been widely used in recent decades to model many real-world problems [14][15][16][17][18][19][20], have the unique ability to identify the complex non-linear relationships between the predictors (inputs) and targets (outputs) without the need for the physical characterization of the system or the requirement of making any underlying assumptions. Many hybrid ensemble ML models with the integration of different data preprocessing techniques such as wavelet transformations, empirical mode decomposition, etc.…”