Abstract

This study was developed using data from a drinking water treatment plant located at Boudouaou, Algeria, it is located at about 7 km from the Keddara dam which supplies potable water to Algiers capital of Algeria. The treatment consists essentially of preliminary disinfection, coagulation–flocculation, settling, filtration and final disinfection. Traditionally, optimum coagulant dosages are determined using jar tests. However, jar tests are relatively expensive and time consuming. In this study, we present a new Artificial Intelligence Techniques model called dynamic evolving neural-fuzzy inference system (DENFIS) based on an evolving clustering method, for modelling coagulant dosage rate used in the coagulation stage. Six online variables of raw water quality including turbidity, conductivity, temperature, apparent colour, ultraviolet absorbance, water pH and alum dosage were used to build the coagulant dosage model. Two DENFIS-based evolving neural-fuzzy inference system are presented and compared. The two DENFIS systems are: (1) Offline-based system named DENFIS-OF and (2) Online-based system, named DENFIS-ON. The performances of the models are evaluated using root-mean square errors (RMSE), mean absolute error and correlation coefficient (CC) statistics. The low RMSE and high CC values were obtained with DENFIS-ON method.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call