Optimization of conventional water treatment plant using dynamic programming

Abstract

In this research, the mathematical models, indicating the capability of various units, such as rapid mixing, coagulation and flocculation, sedimentation, and the rapid sand filtration are used. Moreover, cost functions were used for the formulation of conventional water and wastewater treatment plant by applying Clark's formula (Clark, 1982). Also, by applying dynamic programming algorithm, it is easy to design a conventional treatment system with minimal cost. The application of the model for a case reduced the annual cost. This reduction was approximately in the range of 4.5-9.5% considering variable limitations. Sensitivity analysis and prediction of system's feedbacks were performed for different alterations in proportion from parameters optimized amounts. The results indicated (1) that the objective function is more sensitive to design flow rate (Q), (2) the variations in the alum dosage (A), and (3) the sand filter head loss (H). Increasing the inflow by 20%, the total annual cost would increase to about 12.6%, while 20% reduction in inflow leads to 15.2% decrease in the total annual cost. Similarly, 20% increase in alum dosage causes 7.1% increase in the total annual cost, while 20% decrease results in 7.9% decrease in the total annual cost. Furthermore, the pressure decrease causes 2.95 and 3.39% increase and decrease in total annual cost of treatment plants.

<b>Intelligent system for improving dosage control

“…For example, the use of mathematical modeling in water treatment process foregrounds the study of variations in costs and inputs at various stages, including clotting, according to operating conditions (Malzer & Strugholtz, 2008;Mostafa, Bahareh, Elahe, & Pegah, 2013;Ogwueleka & Ogwueleka, 2009 Intelligent methods, such as fuzzy systems, artificial neural networks and expert systems, are currently employed to overcome the complexity of water treatment processes (Dharman, Chandramouli, & Lingireddy, 2012;Wu & Lo, 2010;Zhang, Achari, Li, Zargar, & Saqid, 2013).…”