Document Type : Original Article

Authors

• Saeed Alimohammadi ¹
• Masoumeh Behrouz ²

¹ Associate Professor of Civil Engineering, Civil, Water and Environmental Engineering Faculty, Shahid Beheshti University, P.O.Box: 16765-1719, Zip Code: 1658953571, Bahar Blvd., Hakimieh, Tehran, Iran.

² Assistant Professor of Civil Engineering, Civil, Water and Environmental Engineering Faculty, Shahid Beheshti University, P.O.Box: 16765-1719, Zip Code: 1658953571, Bahar Blvd., Hakimieh, Tehran, Iran.

Abstract

Riverbank filtration systems offer a useful and reliable method for meeting domestic and industrial demands. In these systems, some wells are constructed in the bank of a river, where the water that flows across porous media into them has a very low pollutant level compared to the river. This study develops a cost-effective optimization model with the objective of minimizing total cost. A simulation model for analyzing these systems is developed as well. In these models, the analytic solutions of the groundwater flow equations and pollutant transport are used. Using the concept of response functions of linear systems, these solutions are generalized in the case of variable pumping. In common RBF systems, the unit pulse response function of drawdown is independent of wells location, and the transient flow equation reaches pseudo steady-state conditions. Two hypothetical example problems are presented, in the first, the design of a system is considered for meeting a given demand. The model solutions give the distance of wells’ alignment from river, distance between wells, and wells’ pumping rates. The model also outputs the pollutant concentration in the wells. The results of the steady optimization problem reveal that unexpectedly the central well’s discharge is greater than the side wells’ discharges. The resulting pumping, conveyance, and treatment costs showed that all these three cost terms are important. The sensitivity analysis revealed that all four considered parameters are sensitive with the sensitivity ranking of: T (transmissivity), λ (decay rate), ϴ (porosity), and Rd (retardation factor). In the second problem, an existing RBF system was analyzed by a simulation model and the variations of the well’s concentration were assessed by altering the four sensitive parameters. The proposed models are useful tools for primary design and analysis of RBF systems and assessing the effects of changing parameters in the system behavior.

Keywords

• Riverbank Filtration
• Response Function of Linear Systems
• Analytic Solution
• Optimization