Sample first page

Download the file

OPTIMIZATION OF GROUNDWATER RESOURCES MANAGEMENT IN POLLUTED AQUIFERS

M. Mouti¹ and K. Ntrogkouli

Division of Hydraulics and Environmental Engineering,

Dept. of Civil Engineering, A.U.Th,

GR- 54124 Thessaloniki, Macedonia, Greece

E-mail:---@--------¹

ABSTRACT

Τhis paper focuses on optimal management of aquifers suffering from pollution problems. Genetic algorithms, outlined in Section 2, are used as the optimization tool. The chromosome evaluation function includes solution of the respective flow and mass transport problem. Infinite homogeneous aquifers, which allow for analytical solution of the groundwater flow problem, are taken into account. Then, advective pollutant transport is simulated by means of a moving point technique, based on local water velocities, which is discussed in Section 3. The respective computational load is rather small and allows for large values of the critical parameters of the genetic algorithm, namely population size and number of generations.

The approach is illustrated by means of two application examples, which are presented in Section 4. In the first we seek the maximum total clean water pumping rate Q_s from three production wells, situated in the same restricted area with two wells that inject polluted water. Finding best location of two of the production wells is part of the optimization process. The choice of the form of the penalty function, used to avoid constraint violation, is particularly discussed. Results show that Q_s depends heavily on the dimensions of the available area.

In the second application example, the total flow rate of the production wells is fixed and the minimum treatment cost of pumped water is sought. The form of the evaluation function, which represents treatment cost, is particularly discussed. Best solutions resulting from a number of runs are classified in three different patterns, which are comparatively evaluated in Section 5, using additional criteria. Finally, some thoughts on taking into account dispersive mass transport are discussed in Section 6. These include increase of pollutant deactivation time and use of an approximate analytical formula, of rather low computational volume.

Keywords

aquifer pollution; optimal management; genetic algorithms; moving points; penalty function