Combining Pumping Flowrate Maximization from Polluted Aquifers with Cost Minimization

Abstract

In this paper, we present a methodology for optimal management of aquifers affected by non-conservative pollutants, by means of application examples. Our multi-criteria approach takes into account: a) Maximization of safe total flowrate from a system of wells; and b) Minimization of pumping cost or of pipe network construction cost. As a preliminary step, we have studied maximization of the total safe well flowrate, without cost restrictions, in order to calculate an upper flowrate bound for all application examples. Then, we have solved two sets of minimization problems, using one of the aforementioned cost criteria in the respective objective function and different values of the required total safe well flowrate as a constraint in all of them. In this way, we have derived a set of Pareto optimal solutions for each problem. We have used the method of genetic algorithms as optimization technique and a moving point code for the simulation of advective pollutant transport. Our paper includes also: a) a discussion of relevant technical details, i.e., moving point arrival at production wells and structure of the penalty function, which is used in the genetic algorithm code; and b) a comparison of the results of the two sets of examples, which shows that the optimal well layouts are rather similar for large total safe flowrate values, while they are quite different for small ones.