Genetic Algorithms for Optimizing the Remediation of Contaminated Aquifer

Produced water constitutes a large amount of waste fluids during the production operation of an oil field. Underground injection for disposing the wastewater from hydrocarbon production is an engineering problem due to the possibility of leakage of injected pollutant material from receiving medium to a drinking water source. This paper describes a method for optimization of polluted aquifer remediation design using one of the artificial intelligence optimization methods, namely Genetic Algorithms (GAs). As a case study, the contaminated area was created by using a groundwater transport simulator, which is based on Method of Characteristics (MOC). Then, the developed computer program was run to find the optimum solution for remediation, and the solution yielded from the program was verified by using a groundwater simulator. The plume was captured and the concentration level of chloride ion within the aquifer was diminished by using extraction wells. The analytical model approach provided different alternatives for appropriate isolation of plume. GAs were used as an optimization technique for making a decision among the alternatives, by considering operation time, number of wells, pumping rate and drawdown as decision variables and constraints.