OPTIMIZATION OF INTERMITTENT PUMPING SCHEDULES FOR AQUIFER REMEDIATION USING A GENETIC ALGORITHM

• Published in: Journal of the American Water Resources Association 2000-12, Vol.36 (6), p.1335-1348
• Main Authors: Liu, Wei-Han, Medina Jr, Miguel A., Thomann, Wayne, Piver, Warren T., Jacobs, Timothy L.
• Format: Article
• Language: English
• Subjects: contaminant transport models; Earth sciences; Earth, ocean, space; Engineering and environment geology. Geothermics; Exact sciences and technology; genetic algorithm; Hydrogeology; Hydrology. Hydrogeology; optimal pumping schedules; Pollution, environment geology; stochastic ground water; USA, North Carolina
• ISSN: 1093-474X; 1752-1688
• DOI: 10.1111/j.1752-1688.2000.tb05730.x

An optimization model was developed to evaluate the pump-and-treat option for the aquifer below the Duke Forest Gate 11 waste site in Durham, NC. The multiple-objective optimization model, based on genetic algorithms, minimizes total pumping costs and reduces contaminant concentrations in the proximity of the pumping wells to concentrations below a prescribed health standard. The model includes solutions to stochastic-flow and mass-transport models, the toxicity standard for the contaminant, and a cost function for installing and operating the extraction pumps. Genetic algorithms were used to solve the discontinuous, chance-constrained, multiple-objective-function optimization model to determine the optimal pumping schedules. The groundwater and contaminant transport models are provided, and model results are presented to show the intermittent pumping schedule optimization and the optimal pumping schedules for remediation of the waste site. The results illustrate that a genetic algorithm provided a more robust search strategy than other heuristic approaches.