Time-varying optimal design for groundwater bioremediation: The pilot-scale study of a western Canadian site

This paper presents the development of an integrated simulation-clustering-optimization (ISCO) approach for supporting the groundwater bioremediation design. To investigate remediation performances, a subsurface model was employed to simulate contaminant transport. A mixed integer nonlinear optimization model was formulated to evaluate different remediation strategies. Multivariate relationships based on stepwise clustering analysis were developed to facilitate the incorporation of a simulation model within a nonlinear optimization framework. Based on the developed statistical relationships, a prediction system was constructed to provide all possible outputs for the proposed remediation design. By using the prediction system, the optimal strategy to achieve the remediation objective can be directly located. The proposed ISCO approach can improve bioremediation designs by providing adjustable schedules for operating wells from stage to stage, which will make the optimization more realistic. This approach was examined through the pilot-scale study of a western Canadian site. The optimization results demonstrated that 31% of the cost can be reduced by using a multi-stage design to achieve the same remediation objective as that of a single-stage design.