HR: 1340h
AN: H13D-1357    [Abstracts]
TI: Bridging the Gap - Interactive Inverse Groundwater Modeling
AU: * Singh, A
EM: asingh8@uiuc.edu
AF: University of Illinois, Urbana-Champaign Department of Civil and Environmental Engineering, 4129, NCEL-MC250, 205 N. Mathews Avenue, Urbana, IL 61801 United States
AU: Minsker, B
EM: minsker@uiuc.edu
AF: University of Illinois, Urbana-Champaign Department of Civil and Environmental Engineering, 3230 NCEL-MC250, 205 N. Mathews Avenue, Urbana, IL 61801 United States
AB: This paper presents a novel approach for solving the inverse problem of estimating heterogeneous aquifer parameters for a groundwater flow model, using interactive multi-objective evolutionary optimization. A hypothetical aquifer, for which the `true' parameter values (in this case hydraulic conductivity) are known, is used as a test case to demonstrate the usefulness of this method. It is shown that using automated calibration techniques without using expert interaction leads to parameter values that are not consistent with site knowledge. In such cases, it is desirable to incorporate expert knowledge in the inversion process to generate more reasonable estimates. An interactive approach is proposed within a multi-objective framework that allows the user to evaluate trade-offs between the expert knowledge and other measures of numerical errors. Using Pilot points and geostatistical parameters as decision variables, numerical optimization is combined with expert knowledge leading to conductivity fields that respect both the observation data and site knowledge that the expert may have. Early results indicate that this approach leads to parameter estimates that are much more consistent with site knowledge. A major issue with interactive approaches, however, is `human fatigue' in evaluating numerous potential solutions. One way of dealing with human fatigue is to use machine learning to model user preferences. This work presents initial results showing that machine learning models can be successfully used to augment user interaction, allowing the interactive genetic algorithm to find good solutions with much less user-effort.
DE: 1829 Groundwater hydrology
DE: 1832 Groundwater transport
DE: 1846 Model calibration (3333)
DE: 1847 Modeling
SC: Hydrology [H]
MN: Fall Meeting 2005