Using Artificial Neural Networks to Forecast Trichloroethylene Concentrations at the Paducah Gaseous Diffusion Plant
- Univ of KY, Dept of Civil Engineering
To determine the future extent of the TCE contamination plume at PGDP, a groundwater and solute transport model has been developed by the Department of Energy (DOE). The model used to perform these calculations is MODFLOWT which is an enhanced groundwater transport model developed by the United States Geological Survey (USGS). MODFLOWT models groundwater movement as well as the transport of species that are subject to adsorption and decay by using a finite difference method (Duffield et al 2001). A significant limitation of MODFLOWT is that it requires large amounts of data. This data can be difficult and expensive to obtain. MODFLOWT also requires excessive computational time to perform one simulation. It is desirable to have a model that can predict the spatial extent of the contaminant plume without as much required data and that does not require excessive computational times. The purpose of this study is to develop and alternative model to MODFLOWT that can produce similar results for possible use in a companion management model. The alternative model used in this study is an artificial neural network (ANN).
- Research Organization:
- Kentucky Research Consortium for Energy and Environment, University of Kentucky, Lexington, KY
- Sponsoring Organization:
- USDOE Office of Environmental Management (EM)
- DOE Contract Number:
- FG05-03OR23032
- OSTI ID:
- 1233312
- Report Number(s):
- UK/KRCEE doc#: 12.2 2007; UK/KRCEE doc#: 12.2 2007
- Country of Publication:
- United States
- Language:
- English
Similar Records
TECHNICAL EVALUATION OF TEMPORAL GROUNDWATER MONITORING VARIABILITY IN MW66 AND NEARBY WELLS, PADUCAH GASEOUS DIFFUSION PLANT
Evaluation of natural attenuation processes for trichloroethylene and technetium-99 in the Northeast and Northwest plumes at the Paducah Gaseous Diffusion Plant, Paducah, Kentucky