APPLICATION OF ARTIFICIAL NEURAL-NETWORK AND GENETIC ALGORITHM IN FLOW AND TRANSPORT SIMULATIONS

Artificial neural network (ANN) is considered to be a powerful tool fo r solving groundwater problems which require a large number of flow an d contaminant transport (GFCT) simulations. Often, GFCT models are non linear, and they are difficult to solve using traditional numerical me thods to simulate specific input-output responses. In order to avoid t hese difficulties, ANN may be used to simulate the GFCT responses expl icitly. In this manuscript, recent research related to the application of ANN in simulating GFCT responses is critically reviewed, and six r esearch areas are identified. In order to study these areas, a one-dim ensional unsaturated flow and transport scenario was developed, and AN N was used to simulate the effects of specific GFCT parameters on over all results. Using these results, ANN concepts related to architecture , sampling, training, and multiple function approximations are studied , and ANN training using back-propagation algorithm (BPA) and genetic algorithm (GA) an compared. These results are summarized, and appropri ate conclusions are made. (C) 1998 Elsevier Science Limited. All right s reserved.