Approaching the inverse problem of parameter estimation in groundwater models by means of artificial neural networks

Mathematical models are widely used to simulate the behaviour of groundwater systems under various physical conditions. The development of these models requires that the system state evolution be predicted by solving appropriate governing equations. This also requires that the values of the parameters imbedded in the models be accurately determined. Since the hydraulic and water quality parameters are in general not directly measurable, their estimation must be obtained by solving the so called inverse parameter identification problem which amounts to determining the unknown physical parameters by fitting observed histories of the system states. In this paper we investigate the feasibility of solving the inverse problem by using artificial neural networks. These computational tools have the ability to learn complex input-output relationships and have a very advantageous property of generalization. To demonstrate the approach, a simple case study is considered consisting of an analytic model of contaminant transport due to a point source in stationary flow field. The model is used to simulate the behaviour of the groundwater systems for different values of the dispersion coefficient. A set of supervised, multilayered, feedforward neural networks are then trained to predict the values of the parameter corresponding to given concentration histories.

[1]  P. Gill,et al.  Aquifer Reclamation Design: The Use of Contaminant Transport Simulation Combined With Nonlinear Programing , 1984 .

[2]  K. H. Coats,et al.  A New Technique for Determining Reservoir Description from Field Performance Data , 1970 .

[3]  Richard S. Sutton,et al.  Neural networks for control , 1990 .

[4]  W. G. Gray,et al.  A Galerkin‐finite element technique for calculating the transient position of the saltwater front , 1975 .

[5]  J. Seinfeld.  Identification of parameters in partial differential equations , 1969 .

[6]  William W.-G. Yeh,et al.  Systems Analysis in Ground-Water Planning and Management , 1992 .

[7]  S. Gorelick.  A review of distributed parameter groundwater management modeling methods , 1983 .

[8]  W. Yeh,et al.  Optimal Identification of Aquifer Diffusivity Using Quasilinearization , 1971 .

[9]  Berndt Müller,et al.  Neural networks: an introduction , 1990 .

[10]  P. Jacquard,et al.  Permeability Distribution From Field Pressure Data , 1965 .

[11]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[12]  W. Yeh,et al.  Identification of parameters in an inhomogeneous aquifer by use of the maximum principle of optimal control and quasi-linearization , 1974 .

[13]  John M. Mulvey,et al.  Contaminated groundwater remediation design using simulation, optimization, and sensitivity theory: 1. Model development , 1988 .

[14]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[15]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[16]  Enrico Zio,et al.  Fault Diagnosis Via Neural Networks: The Boltzmann Machine , 1994 .

[17]  W. Yeh.  Review of Parameter Identification Procedures in Groundwater Hydrology: The Inverse Problem , 1986 .

[18]  L. Kent Thomas,et al.  A Nonlinear Automatic History Matching Technique for Reservoir Simulation Models , 1972 .

[19]  C. Lombardi,et al.  THERMALHYDRAULIC CORRELATIONS BY NEURAL NETWORKS , 1994 .

[20]  W. Yeh,et al.  Identification of parameters in unsteady open channel flows , 1972 .

[21]  C. V. Theis.  The relation between the lowering of the Piezometric surface and the rate and duration of discharge of a well using ground‐water storage , 1935 .

[22]  Clifford Lau,et al.  Neural Networks: Theoretical Foundations and Analysis , 1991 .

[23]  N. R. Ortiz,et al.  Risk methodology for geologic disposal of radioactive waste: Final report , 1987 .

[24]  G. E. Slater,et al.  Adjustment of Reservoir Simulation Models To Match Field Performance , 1971 .

[25]  Russell C. Eberhart,et al.  Neural network PC tools: a practical guide , 1990 .

[26]  M. L. Wasserman,et al.  A New Algorithm for Automatic History Matching , 1974 .

[27]  S. P. Neuman.  Calibration of distributed parameter groundwater flow models viewed as a multiple‐objective decision process under uncertainty , 1973 .

[28]  G. Pinder,et al.  A Numerical Technique for Calculating the Transient Position of the Saltwater Front , 1970 .

[29]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[30]  B.G.J. Thompson,et al.  The development and application of integrated procedures for post-closure assessment, based upon Monte Carlo simulation: the probabilistic systems assessment (PSA) approach , 1993 .

[31]  Geoffrey E. Hinton,et al.  Learning and relearning in Boltzmann machines , 1986 .

[32]  V. Cerný.  Thermodynamical approach to the traveling salesman problem: An efficient simulation algorithm , 1985 .