Pumping optimization of coastal aquifers based on evolutionary algorithms and surrogate modular neural network models

Pumping optimization of coastal aquifers involves complex numerical models. In problems with many decision variables, the computational burden for reaching the optimal solution can be excessive. Artificial Neural Networks (ANN) are flexible function approximators and have been used as surrogate models of complex numerical models in groundwater optimization. However, this approach is not practical in cases where the number of decision variables is large, because the required neural network structure can be very complex and difficult to train. The present study develops an optimization method based on modular neural networks, in which several small subnetwork modules, trained using a fast adaptive procedure, cooperate to solve a complex pumping optimization problem with many decision variables. The method utilizes the fact that salinity distribution in the aquifer, depends more on pumping from nearby wells rather than from distant ones. Each subnetwork predicts salinity in only one monitoring well, and is controlled by relatively few pumping wells falling within certain control distance from the monitoring well. While the initial control area is radial, its shape is adaptively improved using a Hermite interpolation procedure. The modular neural subnetworks are trained adaptively during optimization, and it is possible to retrain only the ones not performing well. As optimization progresses, the subnetworks are adapted to maximize performance near the current search space of the optimization algorithm. The modular neural subnetwork models are combined with an efficient optimization algorithm and are applied to a real coastal aquifer in the Greek island of Santorini. The numerical code SEAWAT was selected for solving the partial differential equations of flow and density dependent transport. The decision variables correspond to pumping rates from 34 wells. The modular subnetwork implementation resulted in significant reduction in CPU time and identified an even better solution than the original numerical model.

[1]  Kourosh Behzadian,et al.  Stochastic sampling design using a multi-objective genetic algorithm and adaptive neural networks , 2009, Environ. Model. Softw..

[2]  Leah L. Rogers,et al.  Solving Problems in Environmental Engineering and Geosciences with Artificial Neural Networks , 1996 .

[3]  Luis A. Garcia,et al.  Using neural networks for parameter estimation in ground water , 2006 .

[4]  Irwin Remson,et al.  AN EFFICIENT RESPONSE MATRIX METHOD FOR COUPLING A GROUNDWATER SIMULATOR AND A REGIONAL AGRICULTURAL MANAGEMENT MODEL1 , 1986 .

[5]  Mohammad N. Almasri,et al.  Modular neural networks to predict the nitrate distribution in ground water using the on-ground nitrogen loading and recharge data , 2005, Environ. Model. Softw..

[6]  Dimitri P. Solomatine,et al.  Baseflow separation techniques for modular artificial neural network modelling in flow forecasting , 2007 .

[7]  Mitsuo Kawato,et al.  Recognition of manipulated objects by motor learning with modular architecture networks , 1991, Neural Networks.

[8]  B. Ataie‐Ashtiani,et al.  Tidal effects on sea water intrusion in unconfined aquifers , 1999 .

[9]  S. Murty Bhallamudi,et al.  Optimal Groundwater Management in Deltaic Regions using Simulated Annealing and Neural Networks , 2003 .

[10]  David P. Ahlfeld,et al.  The impact of numerical precision on the solution of confined and unconfined optimal hydraulic control problems , 1996 .

[11]  Ferenc Szidarovszky,et al.  Artificial Neural Network Approach for Predicting Transient Water Levels in a Multilayered Groundwater System under Variable State, Pumping, and Climate Conditions , 2003 .

[12]  Julie Q. Shang,et al.  Using Complex Permittivity and Artificial Neural Networks for Contaminant Prediction , 2002 .

[13]  A. Cheng,et al.  BEM solution of stochastic seawater intrusion problems , 1999 .

[14]  S. N. Rao,et al.  Planning groundwater development in coastal aquifers / Planification du développement de la ressource en eau souterraine des aquifères côtiers , 2004 .

[15]  Giuseppe Gambolati,et al.  Coupled and partially coupled Eulerian‐Lagrangian Model of freshwater‐seawater mixing , 1992 .

[16]  J. Carrera,et al.  Inverse modelling of seawater intrusion in the Llobregat delta deep aquifer , 1997 .

[17]  W. Kinzelbach,et al.  Coupled flow and salinity transport modelling in semi-arid environments: The Shashe River Valley, Botswana , 2006 .

[18]  Bernd Freisleben,et al.  Approximating a finite element model by neural network prediction for facility optimization in groundwater engineering , 2005, Eur. J. Oper. Res..

[19]  Richard C. Peralta,et al.  Optimal design of aquifer cleanup systems under uncertainty using a neural network and a genetic algorithm , 1999 .

[20]  B. S. Thandaveswara,et al.  Conjunctive Use of Surface and Groundwater for Coastal and Deltaic Systems , 2004 .

[21]  John A. Nelder,et al.  A Simplex Method for Function Minimization , 1965, Comput. J..

[22]  Ashu Jain,et al.  Identification of Unknown Groundwater Pollution Sources Using Artificial Neural Networks , 2004 .

[23]  A. Mantoglou,et al.  Optimal design of pumping networks in coastal aquifers using sharp interface models , 2008 .

[24]  David J. C. MacKay,et al.  Bayesian Interpolation , 1992, Neural Computation.

[25]  Richard H. Bartels,et al.  Interpolating splines with local tension, continuity, and bias control , 1984, SIGGRAPH.

[26]  Pedro J. Restrepo,et al.  A methodology to investigate brackish groundwater desalination coupled with aquifer recharge by treated wastewater as an alternative strategy for water supply in Mediterranean areas , 2001 .

[27]  Douchko Romanov,et al.  Evolution of porosity in the saltwater-freshwater mixing zone of coastal carbonate aquifers : An alternative modelling approach , 2006 .

[28]  Aris Psilovikos.  Response Matrix Minimization Used in Groundwater Management with Mathematical Programming: A Case Study in a Transboundary Aquifer in Northern Greece , 2006 .

[29]  George Kourakos,et al.  Pumping Optimization of Coastal Aquifers using 3-D Density Models and Approximations with Neural Networks , 2006 .

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

[31]  Virginia M. Johnson,et al.  Accuracy of Neural Network Approximators in Simulation-Optimization , 2000 .

[32]  W. E. Engeler,et al.  Neural network learning time: effects of network and training set size , 1989, International 1989 Joint Conference on Neural Networks.

[33]  Farooq Azam,et al.  Biologically Inspired Modular Neural Networks , 2000 .

[34]  G. Dagan,et al.  Seawater-freshwater interface in a stratified aquifer of random permeability distribution , 1998 .

[35]  Michael I. Jordan,et al.  A Competitive Modular Connectionist Architecture , 1990, NIPS.

[36]  Paul P. Wang,et al.  MT3DMS: A Modular Three-Dimensional Multispecies Transport Model for Simulation of Advection, Dispersion, and Chemical Reactions of Contaminants in Groundwater Systems; Documentation and User's Guide , 1999 .

[37]  Ioannis K. Nikolos,et al.  Artificial neural networks as an alternative approach to groundwater numerical modelling and environmental design , 2008 .

[38]  Jagath J. Kaluarachchi,et al.  Application of artificial neural network and genetic algorithm in flow and transport simulations , 1998 .

[39]  David P. Ahlfeld,et al.  Optimal management of flow in groundwater systems , 2000 .

[40]  A. Mantoglou.  Pumping management of coastal aquifers using analytical models of saltwater intrusion , 2003 .

[41]  Mustafa M. Aral,et al.  Multi-objective optimization of pumping rates and well placement in coastal aquifers , 2004 .

[42]  Bithin Datta,et al.  Development of multiobjective management models for coastal aquifers , 1999 .

[43]  Arlen W. Harbaugh,et al.  A modular three-dimensional finite-difference ground-water flow model , 1984 .

[44]  Gedeon Dagan,et al.  The salt-water body in the Northern part of Yarkon-Taninim aquifer: Field data analysis, conceptual model and prediction , 2006 .

[45]  Costantino Masciopinto,et al.  Simulation of coastal groundwater remediation: the case of Nardò fractured aquifer in Southern Italy , 2006, Environ. Model. Softw..

[46]  A. Mahesha,et al.  Control of Seawater Intrusion through Injection-Extraction Well System , 1996 .

[47]  Aristotelis Mantoglou,et al.  Management of coastal aquifers based on nonlinear optimization and evolutionary algorithms , 2004 .

[48]  S. V. N. Rao,et al.  A Computationally Efficient Technique for Source Identification Problems in Three-Dimensional Aquifer Systems Using Neural Networks and Simulated Annealing , 2006 .

[49]  W. Kinzelbach,et al.  Numerical simulation of three-dimensional saltwater-freshwater fingering instabilities observed in a porous medium , 2006 .

[50]  Anthony J. Smith.  Mixed convection and density-dependent seawater circulation in coastal aquifers , 2004 .

[51]  Eric G. Reichard,et al.  Assessment of Regional Management Strategies for Controlling Seawater Intrusion , 2005 .

[52]  P. F. Andersen,et al.  Saltwater intrusion in aquifers: Development and testing of a three‐dimensional finite element model , 1987 .

[53]  David E. Thompson,et al.  Optimization of a Model to Predict Salinity Intrusion in San Francisco Bay Estuary Using a Genetic Algorithm , 2002 .

[54]  Rao S. Govindaraju,et al.  Prediction of watershed runoff using Bayesian concepts and modular neural networks , 2000 .

[55]  B. Datta,et al.  OPTIMIZATION BASED SOLUTION OF DENSITY DEPENDENT SEAWATER INTRUSION IN COASTAL AQUIFERS , 2000 .

[56]  Barbara S. Minsker,et al.  Optimal groundwater remediation design using an Adaptive Neural Network Genetic Algorithm , 2006 .

[57]  C. Voss,et al.  SUTRA (Saturated-Unsaturated Transport). A Finite-Element Simulation Model for Saturated-Unsaturated, Fluid-Density-Dependent Ground-Water Flow with Energy Transport or Chemically-Reactive Single-Species Solute Transport. , 1984 .

[58]  Demetris Koutsoyiannis,et al.  Calibration of a semi-distributed model for conjunctive simulation of surface and groundwater flows / Calage d’un modèle semi-distribué pour la simulation conjointe d’écoulements superficiels et souterrains , 2004 .