Bayesian set pair analysis and machine learning based ensemble surrogates for optimal multi-aquifer system remediation design

Abstract Surrogate models are often adopted to substitute computationally intensive groundwater simulation models for aquifer management due to their high effectiveness and computing efficiency. However, solutions of using only one surrogate model are prone to large prediction uncertainty. This study compares individual surrogate models and an ensemble surrogate model in an optimal groundwater remediation design problem. Three machine learning based surrogate models (response surface regression model, artificial neural network and support vector machine) were developed to replace a high-fidelity solute transport model for predicting saltwater intrusion and assisting saltwater scavenging design. An optimal Latin hypercube design was employed to generate training and testing datasets. Set pair analysis was employed to construct a more reliable ensemble surrogate and to address prediction uncertainty arising from individual surrogate models. Bayesian set pair weights were derived by utilizing full information from both training and testing data and improved typical set pair weights. The individual and ensemble surrogate models were applied to the salinization remediation problem in the Baton Rouge area, southeast Louisiana. The optimal remediation design includes two conflicting objectives: minimizing total groundwater extraction from a horizontal scavenger well while maximizing chloride concentration difference to the MCL (maximum contamination level) at monitoring locations. NSGA-II (Non-dominated Sorting Genetic Algorithm II) was employed to solve the nonlinear optimization model and obtain Pareto-optimal pumping schedules. The optimal pumping schedules from ensemble surrogate models and individual surrogate models were verified by the solute transport model. The study found that Bayesian set pair analysis builds robust ensemble surrogates and accounts for model prediction uncertainty. The ensemble-surrogate-assisted optimization model provides stable and reliable solutions while considerably alleviating computational burden.

[1]  H. P. Williams,et al.  Model Building in Mathematical Programming , 1979 .

[2]  Vimal Savsani,et al.  Multi-Objective Optimization of Vehicle Passive Suspension System Using NSGA-II, SPEA2 and PESA-II , 2016 .

[3]  I-Fan Chang,et al.  Support vector regression for real-time flood stage forecasting , 2006 .

[4]  H. Akaike.  A Bayesian extension of the minimum AIC procedure of autoregressive model fitting , 1979 .

[5]  B. Men,et al.  Evaluation of Sustainable Use of Water Resources in the Beijing-Tianjin-Hebei Region Based on S-Type Functions and Set Pair Analysis , 2018, Water.

[6]  Manuel A. Andrade,et al.  Enhanced Artificial Neural Networks Estimating Water Quality Constraints for the Optimal Water Distribution Systems Design , 2016 .

[7]  Jin Lin,et al.  A comparative research of different ensemble surrogate models based on set pair analysis for the DNAPL-contaminated aquifer remediation strategy optimization. , 2017, Journal of contaminant hydrology.

[8]  Wenxi Lu,et al.  Application of ensemble surrogates and adaptive sequential sampling to optimal groundwater remediation design at DNAPLs-contaminated sites. , 2017, Journal of contaminant hydrology.

[9]  Andrea Castelletti,et al.  Balancing exploration, uncertainty and computational demands in many objective reservoir optimization , 2017 .

[10]  J. Hanor,et al.  Spatial Variations in Subsurface Pore Fluid Properties in a Portion of Southeast Louisiana: Implications for Regional Fluid Flow and Solute Transport , 1990 .

[11]  Chih-Jen Lin,et al.  Asymptotic Behaviors of Support Vector Machines with Gaussian Kernel , 2003, Neural Computation.

[12]  D. A. Barry,et al.  Seawater intrusion processes, investigation and management: Recent advances and future challenges , 2013 .

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

[14]  Anthony J. Jakeman,et al.  A review of surrogate models and their application to groundwater modeling , 2015 .

[15]  N. Draper,et al.  Applied Regression Analysis , 1966 .

[16]  Frank T.-C. Tsai,et al.  Saltwater scavenging optimization under surrogate uncertainty for a multi-aquifer system , 2018, Journal of Hydrology.

[17]  Frank T.-C. Tsai,et al.  Bayesian model averaging assessment on groundwater management under model structure uncertainty , 2010 .

[18]  Jichun Wu,et al.  Replenishing an unconfined coastal aquifer to control seawater intrusion: Injection or infiltration? , 2017 .

[19]  N. McIntyre,et al.  Stochastic Modeling of Groundwater Extractions over a Data‐Sparse Region of Australia , 2019, Ground water.

[20]  Bithin Datta,et al.  Adaptive Management of Coastal Aquifers Using Entropy-Set Pair Analysis–Based Three-Dimensional Sequential Monitoring Network Design , 2019 .

[21]  V. Merwade,et al.  Separation and prioritization of uncertainty sources in a raster based flood inundation model using hierarchical Bayesian model averaging , 2019, Journal of Hydrology.

[22]  Venkatesh Merwade,et al.  Accounting for model structure, parameter and input forcing uncertainty in flood inundation modeling using Bayesian model averaging , 2018, Journal of Hydrology.

[23]  Boxin Tang.  Orthogonal Array-Based Latin Hypercubes , 1993 .

[24]  Vahid Nourani,et al.  Conjunction of radial basis function interpolator and artificial intelligence models for time-space modeling of contaminant transport in porous media , 2017 .

[25]  Jianfeng Wu,et al.  An Elitist Multiobjective Tabu Search for Optimal Design of Groundwater Remediation Systems , 2017, Ground water.

[26]  George P. Karatzas,et al.  Spatiotemporal geostatistical modeling of groundwater levels under a Bayesian framework using means of physical background , 2019, Journal of Hydrology.

[27]  Mohammad Rajabi,et al.  Optimal Management of a Freshwater Lens in a Small Island Using Surrogate Models and Evolutionary Algorithms , 2014 .

[28]  Nima Chitsazan,et al.  Bayesian Chance‐Constrained Hydraulic Barrier Design under Geological Structure Uncertainty , 2015, Ground water.

[29]  Frank T.-C. Tsai,et al.  Conjunctive management of surface and groundwater resources under projected future climate change scenarios , 2016 .

[30]  Bryan A. Tolson,et al.  Review of surrogate modeling in water resources , 2012 .

[31]  Vasileios Christelis,et al.  Pumping Optimization of Coastal Aquifers Assisted by Adaptive Metamodelling Methods and Radial Basis Functions , 2016, Water Resources Management.

[32]  Changkuan Zhang,et al.  Advances in Water Resources and Hydraulic Engineering , 2009 .

[33]  F. Tsai,et al.  Steady-State Approximate Freshwater–Saltwater Interface in a Two-Horizontal-Well Scavenging System , 2019, Journal of hydrologic engineering.

[34]  Lihua Feng,et al.  Statistical Prediction of Changes in Water Resources Trends Based on Set Pair Analysis , 2014, Water Resources Management.

[35]  Kazuro Momii,et al.  Effects of Recharge Wells and Flow Barriers on Seawater Intrusion , 2011, Ground water.

[36]  Bithin Datta,et al.  Multi-objective management of saltwater intrusion in coastal aquifers using genetic programming and modular neural network based surrogate models. , 2010 .

[37]  Wolfgang Nowak,et al.  Explicit treatment for Dirichlet, Neumann and Cauchy boundary conditions in POD-based reduction of groundwater models , 2018 .

[38]  Petros Koumoutsakos,et al.  Reducing the Time Complexity of the Derandomized Evolution Strategy with Covariance Matrix Adaptation (CMA-ES) , 2003, Evolutionary Computation.

[39]  Yi Ji,et al.  Identifying the release history of a groundwater contaminant source based on an ensemble surrogate model , 2019, Journal of Hydrology.

[40]  Frank T.-C. Tsai,et al.  Bayesian model averaging for groundwater head prediction and uncertainty analysis using multimodel and multimethod , 2009 .

[41]  Michael Andrew Christie,et al.  Surrogate accelerated sampling of reservoir models with complex structures using sparse polynomial chaos expansion , 2015 .

[42]  F. Tsai,et al.  Modeling complex aquifer systems: a case study in Baton Rouge, Louisiana (USA) , 2017, Hydrogeology Journal.

[43]  Jian Song,et al.  Adaptive surrogate model based multiobjective optimization for coastal aquifer management , 2018, Journal of Hydrology.

[44]  Atharv Bhosekar,et al.  Advances in surrogate based modeling, feasibility analysis, and optimization: A review , 2018, Comput. Chem. Eng..

[45]  J. McCray,et al.  Evaluation of Groundwater Levels in the Arapahoe Aquifer Using Spatiotemporal Regression Kriging , 2019, Water Resources Research.

[46]  A. Hense,et al.  A Bayesian approach to climate model evaluation and multi‐model averaging with an application to global mean surface temperatures from IPCC AR4 coupled climate models , 2006 .

[47]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[48]  Harish Garg,et al.  Connection number of set pair analysis based TOPSIS method on intuitionistic fuzzy sets and their application to decision making , 2017, Applied Intelligence.

[49]  P. Barlow,et al.  Saltwater intrusion in coastal regions of North America , 2010 .

[50]  Guoru Huang,et al.  Set pair analysis for karst waterlogging risk assessment based on AHP and entropy weight , 2018 .

[51]  W. M. Bolstad.  Introduction to Bayesian Statistics , 2004 .

[52]  Mohammad Rajabi,et al.  Uncertainty-based simulation-optimization using Gaussian process emulation: Application to coastal groundwater management , 2017 .

[53]  Rao S. Govindaraju,et al.  COMPARISON OF ANNS AND EMPIRICAL APPROACHES FOR PREDICTING WATERSHED RUNOFF , 2000 .

[54]  Frank T.-C. Tsai,et al.  Uncertainty Segregation and Comparative Evaluation in Groundwater Remediation Designs: A Chance-Constrained Hierarchical Bayesian Model Averaging Approach , 2015 .

[55]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[56]  Azizallah Izady,et al.  An efficient surrogate-based simulation-optimization method for calibrating a regional MODFLOW model , 2017 .

[57]  Mac McKee,et al.  Applicability of statistical learning algorithms in groundwater quality modeling , 2005 .

[58]  Bithin Datta,et al.  Coupled simulation‐optimization model for coastal aquifer management using genetic programming‐based ensemble surrogate models and multiple‐realization optimization , 2011 .