Surrogate-Based Stochastic Multiobjective Optimization for Coastal Aquifer Management under Parameter Uncertainty

Linked simulation-optimization (S/O) approaches have been extensively used as tools in coastal aquifer management. However, parameter uncertainties seawater intrusion (SI) simulation models often undermine the reliability of derived solutions. In this study, a stochastic S/O framework is presented and applied to real-world case Longkou China. The three conflicting objectives maximizing total pumping rate, minimizing injection solute mass increase are considered optimization model. uncertain parameters contained both constraints objective functions. A multiple realization approach utilized address uncertainty model parameters, new multiobjective evolutionary algorithm (EN-NSGA2) proposed solve EN-NSGA2 overcomes some inherent limitations traditional nondominated sorting genetic algorithm-II (NSGA-II) by introducing information entropy theory. comparison results indicate that can effectively ameliorate diversity Pareto-optimal For computational challenge process, surrogate based on multigene programming (MGGP) method developed substitute for numerical show MGGP tremendously reduce burden while ensuring an acceptable level accuracy.