Aggregated surrogate simulator for groundwater-surface water management via simulation-optimization modeling: Theory, development and tests. (October 2017)

• Record Type: Journal Article
• Authors: Timani, Bassel; Peralta, Richard
• Is Part Of: Environmental modelling & software. Volume 96(2017)
• Page Start: 210
• Page End: 231
• Publication Date: 2017-10
• Subjects: Surrogate simulators -- Simulation-optimization -- Groundwater modeling -- Water management -- Groundwater/surface water interaction -- Metamodeling -- Convolution -- Linearization; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2017.06.014
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Computational requirements sometimes discourage using mathematical optimization for groundwater management. To dramatically reduce computation time, the presented hybrid response matrix method (RMM), Coefficient Generation and Use method 4 (CGU4), prepares surrogate simulators used during optimization. CGU4 reduces numerical model simulations needed to populate the surrogates (linearized convolution equations, LCEs). LCEs often represent flow and head constraints within groundwater flow optimization problems. CGU4 reduces computations for problems having: varying time period sizes; eras of sequential time periods of equal duration; and system (non)linearity. For a situation having: linear, piece-wise, and nonlinear groundwater flows; 20 periods of varying and sequentially constant durations; and optimization problems employing (non)linear objective functions and linear head and aquifer-surface seepage constraints, CGU4 requires 22–61% fewer simulations to compute optimal objective function values within 0.001–0.003% of the best alternative RMM. For hypothetical (non)linear dynamic stream-aquifer problems, CGU4 required the same or 63–89% less time than previous RMMs. Highlights: CGU4 reduces computation time to populate linearized convolution equations (LCEs). LCEs are flow and head constraints in groundwater management optimization problems. CGU4 is applicable for (non)linear systems and objective functions. CGU4 reduced computation time up to 89% versus other LCE-preparation methods.