Review of surrogate modeling in water resources

By Saman Razavi, Bryan Tolson, Donald H. Burn. 2012

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

An evaluation of adaptive surrogate modeling based optimization with two benchmark problems (2014, methodology): “…families of surrogates: (1) response surface surrogates, which are statistical or empirical data-driven models emulating the responses of a high-fidelity simulation model; and (2) lower-fidelity physical based surrogates, which are simplified models of the original system (Razavi et al., 2012b).”
Model-based methods for continuous and discrete global optimization (2017, methodology): “Further applications from the water industry are published by Razavi et al [49].”
Evaluating marginal likelihood with thermodynamic integration method and comparison with several other numerical methods (2016, methodology): “This is feasible, given that surrogate modeling has been widely used in uncertainty quantification of groundwater modeling [Razavi et al., 2012; Laloy et al., 2013; Zhang et al., 2013].”
Variance-based global sensitivity analysis for multiple scenarios and models with implementation using sparse grid collocation (2015, methodology): “The quantitative measures have been used recently for model structure diagnosis with respect to model complexity and model structure inadequacy (Rosolem et al., 2012; Gupta et al., 2012).”
The Future of Sensitivity Analysis: An essential discipline for systems modeling and policy support (2020, unknown): “Surrogate models, also called response surface models, metamodels or emulators, are used in lieu of computationally intensive models and can be statistical (i.e., ‘response surface surrogates ’ ) or mechanistic (i.e., ‘lower-fidelity mechanistic surrogates ’ ) (Razavi et al., 2012).”
An adaptive sparse‐grid high‐order stochastic collocation method for Bayesian inference in groundwater reactive transport modeling (2013, methodology): “…quantification of model uncertainty, because quantifying parametric uncertainty is the basis of quantifying model structure uncertainty in the popular multimodel analysis methods [Neuman, 2003; Ye et al., 2004; Poeter and Hill, 2007; Refsgaard et al., 2012; Neuman et al., 2012; Lu et al., 2012a].”
Polynomial chaos expansions for uncertainty propagation and moment independent sensitivity analysis of seawater intrusion simulations (2015, background): “…simplified and less-detailed versions of the original model while retaining the key physical characteristics (e.g., Kerrou and Renard, 2010), and (4) response surface surrogate modeling (i.e. developing data-driven and physics-free approximations of the model response (Razavi et al., 2012)).”
Digital Twin: Values, Challenges and Enablers From a Modeling Perspective (2019, unknown): “[256] S.”
A machine learning framework to forecast wave conditions (2017, methodology): “Here, we define a surrogate model (Razavi et al., 2012) as a data-driven technique to empirically approximate the response surface of a physics-based model.”
Deep Probabilistic Surrogate Networks for Universal Simulator Approximation (2019, methodology): “Currently, the most commonly used methods for constructing deterministic surrogate models (Razavi et al., 2012) include Kriging (Simpson et al.”
Analysis of regression confidence intervals and Bayesian credible intervals for uncertainty quantification (2012, methodology): “…credible intervals can be reduced by using surrogate modeling such as response surface surrogates and low-fidelity physically based surrogates [Razavi et al., 2012], surrogate modeling generally requires thousands of model runs to establish the response surface surrogate and the 1Department…”
Deep Convolutional Encoder‐Decoder Networks for Uncertainty Quantification of Dynamic Multiphase Flow in Heterogeneous Media (2018, methodology): “Computationally eﬃcient surrogate methods have achieved an increased popularity over the past decade [ Asher et al. , 2015; Razavi et al. , 2012].”
On closures for reduced order models—A spectrum of first-principle to machine-learned avenues (2021, unknown): “(33) to correctly resolve the time dynamics of W , where C represents the closure model that needs to be determined.”
Remarks on multi-output Gaussian process regression (2018, unknown): “usually employ surrogates to approximate the input-output relationship in order to relieve computational budget [1, 2, 3].”