An adaptive Kriging surrogate method for efficient joint estimation of hydraulic and biochemical parameters in reactive transport modeling

Jun, Zhou; Su, Xiaosi; Cui, Geng

doi:10.1016/j.jconhyd.2018.08.005

Cited by 24 publications

(12 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To further mitigate the computational burden and approximate errors, in the recent years many strategies have been introduced to enable efficient data-driven model reconstruction, for example, compressed sensing, adaptive and/or multilevel, and multifidelity strategies (Gong et al, 2016;Ju et al, 2018;Laloy et al, 2013;Mo et al, 2017;Zhang et al, 2017Zhang et al, , 2018Zhang et al, , 2020Zhou et al, 2018). For example, Adam et al (2020) incorporate a TSVD (truncated singular value decomposition)-based dimensionality reduction method to reduce the number of variables and thereby decrease the HFM runs needed in GPR surrogate.…”

Section: Introductionmentioning

confidence: 99%

“…To further mitigate the computational burden and approximate errors, in the recent years many strategies have been introduced to enable efficient data‐driven model reconstruction, for example, compressed sensing, adaptive and/or multilevel, and multifidelity strategies (Gong et al., 2016; Ju et al., 2018; Laloy et al., 2013; Mo et al., 2017; Zhang et al., 2017, 2018, 2020; Zhou et al., 2018). For example, Adam et al.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Surrogate‐Based Joint Estimation of Subsurface Geological and Relative Permeability Parameters for High‐Dimensional Inverse Problem by Use of Smooth Local Parameterization

Xiao

Tian

2020

Water Resources Research

View full text Add to dashboard Cite

This paper introduces an efficient surrogate model with the aim of accelerating joint estimation of subsurface geological properties and relative permeability parameters for high‐dimensional inversion problems. We fully replace the high‐fidelity model with a set of subdomain linear models through integrating model linearization with smooth local parameterization where the Gaussian geological parameters and non‐Gaussian facies indicators are locally parameterized. These subdomain linear models with smooth local parameterization, referred to as SLM‐SLP, are constructed in each subdomain individually using only a few high‐fidelity model simulations. An adaptive scheme, that is, weighting smooth local parameterization (WSLP), is introduced as well to mitigate the negative effects of inappropriate domain decomposition schemes by adaptively optimizing the domain decomposition strategy. The computational advantages of the proposed algorithm are demonstrated on a synthetic non‐Gaussian facies model and a real‐world high‐dimensional Gaussian model. The amount of computational cost has been drastically reduced while reasonable accuracy remains. Specifically, SLM‐SLP only needs 220 fidelity simulations to optimize 302 parameters. Compared to ensemble smoother with multiple data assimilation (ES‐MDA), SLM‐SLP effectively and efficiently mitigates the ensemble collapse problem in the course of uncertain quantification.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Surrogate‐Based Joint Estimation of Subsurface Geological and Relative Permeability Parameters for High‐Dimensional Inverse Problem by Use of Smooth Local Parameterization

Xiao

Tian

2020

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…To help alleviate this problem, the Kriging method (Zhou et al 2018) is introduced in order to construct a surrogate model to replace the seawater intrusion simulation model, which can then be coupled with the optimization model, greatly saving on calculation costs.…”

Section: Simulation-optimization Methodsmentioning

confidence: 99%

Optimal location of cutoff walls for seawater intrusion

Zheng

Fan

et al. 2021

Appl Water Sci

View full text Add to dashboard Cite

In this paper, the simulation–optimization method is used to study the optimal location of cutoff walls for seawater intrusion. The optimization model is based on minimizing the chlorine concentration of two water sources after 50 years. In order to reduce the computational complexity, a Kriging surrogate model simulation is coupled with the optimization model. Finally, a hypothetical case is used to evaluate the accuracy of the surrogate model and the performance of the optimization model. The results show that the outputs of the Kriging surrogate model and the variable density groundwater simulation model for the same cutoff wall design fit well, and the average relative error of the two outputs is only 2.2% which proves that it is feasible to apply the Kriging surrogate model to this problem. By solving the optimization model, the location of the cutoff wall which minimizes the sum of chlorine concentration of the two water sources after 50 years is obtained. This provides a stable and reliable method for the site selection of cutoff walls for future projects intended to prevent and control seawater intrusion.

show abstract

“…Several surrogate model methods have been proposed in prior studies. This includes and is not limited to polynomial chaos expansion (Laloy et al., 2013; Xiu & Karniadakis, 2002), Gaussian process (H. Wang & Li, 2018; J. Zhang et al., 2016), Kriging surrogate modeling (X. Yan et al., 2019; J. Zhou et al., 2018), support vector machine (Lal & Datta, 2018; Xingpo et al., 2021), conventional artificial neural network (i.e., single hidden layer neural network; Kourakos & Mantoglou, 2009; Shin et al., 2019), and radial basis function (Y. Liu, Wang, et al., 2019; Xing et al., 2019). Unfortunately, these conventional methods are faced with the problem of dimensionality (i.e., “curse of dimensionality”), since the computational cost for constructing surrogate models increases exponentially as the input dimensionality increases (Asher et al., 2015; Liao et al., 2017; Mo, Zabaras, et al., 2019).…”

Section: Introductionmentioning

confidence: 99%

“…Zhang et al, 2016), Kriging surrogate modeling (X. Yan et al, 2019;J. Zhou et al, 2018), support vector machine (Lal & Datta, 2018;Xingpo et al, 2021), conventional artificial neural network (i.e., single hidden layer neural network; Kourakos & Mantoglou, 2009;Shin et al, 2019), and radial basis function (Y.…”

mentioning

confidence: 99%

An Improved Tandem Neural Network Architecture for Inverse Modeling of Multicomponent Reactive Transport in Porous Media

Chen

Dai

Yang

et al. 2021

Water Resources Research

View full text Add to dashboard Cite

show abstract

An adaptive Kriging surrogate method for efficient joint estimation of hydraulic and biochemical parameters in reactive transport modeling

Cited by 24 publications

References 33 publications

Surrogate‐Based Joint Estimation of Subsurface Geological and Relative Permeability Parameters for High‐Dimensional Inverse Problem by Use of Smooth Local Parameterization

Surrogate‐Based Joint Estimation of Subsurface Geological and Relative Permeability Parameters for High‐Dimensional Inverse Problem by Use of Smooth Local Parameterization

Optimal location of cutoff walls for seawater intrusion

An Improved Tandem Neural Network Architecture for Inverse Modeling of Multicomponent Reactive Transport in Porous Media

Contact Info

Product

Resources

About