A comparative study of machine learning models for predicting the state of reactive mixing
Abstract
Mixing phenomena are important mechanisms controlling flow, species transport, and reaction processes in fluids and porous media. Accurate predictions of reactive mixing are critical for many Earth and environmental science problems such as contaminant fate and remediation, macroalgae growth, and plankton biomass evolution. Here, to investigate the evolution of mixing dynamics under different scenarios (e.g., anisotropy, fluctuating velocity fields), a finite-element-based numerical model was built to solve the fast, irreversible bimolecular reaction-diffusion equations to simulate a range of reactive-mixing scenarios. A total of 2,315 simulations were performed using different sets of model input parameters comprising various spatial scales of vortex structures in the velocity field, time-scales associated with velocity oscillations, the perturbation parameter for the vortex-based velocity, anisotropic dispersion contrast (i.e., ratio of longitudinal-to-transverse dispersion), and molecular diffusion. The outputs comprised concentration profiles of reactants and products. The inputs to and outputs from these simulations were concatenated into feature and label matrices, respectively, to train 20 different machine learning (ML) models intended to emulate system behavior. These 20 ML emulators, based on linear methods, Bayesian methods, ensemble learning methods, and multilayer perceptrons (MLPs), were trained to classify the state of mixing and predict three quantities of interest (QoIs) characterizing speciesmore »
- Authors:
-
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Baylor Univ., Waco, TX (United States)
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
- Baylor Univ., Waco, TX (United States)
- Publication Date:
- Research Org.:
- Pacific Northwest National Laboratory (PNNL), Richland, WA (United States); Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Org.:
- USDOE Office of Fossil Energy (FE); USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE National Nuclear Security Administration (NNSA); USDOE Office of Science (SC), Biological and Environmental Research (BER); USDOE Office of Science (SC), Biological and Environmental Research (BER). Earth and Environmental Systems Science Division
- OSTI Identifier:
- 1765121
- Alternate Identifier(s):
- OSTI ID: 1762732; OSTI ID: 1775931; OSTI ID: 1778767
- Report Number(s):
- PNNL-SA-157340; LA-UR-20-21737
Journal ID: ISSN 0021-9991; TRN: US2206433
- Grant/Contract Number:
- AC05-76RL01830; 20150693ECR; 20190020DR; 89233218CNA000001; AC02-05CH11231
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Journal of Computational Physics
- Additional Journal Information:
- Journal Volume: 432; Journal ID: ISSN 0021-9991
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; surrogate modeling; machine learning; reaction-diffusion equations; random forests; ensemble methods; artificial neural networks; computer science; earth sciences; energy sciences; information science; mathematics
Citation Formats
Ahmmed, Bulbul, Mudunuru, Maruti K., Karra, Satish, James, Scott C., and Vesselinov, Velimir V. A comparative study of machine learning models for predicting the state of reactive mixing. United States: N. p., 2021.
Web. doi:10.1016/j.jcp.2021.110147.
Ahmmed, Bulbul, Mudunuru, Maruti K., Karra, Satish, James, Scott C., & Vesselinov, Velimir V. A comparative study of machine learning models for predicting the state of reactive mixing. United States. https://doi.org/10.1016/j.jcp.2021.110147
Ahmmed, Bulbul, Mudunuru, Maruti K., Karra, Satish, James, Scott C., and Vesselinov, Velimir V. Tue .
"A comparative study of machine learning models for predicting the state of reactive mixing". United States. https://doi.org/10.1016/j.jcp.2021.110147. https://www.osti.gov/servlets/purl/1765121.
@article{osti_1765121,
title = {A comparative study of machine learning models for predicting the state of reactive mixing},
author = {Ahmmed, Bulbul and Mudunuru, Maruti K. and Karra, Satish and James, Scott C. and Vesselinov, Velimir V.},
abstractNote = {Mixing phenomena are important mechanisms controlling flow, species transport, and reaction processes in fluids and porous media. Accurate predictions of reactive mixing are critical for many Earth and environmental science problems such as contaminant fate and remediation, macroalgae growth, and plankton biomass evolution. Here, to investigate the evolution of mixing dynamics under different scenarios (e.g., anisotropy, fluctuating velocity fields), a finite-element-based numerical model was built to solve the fast, irreversible bimolecular reaction-diffusion equations to simulate a range of reactive-mixing scenarios. A total of 2,315 simulations were performed using different sets of model input parameters comprising various spatial scales of vortex structures in the velocity field, time-scales associated with velocity oscillations, the perturbation parameter for the vortex-based velocity, anisotropic dispersion contrast (i.e., ratio of longitudinal-to-transverse dispersion), and molecular diffusion. The outputs comprised concentration profiles of reactants and products. The inputs to and outputs from these simulations were concatenated into feature and label matrices, respectively, to train 20 different machine learning (ML) models intended to emulate system behavior. These 20 ML emulators, based on linear methods, Bayesian methods, ensemble learning methods, and multilayer perceptrons (MLPs), were trained to classify the state of mixing and predict three quantities of interest (QoIs) characterizing species production, decay (i.e., average concentration, square of average concentration), and degree of mixing (i.e., variances of species concentration). Unsurprisingly, linear classifiers and regressors failed to reproduce the QoIs; however, ensemble methods (classifiers and regressors) and the MLP model accurately classified the state of reactive mixing and the QoIs. Among ensemble methods, random forest and decision-tree-based AdaBoost faithfully predicted the QoIs. At run time, trained ML emulators produced results times faster than the finite-element simulations. Due to their low computational expense and high accuracy, ensemble and MLP models are excellent emulators for these numerical simulations and great utilities in uncertainty quantification exercises, which can require 1,000s of forward model runs.},
doi = {10.1016/j.jcp.2021.110147},
journal = {Journal of Computational Physics},
number = ,
volume = 432,
place = {United States},
year = {Tue Jan 19 00:00:00 EST 2021},
month = {Tue Jan 19 00:00:00 EST 2021}
}
Works referenced in this record:
Machine learning in concrete strength simulations: Multi-nation data analytics
journal, December 2014
- Chou, Jui-Sheng; Tsai, Chih-Fong; Pham, Anh-Duc
- Construction and Building Materials, Vol. 73
A framework for coupled deformation-diffusion analysis with application to degradation/healing: A framework for coupled deformation-diffusion analysis
journal, November 2011
- Mudunuru, M. K.; Nakshatrala, K. B.
- International Journal for Numerical Methods in Engineering, Vol. 89, Issue 9
Supervised machine learning on a network scale: application to seismic event classification and detection
journal, May 2017
- Reynen, Andrew; Audet, Pascal
- Geophysical Journal International, Vol. 210, Issue 3
Extremely randomized trees
journal, March 2006
- Geurts, Pierre; Ernst, Damien; Wehenkel, Louis
- Machine Learning, Vol. 63, Issue 1
Boosting a Weak Learning Algorithm by Majority
journal, September 1995
- Freund, Y.
- Information and Computation, Vol. 121, Issue 2
Deep learning and process understanding for data-driven Earth system science
journal, February 2019
- Reichstein, Markus; Camps-Valls, Gustau; Stevens, Bjorn
- Nature, Vol. 566, Issue 7743
Mixing and Reactive Fronts in the Subsurface
journal, September 2019
- Rolle, Massimo; Le Borgne, Tanguy
- Reviews in Mineralogy and Geochemistry, Vol. 85, Issue 1
On enforcing maximum principles and achieving element-wise species balance for advection–diffusion–reaction equations under the finite element method
journal, January 2016
- Mudunuru, M. K.; Nakshatrala, K. B.
- Journal of Computational Physics, Vol. 305
Similarity of fast and slow earthquakes illuminated by machine learning
journal, December 2018
- Hulbert, Claudia; Rouet-Leduc, Bertrand; Johnson, Paul A.
- Nature Geoscience, Vol. 12, Issue 1
An integrated framework that combines machine learning and numerical models to improve wave-condition forecasts
journal, October 2018
- O’Donncha, Fearghal; Zhang, Yushan; Chen, Bei
- Journal of Marine Systems, Vol. 186
Industrial Deployment of Reactive Transport Simulation: An Application to Uranium In situ Recovery
journal, September 2019
- Lagneau, Vincent; Regnault, Olivier; Descostes, Michaël
- Reviews in Mineralogy and Geochemistry, Vol. 85, Issue 1
Advancing Graph-Based Algorithms for Predicting Flow and Transport in Fractured Rock
journal, September 2018
- Viswanathan, H. S.; Hyman, J. D.; Karra, S.
- Water Resources Research, Vol. 54, Issue 9
A new strategy to design eutectic high-entropy alloys using mixing enthalpy
journal, December 2017
- Lu, Yiping; Jiang, Hui; Guo, Sheng
- Intermetallics, Vol. 91
A machine learning approach to geochemical mapping
journal, August 2016
- Kirkwood, Charlie; Cave, Mark; Beamish, David
- Journal of Geochemical Exploration, Vol. 167
Machine learning for data-driven discovery in solid Earth geoscience
journal, March 2019
- Bergen, Karianne J.; Johnson, Paul A.; de Hoop, Maarten V.
- Science, Vol. 363, Issue 6433
A supervised machine-learning approach towards geochemical predictive modelling in archaeology
journal, July 2015
- Oonk, Stijn; Spijker, Job
- Journal of Archaeological Science, Vol. 59
Mapping groundwater contamination risk of multiple aquifers using multi-model ensemble of machine learning algorithms
journal, April 2018
- Barzegar, Rahim; Moghaddam, Asghar Asghari; Deo, Ravinesh
- Science of The Total Environment, Vol. 621
Machine Learning of Mineralization-Related Geochemical Anomalies: A Review of Potential Methods
journal, May 2017
- Zuo, Renguang
- Natural Resources Research, Vol. 26, Issue 4
GPU implementation of neural networks
journal, June 2004
- Oh, Kyoung-Su; Jung, Keechul
- Pattern Recognition, Vol. 37, Issue 6
Multiscale Approaches in Reactive Transport Modeling
journal, September 2019
- Molins, Sergi; Knabner, Peter
- Reviews in Mineralogy and Geochemistry, Vol. 85, Issue 1
A machine learning framework to forecast wave conditions
journal, July 2018
- James, Scott C.; Zhang, Yushan; O'Donncha, Fearghal
- Coastal Engineering, Vol. 137
Methods for interpreting and understanding deep neural networks
journal, February 2018
- Montavon, Grégoire; Samek, Wojciech; Müller, Klaus-Robert
- Digital Signal Processing, Vol. 73
Using Machine Learning to Discern Eruption in Noisy Environments: A Case Study Using CO2‐Driven Cold‐Water Geyser in Chimayó, New Mexico
journal, February 2019
- Yuan, Baichuan; Tan, Yen Joe; Mudunuru, Maruti K.
- Seismological Research Letters, Vol. 90, Issue 2A
Mapping geology and volcanic-hosted massive sulfide alteration in the Hellyer–Mt Charter region, Tasmania, using Random Forests™ and Self-Organising Maps
journal, November 2013
- Cracknell, M. J.; Reading, A. M.; McNeill, A. W.
- Australian Journal of Earth Sciences, Vol. 61, Issue 2
Pore scale study of multiphase multicomponent reactive transport during CO2 dissolution trapping
journal, June 2018
- Chen, Li; Wang, Mengyi; Kang, Qinjun
- Advances in Water Resources, Vol. 116
A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting
journal, August 1997
- Freund, Yoav; Schapire, Robert E.
- Journal of Computer and System Sciences, Vol. 55, Issue 1
Robust system size reduction of discrete fracture networks: a multi-fidelity method that preserves transport characteristics
journal, September 2018
- Srinivasan, Shriram; Hyman, Jeffrey; Karra, Satish
- Computational Geosciences, Vol. 22, Issue 6
Machine Learning Predicts Laboratory Earthquakes: MACHINE LEARNING PREDICTS LAB QUAKES
journal, September 2017
- Rouet-Leduc, Bertrand; Hulbert, Claudia; Lubbers, Nicholas
- Geophysical Research Letters, Vol. 44, Issue 18
Predicting the evolution of fast chemical reactions in chaotic flows
journal, August 2009
- Tsang, Yue-Kin
- Physical Review E, Vol. 80, Issue 2
Mixing, spreading and reaction in heterogeneous media: A brief review
journal, March 2011
- Dentz, Marco; Le Borgne, Tanguy; Englert, Andreas
- Journal of Contaminant Hydrology, Vol. 120-121
On mesh restrictions to satisfy comparison principles, maximum principles, and the non-negative constraint: Recent developments and new results
journal, September 2016
- Mudunuru, Maruti; Nakshatrala, Kalyana Babu
- Mechanics of Advanced Materials and Structures, Vol. 24, Issue 7
Generalized Inverses, Ridge Regression, Biased Linear Estimation, and Nonlinear Estimation
journal, August 1970
- Marquaridt, Donald W.
- Technometrics, Vol. 12, Issue 3
A numerical framework for diffusion-controlled bimolecular-reactive systems to enforce maximum principles and the non-negative constraint
journal, November 2013
- Nakshatrala, K. B.; Mudunuru, M. K.; Valocchi, A. J.
- Journal of Computational Physics, Vol. 253
A Parallel Random Forest Algorithm for Big Data in a Spark Cloud Computing Environment
journal, April 2017
- Chen, Jianguo; Li, Kenli; Tang, Zhuo
- IEEE Transactions on Parallel and Distributed Systems, Vol. 28, Issue 4
Physics-aware Gaussian processes in remote sensing
journal, July 2018
- Camps-Valls, Gustau; Martino, Luca; Svendsen, Daniel H.
- Applied Soft Computing, Vol. 68
Acid Water–Rock–Cement Interaction and Multicomponent Reactive Transport Modeling
journal, September 2019
- Cama, Jordi; Soler, Josep M.; Ayora, Carles
- Reviews in Mineralogy and Geochemistry, Vol. 85, Issue 1
Unsupervised machine learning based on non-negative tensor factorization for analyzing reactive-mixing
journal, October 2019
- Vesselinov, V. V.; Mudunuru, M. K.; Karra, S.
- Journal of Computational Physics, Vol. 395
Machine learning predictive models for mineral prospectivity: An evaluation of neural networks, random forest, regression trees and support vector machines
journal, December 2015
- Rodriguez-Galiano, V.; Sanchez-Castillo, M.; Chica-Olmo, M.
- Ore Geology Reviews, Vol. 71
A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting
journal, August 1997
- Freund, Yoav; Schapire, Robert E.
- Journal of Computer and System Sciences, Vol. 55, Issue 1
On enforcing maximum principles and achieving element-wise species balance for advection–diffusion–reaction equations under the finite element method
journal, January 2016
- Mudunuru, M. K.; Nakshatrala, K. B.
- Journal of Computational Physics, Vol. 305
Ensemble model aggregation using a computationally lightweight machine-learning model to forecast ocean waves
journal, November 2019
- O’Donncha, Fearghal; Zhang, Yushan; Chen, Bei
- Journal of Marine Systems, Vol. 199
A spectral approach to reaction/diffusion kinetics in chaotic flows
journal, January 2002
- Adrover, Alessandra; Cerbelli, Stefano; Giona, Massimiliano
- Computers & Chemical Engineering, Vol. 26, Issue 1
Supervised machine learning on a network scale: application to seismic event classification and detection
journal, May 2017
- Reynen, Andrew; Audet, Pascal
- Geophysical Journal International, Vol. 210, Issue 3
Mixing and Reactive Fronts in the Subsurface
journal, September 2019
- Rolle, Massimo; Le Borgne, Tanguy
- Reviews in Mineralogy and Geochemistry, Vol. 85, Issue 1
A framework for coupling flow and deformation of the porous solid
preprint, January 2013
- Turner, D. Z.; Nakshatrala, K. B.; Martinez, M. J.
- arXiv
API design for machine learning software: experiences from the scikit-learn project
text, January 2013
- Buitinck, Lars; Louppe, Gilles; Blondel, Mathieu
- arXiv
Works referencing / citing this record:
A deep learning modeling framework to capture mixing patterns in reactive-transport systems
text, January 2021
- Jagtap, N. V.; Mudunuru, M. K.; Nakshatrala, K. B.
- arXiv