Unsupervised machine learning based on non-negative tensor factorization for analyzing reactive-mixing
Abstract
Analysis of reactive-diffusion simulations representing complex mixing processes requires a large number of independent model runs. For each high-fidelity model simulation, the model inputs are varied and the predicted mixing behavior is represented by temporal and spatial changes in species concentration. It is then required to discern how the model inputs (such as diffusivity, dispersion, anisotropy, and velocity field properties) impact the mixing process. This task is challenging and typically involves interpretation of large model outputs representing temporal and spatial changes of species concentration within the model domain. Yet, the task can be automated and substantially simplified by applying Machine Learning (ML) methods. Here, we present an application of an unsupervised ML method (called NTFk) using Non-negative Tensor Factorization (NTF) coupled with a custom clustering procedure based on k-means to reveal the temporal and spatial features in product concentrations. An attractive and unique aspect of the proposed ML method is that it ensures the extracted features are non-negative, which is important to obtain a meaningful deconstruction of the mixing processes. The ML methodology is applied to a large set of high-resolution finite-element model simulations representing anisotropic reaction-diffusion processes in perturbed vortex-based velocity fields. The applied finite-element method ensures that spatialmore »
- Authors:
-
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Univ. of Maryland Baltimore County (UMBC), Baltimore, MD (United States)
- Publication Date:
- Research Org.:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Org.:
- USDOE National Nuclear Security Administration (NNSA), Office of Defense Nuclear Nonproliferation
- OSTI Identifier:
- 1529535
- Alternate Identifier(s):
- OSTI ID: 1527034
- Report Number(s):
- LA-UR-18-20868
Journal ID: ISSN 0021-9991
- Grant/Contract Number:
- 89233218CNA000001; AC52-06NA25396
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Journal of Computational Physics
- Additional Journal Information:
- Journal Volume: 395; Journal Issue: C; Journal ID: ISSN 0021-9991
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; Non-negative tensor factorization; Unsupervised machine learning; Structure-preserving feature extraction; Reactive-mixing; Anisotropic dispersion; Non-negative solutions
Citation Formats
Vesselinov, V. V., Mudunuru, M. K., Karra, S., O'Malley, D., and Alexandrov, B. S. Unsupervised machine learning based on non-negative tensor factorization for analyzing reactive-mixing. United States: N. p., 2019.
Web. doi:10.1016/j.jcp.2019.05.039.
Vesselinov, V. V., Mudunuru, M. K., Karra, S., O'Malley, D., & Alexandrov, B. S. Unsupervised machine learning based on non-negative tensor factorization for analyzing reactive-mixing. United States. https://doi.org/10.1016/j.jcp.2019.05.039
Vesselinov, V. V., Mudunuru, M. K., Karra, S., O'Malley, D., and Alexandrov, B. S. Tue .
"Unsupervised machine learning based on non-negative tensor factorization for analyzing reactive-mixing". United States. https://doi.org/10.1016/j.jcp.2019.05.039. https://www.osti.gov/servlets/purl/1529535.
@article{osti_1529535,
title = {Unsupervised machine learning based on non-negative tensor factorization for analyzing reactive-mixing},
author = {Vesselinov, V. V. and Mudunuru, M. K. and Karra, S. and O'Malley, D. and Alexandrov, B. S.},
abstractNote = {Analysis of reactive-diffusion simulations representing complex mixing processes requires a large number of independent model runs. For each high-fidelity model simulation, the model inputs are varied and the predicted mixing behavior is represented by temporal and spatial changes in species concentration. It is then required to discern how the model inputs (such as diffusivity, dispersion, anisotropy, and velocity field properties) impact the mixing process. This task is challenging and typically involves interpretation of large model outputs representing temporal and spatial changes of species concentration within the model domain. Yet, the task can be automated and substantially simplified by applying Machine Learning (ML) methods. Here, we present an application of an unsupervised ML method (called NTFk) using Non-negative Tensor Factorization (NTF) coupled with a custom clustering procedure based on k-means to reveal the temporal and spatial features in product concentrations. An attractive and unique aspect of the proposed ML method is that it ensures the extracted features are non-negative, which is important to obtain a meaningful deconstruction of the mixing processes. The ML methodology is applied to a large set of high-resolution finite-element model simulations representing anisotropic reaction-diffusion processes in perturbed vortex-based velocity fields. The applied finite-element method ensures that spatial and temporal species concentration are always non-negative, even in the case of high anisotropic contrasts. The simulated reaction is a fast irreversible bimolecular reaction A+B→C , where species A and B react to form species C. The reactive-diffusion model input parameters that control mixing include properties of the velocity field (such as vortex structures), anisotropic dispersion, and molecular diffusion. We demonstrate the applicability of the ML feature extraction method to produce a meaningful deconstruction of model outputs to discriminate between different physical processes impacting the reactants, their mixing, and the spatial distribution of the product C. The introduced ML analysis allowed us to identify additive temporal and spatial features that characterize mixing behavior. The application of the proposed NTFk approach is not limited to reactive-mixing. NTFk can be readily applied to any observed or simulated datasets that can be represented as tensors (multi-dimensional arrays) and have separable latent signatures or features.},
doi = {10.1016/j.jcp.2019.05.039},
journal = {Journal of Computational Physics},
number = C,
volume = 395,
place = {United States},
year = {Tue Oct 15 00:00:00 EDT 2019},
month = {Tue Oct 15 00:00:00 EDT 2019}
}
Web of Science
Figures / Tables:
Works referenced in this record:
Effect of inhomogeneous mixing on atmospheric photochemical reactions
journal, September 1972
- Donaldson, Coleman duP.; Hilst, Glenn R.
- Environmental Science & Technology, Vol. 6, Issue 9
Mixing, Chaotic Advection, and Turbulence
journal, January 1990
- Ottino, J. M.
- Annual Review of Fluid Mechanics, Vol. 22, Issue 1
On the Nature of Turbulence in a Stratified Fluid. Part I: The Energetics of Mixing
journal, May 1991
- Ivey, G. N.; Imberger, J.
- Journal of Physical Oceanography, Vol. 21, Issue 5
Experimental and theoretical scaling laws for transverse diffusive broadening in two-phase laminar flows in microchannels
journal, April 2000
- Ismagilov, Rustem F.; Stroock, Abraham D.; Kenis, Paul J. A.
- Applied Physics Letters, Vol. 76, Issue 17
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
Micromixers—a review on passive and active mixing principles
journal, April 2005
- Hessel, Volker; Löwe, Holger; Schönfeld, Friedhelm
- Chemical Engineering Science, Vol. 60, Issue 8-9, p. 2479-2501
Evaluation of the Effects of Porous Media Structure on Mixing-Controlled Reactions Using Pore-Scale Modeling and Micromodel Experiments
journal, May 2008
- Willingham, Thomas W.; Werth, Charles J.; Valocchi, Albert J.
- Environmental Science & Technology, Vol. 42, Issue 9
A Critical Review of the Risks to Water Resources from Unconventional Shale Gas Development and Hydraulic Fracturing in the United States
journal, March 2014
- Vengosh, Avner; Jackson, Robert B.; Warner, Nathaniel
- Environmental Science & Technology, Vol. 48, Issue 15
Simultaneous parameter estimation and contaminant source characterization for coupled groundwater flow and contaminant transport modelling
journal, July 1992
- Wagner, Brian J.
- Journal of Hydrology, Vol. 135, Issue 1-4
Comparison of inverse methods for reconstructing the release history of a groundwater contamination source
journal, September 2000
- Neupauer, Roseanna M.; Borchers, Brian; Wilson, John L.
- Water Resources Research, Vol. 36, Issue 9
Pollution source identification in heterogeneous porous media
journal, August 2001
- Atmadja, Juliana; Bagtzoglou, Amvrossios C.
- Water Resources Research, Vol. 37, Issue 8
Estimation of historical groundwater contaminant distribution using the adjoint state method applied to geostatistical inverse modeling: ESTIMATION OF GROUNDWATER CONTAMINANT DISTRIBUTION
journal, August 2004
- Michalak, Anna M.; Kitanidis, Peter K.
- Water Resources Research, Vol. 40, Issue 8
Identification of Contaminant Sources in Water Distribution Systems Using Simulation–Optimization Method: Case Study
journal, July 2006
- Guan, Jiabao; Aral, Mustafa M.; Maslia, Morris L.
- Journal of Water Resources Planning and Management, Vol. 132, Issue 4
Point source identification in nonlinear advection–diffusion–reaction systems
journal, March 2013
- Mamonov, A. V.; Tsai, Y-H R.
- Inverse Problems, Vol. 29, Issue 3
Inverse source problem in a one-dimensional evolution linear transport equation with spatially varying coefficients: application to surface water pollution
journal, September 2013
- Hamdi, Adel; Mahfoudhi, Imed
- Inverse Problems in Science and Engineering, Vol. 21, Issue 6
Identification of a time-dependent source term in nonlinear hyperbolic or parabolic heat equation
journal, December 2015
- Borukhov, V. T.; Zayats, G. M.
- International Journal of Heat and Mass Transfer, Vol. 91
Introduction: mixing in microfluidics
journal, May 2004
- Ottino, Julio M.; Wiggins, Stephen
- Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, Vol. 362, Issue 1818
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
Large-Scale Optimization-Based Non-negative Computational Framework for Diffusion Equations: Parallel Implementation and Performance Studies
journal, July 2016
- Chang, J.; Karra, S.; Nakshatrala, K. B.
- Journal of Scientific Computing, Vol. 70, Issue 1
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
Predicting the evolution of fast chemical reactions in chaotic flows
journal, August 2009
- Tsang, Yue-Kin
- Physical Review E, Vol. 80, Issue 2
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
Application of Nonnegative Tensor Factorization for neutron-gamma discrimination of Monte Carlo simulated fission chamber’s output signals
journal, January 2017
- Laassiri, Mounia; Hamzaoui, El-Mehdi; Cherkaoui El Moursli, Rajaa
- Results in Physics, Vol. 7
Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values
journal, June 1994
- Paatero, Pentti; Tapper, Unto
- Environmetrics, Vol. 5, Issue 2
The Cocktail Party Problem
journal, September 2005
- Haykin, Simon; Chen, Zhe
- Neural Computation, Vol. 17, Issue 9
PARAFAC: Parallel factor analysis
journal, August 1994
- Harshman, Richard A.; Lundy, Margaret E.
- Computational Statistics & Data Analysis, Vol. 18, Issue 1
Some mathematical notes on three-mode factor analysis
journal, September 1966
- Tucker, Ledyard R.
- Psychometrika, Vol. 31, Issue 3
The N-way Toolbox for MATLAB
journal, August 2000
- Andersson, Claus A.; Bro, Rasmus
- Chemometrics and Intelligent Laboratory Systems, Vol. 52, Issue 1
Unsupervised pattern-recognition techniques to investigate metal pollution in estuaries
journal, May 2013
- Gredilla, A.; Fdez-Ortiz de Vallejuelo, S.; de Diego, A.
- TrAC Trends in Analytical Chemistry, Vol. 46
Algorithms for Sparse Nonnegative Tucker Decompositions
journal, August 2008
- Mørup, Morten; Hansen, Lars Kai; Arnfred, Sidse M.
- Neural Computation, Vol. 20, Issue 8
Learning the parts of objects by non-negative matrix factorization
journal, October 1999
- Lee, Daniel D.; Seung, H. Sebastian
- Nature, Vol. 401, Issue 6755
A Block Coordinate Descent Method for Regularized Multiconvex Optimization with Applications to Nonnegative Tensor Factorization and Completion
journal, January 2013
- Xu, Yangyang; Yin, Wotao
- SIAM Journal on Imaging Sciences, Vol. 6, Issue 3
Alternating proximal gradient method for sparse nonnegative Tucker decomposition
journal, May 2014
- Xu, Yangyang
- Mathematical Programming Computation, Vol. 7, Issue 1
Blind source separation for groundwater pressure analysis based on nonnegative matrix factorization
journal, September 2014
- Alexandrov, Boian S.; Vesselinov, Velimir V.
- Water Resources Research, Vol. 50, Issue 9
Silhouettes: A graphical aid to the interpretation and validation of cluster analysis
journal, November 1987
- Rousseeuw, Peter J.
- Journal of Computational and Applied Mathematics, Vol. 20
Nonnegative Matrix Factorization for identification of unknown number of sources emitting delayed signals
journal, March 2018
- Iliev, Filip L.; Stanev, Valentin G.; Vesselinov, Velimir V.
- PLOS ONE, Vol. 13, Issue 3
Identification of release sources in advection–diffusion system by machine learning combined with Green’s function inverse method
journal, August 2018
- Stanev, Valentin G.; Iliev, Filip L.; Hansen, Scott
- Applied Mathematical Modelling, Vol. 60
Contaminant source identification using semi-supervised machine learning
journal, May 2018
- Vesselinov, Velimir V.; Alexandrov, Boian S.; O’Malley, Daniel
- Journal of Contaminant Hydrology, Vol. 212
Silhouettes: A graphical aid to the interpretation and validation of cluster analysis
journal, November 1987
- Rousseeuw, Peter J.
- Journal of Computational and Applied Mathematics, Vol. 20
Contaminant source identification using semi-supervised machine learning
journal, May 2018
- Vesselinov, Velimir V.; Alexandrov, Boian S.; O’Malley, Daniel
- Journal of Contaminant Hydrology, Vol. 212
A Multilinear Singular Value Decomposition
journal, January 2000
- De Lathauwer, Lieven; De Moor, Bart; Vandewalle, Joos
- SIAM Journal on Matrix Analysis and Applications, Vol. 21, Issue 4
Mixing, Chaotic Advection, and Turbulence
journal, January 1990
- Ottino, J. M.
- Annual Review of Fluid Mechanics, Vol. 22, Issue 1
Algorithms for Sparse Nonnegative Tucker Decompositions
journal, August 2008
- Mørup, Morten; Hansen, Lars Kai; Arnfred, Sidse M.
- Neural Computation, Vol. 20, Issue 8
Numerical methods for convection-diffusion problems or The 30 years war
preprint, January 2013
- Stynes, Martin
- arXiv
Figures / Tables found in this record: