Search NASA⌕ Search

SEARCH · Search NASA

Results for “dynamic parameter estimation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 595 records · Page 33
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Communication: Molecular dynamics and {sup 1}H NMR of n-hexane in liquid crystals

The NMR spectrum of n-hexane orientationally ordered in the nematic liquid crystal ZLI-1132 is analysed using covariance matrix adaptation evolution strategy (CMA-ES). The spectrum contains over 150 000 transitions, with many sharp features appearing above a broad, underlying background signal that results from the plethora of overlapping transitions from the n-hexane as well as from the liquid crystal. The CMA-ES requires initial search ranges for NMR spectral parameters, notably the direct dipolar couplings. Several sets of such ranges were utilized, including three from MD simulations and others from the modified chord model that is specifically designed to predict hydrocarbon-chain dipolar couplings. In the end, only inaccurate dipolar couplings from an earlier study utilizing proton-proton double quantum 2D-NMR techniques on partially deuterated n-hexane provided the necessary estimates. The precise set of dipolar couplings obtained can now be used to investigate conformational averaging of n-hexane in a nematic environment.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2013

The Gottingen Minipig Is a Model of the Hematopoietic Acute Radiation Syndrome: G-Colony Stimulating Factor Stimulates Hematopoiesis and Enhances Survival From Lethal Total-Body γ-Irradiation

Purpose: We are characterizing the Gottingen minipig as an additional large animal model for advanced drug testing for the acute radiation syndrome (ARS) to enhance the discovery and development of novel radiation countermeasures. Among the advantages provided by this model, the similarities to human hematologic parameters and dynamics of cell loss/recovery after irradiation provide a convenient means to compare the efficacy of drugs known to affect bone marrow cellularity and hematopoiesis. Methods and Materials: Male Gottingen minipigs, 4 to 5 months old and weighing 9 to 11 kg, were used for this study. We tested the standard off-label treatment for ARS, rhG-CSF (Neupogen, 10 μg/kg/day for 17 days), at the estimated LD70/30 total-body γ-irradiation (TBI) radiation dose for the hematopoietic syndrome, starting 24 hours after irradiation. Results: The results indicated that granulocyte colony stimulating factor (G-CSF) enhanced survival, stimulated recovery from neutropenia, and induced mobilization of hematopoietic progenitor cells. In addition, the administration of G-CSF resulted in maturation of monocytes/macrophages. Conclusions: These results support continuing efforts toward validation of the minipig as a large animal model for advanced testing of radiation countermeasures and characterization of the pathophysiology of ARS, and they suggest that the efficacy of G-CSF in improving survival after total body irradiation may involve mechanisms other than increasing the numbers of circulating granulocytes.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗
DOE OSTI2022

Quantum criticality using a superconducting quantum processor

Quantum criticality emerges from the collective behavior of many interacting quantum particles, often at the transition between different phases of matter. It is one of the cornerstones of condensed matter physics, which we access on noisy intermediate-scale (NISQ) quantum devices by leveraging a dynamically driven phenomenon. Here we probe the critical properties of the one-dimensional quantum Ising model on a programmable superconducting quantum chip via a Kibble-Zurek process, obtain scaling laws, and estimate critical exponents despite inherent sources of errors on the hardware. In addition, we investigate how the improvement of NISQ computers (more qubits, less noise) will consolidate the computation of those universal physical properties. A one-parameter noise model captures the effect of imperfections and reproduces the experimental data. Its systematic study reveals that the noise, analogously to temperature, induces a new length scale in the system. We introduce and successfully verify modified scaling laws, directly accounting for the noise without any prior knowledge. It makes data analyses for extracting physical properties transparent to noise. By understanding how imperfect quantum hardware modifies the genuine properties of quantum states of matter, we enhance the power of NISQ processors considerably for addressing quantum criticality and potentially other phenomena and algorithms.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2022

Fed-DeepONet: Stochastic Gradient-Based Federated Training of Deep Operator Networks

The Deep Operator Network (DeepONet) framework is a different class of neural network architecture that one trains to learn nonlinear operators, i.e., mappings between infinite-dimensional spaces. Traditionally, DeepONets are trained using a centralized strategy that requires transferring the training data to a centralized location. Such a strategy, however, limits our ability to secure data privacy or use high-performance distributed/parallel computing platforms. To alleviate such limitations, in this paper, we study the federated training of DeepONets for the first time. That is, we develop a framework, which we refer to as Fed-DeepONet, that allows multiple clients to train DeepONets collaboratively under the coordination of a centralized server. To achieve Fed-DeepONets, we propose an efficient stochastic gradient-based algorithm that enables the distributed optimization of the DeepONet parameters by averaging first-order estimates of the DeepONet loss gradient. Then, to accelerate the training convergence of Fed-DeepONets, we propose a moment-enhanced (i.e., adaptive) stochastic gradient-based strategy. Finally, we verify the performance of Fed-DeepONet by learning, for different configurations of the number of clients and fractions of available clients, (i) the solution operator of a gravity pendulum and (ii) the dynamic response of a parametric library of pendulums.

Moya, Christian↗
DOE OSTI2016

ON HELIUM MIXING IN QUASI-GLOBAL SIMULATIONS OF THE INTRACLUSTER MEDIUM

The assumption of a spatially uniform helium distribution in the intracluster medium (ICM) can lead to biases in the estimates of key cluster parameters if composition gradients are present. The helium concentration profile in galaxy clusters is unfortunately not directly observable. Current models addressing the putative sedimentation are one-dimensional and parametrize the presence of magnetic fields in a crude way, ignoring the weakly collisional, magnetized nature of the medium. When these effects are considered, a wide variety of instabilities can play an important role in the plasma dynamics. In a series of recent papers, we have developed the local, linear theory of these instabilities and addressed their nonlinear development with a modified version of Athena. Here, we extend our study by developing a quasi-global approach that we use to simulate the mixing of helium as induced by generalizations of the heat-flux-driven buoyancy instability (HBI) and the magnetothermal instability, which feed off thermal and composition gradients. In the inner region of the ICM, mixing can occur over a few gigayears, after which the average magnetic field inclination angle is ∼30°–50°, resulting in an averaged Spitzer parameter higher by about 20% than the value obtained in homogeneous simulations. In the cluster outskirts the instabilities are rather inefficient, due to the shallow gradients. This suggests that composition gradients in cluster cores might be shallower than one-dimensional models predict. More quantitative statements demand more refined models that can incorporate the physics driving the sedimentation process and simultaneously account for the weakly collisional nature of the plasma.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2018

Galaxy Cluster Mass Reconstruction Project – IV. Understanding the effects of imperfect membership on cluster mass estimation

Here, the primary difficulty in measuring dynamical masses of galaxy clusters from galaxy data lies in the separation between true cluster members from interloping galaxies along the line of sight. We study the impact of membership contamination and incompleteness on cluster mass estimates obtained with 25 commonly used techniques applied to nearly 1000 mock clusters with precise spectroscopic redshifts. We show that all methods overestimate or underestimate cluster masses when applied to contaminated or incomplete galaxy samples, respectively. This appears to be the main source of the intrinsic scatter in the mass scaling relation. Applying corrections based on a prior knowledge of contamination and incompleteness can reduce the scatter to the level of shot noise expected for poorly sampled clusters. We establish an empirical model quantifying the effect of imperfect membership on cluster mass estimation and discuss its universal and method-dependent features. We find that both imperfect membership and the response of the mass estimators depend on cluster mass, effectively causing a flattening of the estimated–true mass relation. Imperfect membership thus alters cluster counts determined from spectroscopic surveys, hence the cosmological parameters that depend on such counts.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Paramagnon dispersion and damping in doped Na x Ca 2 – x CuO 2 Cl 2

Here, using resonant inelastic x-ray scattering, we measure the paramagnon dispersion and damping of undoped, antiferromagnetic Ca 2 CuO 2 Cl 2 as well as doped, superconducting Na x Ca 2–x CuO 2 Cl 2 . Our estimation of the spin-exchange parameter and width of the paramagnon peak at the zone boundary X=(0.5, 0) confirms that no simple relation can be drawn between these parameters and the critical temperature T c . Consistently with other cuprate compounds, we show that upon doping there is a slight softening at (0.25,0) but not at the zone boundary X. In combination with these measurements we perform calculations of the dynamical spin structure factor of the one-band Hubbard model using cluster dynamical mean-field theory. The calculations are in excellent agreement with the experiment in the undoped case, both in terms of energy position and width. While the increase in width is also captured upon doping, the dynamical spin structure factor shows a sizable softening at X, which provides insightful information on the length-scale of the spin fluctuations in doped cuprates.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2019

Quantification of MagLIF Morphology using the Mallat Scattering Transformation

The morphology of the stagnated plasma resulting from Magnetized Liner Inertial Fusion (MagLIF) is measured by imaging the self-emission x-rays coming from the multi-keV plasma, and the evolution of the imploding liner is measured by radiographs. Equivalent diagnostic response can be derived from integrated rad-MHD simulations from programs such as Hydra and Gorgon. There have been only limited quantitative ways to compare the image morphology, that is the texture, of simulations and experiments. We have developed a metric of image morphology based on the Mallat Scattering Transformation (MST), a transformation that has proved to be effective at distinguishing textures, sounds, and written characters. This metric has demonstrated excellent performance in classifying ensembles of synthetic stagnation images. We use this metric to quantitatively compare simulations to experimental images, cross experimental images, and to estimate the parameters of the images with uncertainty via a linear regression of the synthetic images to the parameter used to generate them. This coordinate space has proved very adept at doing a sophisticated relative back-ground subtraction in the MST space. This was needed to compare the experimental self emission images to the rad-MHD simulation images. We have also developed theory that connects the transformation to the causal dynamics of physical systems. This has been done from the classical kinetic perspective and from the field theory perspective, where the MST is the generalized Green's function, or S-matrix of the field theory in the scale basis. From both perspectives the first order MST is the current state of the system, and the second order MST are the transition rates from one state to another. An efficient, GPU accelerated, Python implementation of the MST was developed. Future applications are discussed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2016

Determination of Coronal Mass Ejection physical parameters from a combination of polarized visible light and UV Lyα observations

Visible-light observations of Coronal Mass Ejections (CMEs) performed with coronagraphs and heliospheric imagers (in primis on board the Solar and Heliospheric Observatory and STEREO missions) have offered the best way to study the kinematics and geometrical structure of these fundamental events so far. Nevertheless, it has been widely demonstrated that only combination of multi-wavelength data (including X-ray spectra, EUV images, EUV-UV spectra, and radio dynamic spectra) can provide complete information on the plasma temperature and density distributions, non-thermal motions, magnetic fields, and other physical parameters, for both CMEs and CME-related phenomena. In this work, we analyze three CMEs by combining simultaneous data acquired in the polarized visible light by the LASCO-C2 coronagraph and in the UV H i Lyα line (1216 Å) by the UVCS spectrometer, in order to estimate the CME plasma electron density (using the polarization-ratio technique to infer the 3D structure of the CME) and temperature (from the comparison between the expected and measured Lyα intensities) along the UVCS field of view. This analysis is primarily aimed at testing the diagnostic methods that will be applied to coronagraphic observations of CMEs delivered by the Metis instrument on board the next ESA-Solar Orbiter mission. We find that CME cores are usually associated with cooler plasma (T∼10{sup 6} K), and that a significant increase of the electron temperatures is observed from the core to the front of the CME (where T>10{sup 6.3} K), which seems to be correlated, in all cases, with the morphological structure of the CME as derived from visible-light images.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Evaluating the effect of alternative carbon allocation schemes in a land surface model (CLM4.5) on carbon fluxes, pools, and turnover in temperate forests

Abstract. How carbon (C) is allocated to different plant tissues (leaves, stem, and roots) determines how long C remains in plant biomass and thus remains a central challenge for understanding the global C cycle. We used a diverse set of observations (AmeriFlux eddy covariance tower observations, biomass estimates from tree-ring data, and leaf area index (LAI) measurements) to compare C fluxes, pools, and LAI data with those predicted by a land surface model (LSM), the Community Land Model (CLM4.5). We ran CLM4.5 for nine temperate (including evergreen and deciduous) forests in North America between 1980 and 2013 using four different C allocation schemes: i. dynamic C allocation scheme (named "D-CLM4.5") with one dynamic allometric parameter, which allocates C to the stem and leaves to vary in time as a function of annual net primary production (NPP); ii. an alternative dynamic C allocation scheme (named "D-Litton"), where, similar to (i), C allocation is a dynamic function of annual NPP, but unlike (i) includes two dynamic allometric parameters involving allocation to leaves, stem, and coarse roots; iii.–iv. a fixed C allocation scheme with two variants, one representative of observations in evergreen (named "F-Evergreen") and the other of observations in deciduous forests (named "F-Deciduous"). D-CLM4.5 generally overestimated gross primary production (GPP) and ecosystem respiration, and underestimated net ecosystem exchange (NEE). In D-CLM4.5, initial aboveground biomass in 1980 was largely overestimated (between 10 527 and 12 897 g C m −2 ) for deciduous forests, whereas aboveground biomass accumulation through time (between 1980 and 2011) was highly underestimated (between 1222 and 7557 g C m −2 ) for both evergreen and deciduous sites due to a lower stem turnover rate in the sites than the one used in the model. D-CLM4.5 overestimated LAI in both evergreen and deciduous sites because the leaf C–LAI relationship in the model did not match the observed leaf C–LAI relationship at our sites. Although the four C allocation schemes gave similar results for aggregated C fluxes, they translated to important differences in long-term aboveground biomass accumulation and aboveground NPP. For deciduous forests, D-Litton gave more realistic C stem ∕ C leaf ratios and strongly reduced the overestimation of initial aboveground biomass and aboveground NPP for deciduous forests by D-CLM4.5. We identified key structural and parameterization deficits that need refinement to improve the accuracy of LSMs in the near future. These include changing how C is allocated in fixed and dynamic schemes based on data from current forest syntheses and different parameterization of allocation schemes for different forest types. Our results highlight the utility of using measurements of aboveground biomass to evaluate and constrain the C allocation scheme in LSMs, and suggest that stem turnover is overestimated by CLM4.5 for these AmeriFlux sites. Understanding the controls of turnover will be critical to improving long-term C processes in LSMs.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2013

Is Dynamic Autocrine Insulin Signaling Possible? A Mathematical Model Predicts Picomolar Concentrations of Extracellular Monomeric Insulin within Human Pancreatic Islets

Insulin signaling is essential for β-cell survival and proliferation in vivo. Insulin also has potent mitogenic and anti-apoptotic actions on cultured β-cells, with maximum effect in the high picomolar range and diminishing effect at high nanomolar doses. In order to understand whether these effects of insulin are constitutive or can be subjected to physiological modulation, it is essential to estimate the extracellular concentration of monomeric insulin within an intact islet. Unfortunately, the in vivo concentration of insulin monomers within the islet cannot be measured directly with current technology. Here, we present the first mathematical model designed to estimate the levels of monomeric insulin within the islet extracellular space. Insulin is released as insoluble crystals that exhibit a delayed dissociation into hexamers, dimers, and eventually monomers, which only then can act as signaling ligands. The rates at which different forms of insulin dissolve in vivo have been estimated from studies of peripheral insulin injection sites. We used this and other information to formulate a mathematical model to estimate the local insulin concentration within a single islet as a function of glucose. Model parameters were estimated from existing literature. Components of the model were validated using experimental data, if available. Model analysis predicted that the majority of monomeric insulin in the islet is that which has been returned from the periphery, and the concentration of intra-islet monomeric insulin varies from β50–300 pM when glucose is in the physiological range. Thus, our results suggest that the local concentration of monomeric insulin within the islet is in the picomolar ‘sweet spot’ range of insulin doses that activate the insulin receptor and have the most potent effects on β-cells in vitro. Together with experimental data, these estimations support the concept that autocrine/paracrine insulin signalling within the islet is dynamic, rather than constitutive and saturated.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Leaf 13 C data constrain the uncertainty of the carbon dynamics of temperate forest ecosystems

Stable carbon isotope discrimination occurred in plant biophysical and biogeochemical processes can help understand plant physiology and soil biogeochemistry with respect to carbon cycling. Here, we incorporated the photosynthetic carbon isotope discrimination into a process-based land surface model (iTEM) to test if stable carbon isotope composition (δ 13 C) can impose additional constraint on model parameters. Sequential data assimilation was implemented at six eddy covariance flux tower sites using carbon flux observations including gross primary productivity (GPP) and net ecosystem exchange (NEE) with and without considering foliar δ 13 C (δ 13 C f ) measurement constraints, respectively. Our model-data fusion showed that δ 13 C f can provide useful constraint on photosynthetic (V cmax25 , the maximum rate of carboxylation at 25°C) and stomatal (g1, the slope of stomatal function) parameters as well as the posterior carbon fluxes. When including δ 13 C f measurement, g1 spatially varies among the six sites and is significantly correlated with annual precipitation. We incorporated the statistical relationship between g1 and annual precipitation into iTEM, which is then used to quantify the regional carbon dynamic in temperate forest ecosystems of the Northern Hemisphere. Compared with the simulation only conditioned on carbon flux observations, regional carbon flux estimations performed slightly better against the FLUXCOM products and the uncertainties of modeled carbon fluxes were reduced by 27%. Our study demonstrated that δ 13 C f data constrains carbon flux uncertainties across space.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2012

A Phase-Field Model for Phase Transformations in Glass-Forming Alloys

A phase-field model is proposed for phase transformations in glass-forming alloys. The glass transition is introduced as a structural relaxation, and the competition between the glass and crystalline phases is investigated. The simulations are performed for Cu-Zr alloys, employing thermodynamic and kinetic parameters derived from reported thermodynamic modeling and molecular dynamics simulation results,[1–3] respectively. Four distinct phase fields are treated with a multi-phase-field approach, representing the liquid/glass, Cu10Zr7, CuZr, and CuZr2 phases. In addition, a continuum-field method is applied to the liquid to accommodate the liquid–glass transformation. The combined phase-field approach is used to investigate the glass formation tendency, and critical cooling rates are estimated and compared with the reported experimental values.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Quantifying spatiotemporal variability in occupant exposure to an indoor airborne contaminant with an uncertain source location

Well-mixed zone models are often employed to compute indoor air quality and occupant exposures. While effective, a potential downside to assuming instantaneous, perfect mixing is underpredicting exposures to high intermittent concentrations within a room. When such cases are of concern, more spatially resolved models, like computational-fluid dynamics methods, are used for some or all of the zones. But, these models have higher computational costs and require more input information. A preferred compromise would be to continue with a multi-zone modeling approach for all rooms, but with a better assessment of the spatial variability within a room. Here to do so, we present a quantitative method for estimating a room's spatiotemporal variability, based on influential room parameters. Our proposed method disaggregates variability into the variability in a room's average concentration, and the spatial variability within the room relative to that average. This enables a detailed assessment of how variability in particular room parameters impacts the uncertain occupant exposures. To demonstrate the utility of this method, we simulate contaminant dispersion for a variety of possible source locations. We compute breathing-zone exposure during the releasing (source is active) and decaying (source is removed) periods. Using CFD methods, we found after a 30 minutes release the average standard deviation in the spatial distribution of exposure was approximately 28% of the source average exposure, whereas variability in the different average exposures was lower, only 10% of the total average. We also find that although uncertainty in the source location leads to variability in the average magnitude of transient exposure, it does not have a particularly large influence on the spatial distribution during the decaying period, or on the average contaminant removal rate. By systematically characterizing a room's average concentration, its variability, and the spatial variability within the room important insights can be gained as to how much uncertainty is introduced into occupant exposure predictions by assuming a uniform in-room contaminant concentration. We discuss how the results of these characterizations can improve our understanding of the uncertainty in occupant exposures relative to well-mixed models.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2013

THE MASS OF THE MILKY WAY AND M31 USING THE METHOD OF LEAST ACTION

We constrain the most likely range of masses for the Milky Way (MW) and M31 using an application of the numerical action method (NAM) that optimizes the fit to observed parameters over a large ensemble of NAM-generated solutions. Our 95% confidence level mass ranges, 1.5-4.5 × 10{sup 12} M{sub ☉} for MW and 1.5-5.5 × 10{sup 12} M{sub ☉} for M31, are consistent with the upper range of estimates from other methods and suggests that a larger proportion of the total mass becomes detectable when the peculiar motions of many nearby satellites are taken into account in the dynamical analysis. We test the method against simulated Local Group catalogs extracted from the Millennium Run to confirm that mass predictions are consistent with actual galaxy halo masses.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

An economics-by-design approach to a radiant integrated thermophotovoltaic microreactor system

This paper presents an application of the economics-by-design approach to the Radiant Integrated Thermo-photovoltaic Microreactor System (RITMS). The RITMS design is unique in that it directly couples a critical fission reactor with thermo-photovoltaic (TPV) panels for high efficiency energy conversion. This significant shift from electric conversion using traditional dynamic heat pumps leads to a simpler and more reliable system without turbomachinery and high pressurization. In working towards wrapping up the early design work, the economics-by-design approach, which centers economic competitiveness as the optimization parameter, is well suited in making a final determination on viability. This paper describes the computational sequence that was developed to couple the radiative and conductive heat transfer and feed operational performance to cost estimation. The framework was applied to maximize the power of the system, while minimizing the fuel enrichment. This method is applied to a reference RITMS design as part of a parametric sensitivity study, which revealed that the system is under moderated and that single unit plants could produce power as low as 300 dollar/MWh. This price point supports the notion that early in the RITMS design implementation, adoption into niche markets as a first-of-a-kind technology is possible. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2019

Statistical Analysis of the Eigenmode Spectrum in the SRF Cavities with Mechanical Imperfections

The superconducting radio frequency (SRF) technology is progressing rapidly over last decades toward high accelerating gradients and low surface resistance making feasible the particle accelerators operation with high beam currents and long duty factors. However, the coherent RF losses due to high order modes (HOMs) excitation becomes a limiting factor for these regimes. In spite of the operating mode, which is tuned separately, the parameters of HOMs vary from one cavity to another due to finite mechanical tolerances during cavities fabrication. It is vital to know in advance the spread of HOM parameters in order to predict unexpected cryogenic losses, overheating of beam line components and to keep stable beam dynamics. In this paper we present the method of generating the unique cavity geometry with imperfections while preserving operating mode frequency and field flatness. Based on the eigenmode spectrum calculation of series of randomly generated cavities we can accumulate the data for the evaluation the HOM statistics. Finally, we describe the procedure for the estimation of the probability of the resonant HOM losses in the SRF resonators. The study of these effects leads to specifications of SC cavity and cryomodule and can significantly impact on the efficiency and reliability of the machine operation

43 PARTICLE ACCELERATORS↗
DOE OSTI2024

Comparative study of machine learning techniques for post-combustion carbon capture systems

Computational analysis of countercurrent flows in packed absorption columns, often used in solvent-based post-combustion carbon capture systems (CCSs), is challenging. Typically, computational fluid dynamics (CFD) approaches are used to simulate the interactions between a solvent, gas, and column's packing geometry while accounting for the thermodynamics, kinetics, heat, and mass transfer effects of the absorption process. These simulations can then be used explain a column's hydrodynamic characteristics and evaluate its CO 2 -capture efficiency. However, these approaches are computationally expensive, making it difficult to evaluate numerous designs and operating conditions to improve efficiency at industrial scales. In this work, we comprehensively explore the application of statistical ML methods, convolutional neural networks (CNNs), and graph neural networks (GNNs) to aid and accelerate the scale-up and design optimization of solvent-based post-combustion CCSs. We apply these methods to CFD datasets of countercurrent flows in absorption columns with structured packings characterized by several geometric parameters. We train models to use these parameters, inlet velocity conditions, and other model-specific representations of the column to estimate key determinants of CO 2 -capture efficiency without having to simulate additional CFD datasets. We also evaluate the impact of different input types on the accuracy and generalizability of each model. We discuss the strengths and limitations of each approach to further elucidate the role of CNNs, GNNs, and other machine learning approaches for CO 2 -capture property prediction and design optimization.

97 MATHEMATICS AND COMPUTING↗