Search NASA⌕ Search

SEARCH · Search NASA

Results for “model interpretability”

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 1,495 records · Page 83

On Transferability of Performance Metrics for Redox-Active Molecules

Because of their lower material costs, redox-active organic molecules (ROMs) are increasingly used in flow cells. Quantification of their electrochemical stability is important for choosing between different battery chemistries. Ad hoc performance metrics such as cycle numbers have been used to guide materials discovery, but it is unclear whether such metrics are suitable for the intended use. Here, we suggest a kinetic model allowing a narrower definition of cyclability by interpreting it through intrinsic parameters. Using this model, we examine systematic data on cycling and calendar stabilities for dialkoxyarene ROMs in stationary cells. Even for these closely related ROMs, the cycling numbers strongly depend on minor changes in test protocols and structure, which argue against the universality and transferability of these commonly used metrics. Recommendations regarding the standardization of testing protocols are suggested in light of these findings.

25 ENERGY STORAGE↗

SMART Deliverable 6.1.2a: Application of the ORION tool to the IBDP Carbon Storage Site

Forecasting and managing potential induced seismic activity is one of the challenges facing commercialscale geologic carbon sequestration (GCS), as well as other geologic energy extraction and byproduct disposal technologies. Historically, the process to develop robust, science-based forecasts of induced seismicity has required an integrated effort from experts in seismology, geomechanics, and reservoir engineering to manage data, develop and evaluate models of subsurface processes, and to calibrate and interpret the results from a range of models to understand site behavior relative to prescribed standards and in the context of uncertainty in geologic characterization data, forecasting models, and operational scenario uncertainty. The Operational Forecasting of Induced Seismicity (ORION) toolkit is an open-source, observation-based forecasting toolkit that is being co-developed by two U.S. DOE-funded initiatives: the National Risk Assessment Partnership (NRAP) and the Scienceinformed Machine Learning for Accelerating Real Time Decisions in Subsurface Applications (SMART) Initiative. ORION is designed to provide functionality to support decision making about seismic hazard analysis and risk management for GCS stakeholders ranging from the public to site operators to expert seismologists. The tool, which is written as open-source code in the Python programming language, is composed of a desktop graphical user interface (GUI) and an underlying forecasting engine. The forecasting engine uses available reservoir properties, well and fluid injection scenario details, and observed seismic catalog data as inputs to produce a set of temporal and spatio-temporal seismic forecasts.

58 GEOSCIENCES↗

Transient thermal modeling of the nonscanning ERBE detector

A numerical model to predict the transient thermal response of the ERBE nonscanning wide field of view total radiometer channel was developed. The model, which uses Monte Carlo techniques to characterize the radiative component of heat transfer, is described and a listing of the computer program is provided. Application of the model to simulate the actual blackbody calibration procedure is discussed. The use of the model to establish a real time flight data interpretation strategy is recommended. Modification of the model to include a simulated Earth radiation source field and a filter dome is indicated.

Mahan, J. R.↗

A simple model for the interaction of planetary and synoptic-scale waves in the troposphere

The time-mean circulation of the troposphere during the Northern hemisphere winter is dominated by planetary-scale stationary waves superposed on longitudinal-mean westerly flow. The movement and generation of weather-making, synoptic-scale transient waves are controlled by this large-scale circulation. The interannual variablity of planetary scale waves determines the severity of the winter season over broad geographic regions through such factors as frequency of storms, precipitation and temperature. A conceptual model that incorporates these feedbacks between the two scales is developed and guides the formulation of the quantitative low-order model which retains a few basic physical processes and wave components. In the low-order model the fluxes of sensible heat and momentum and latent heat release by transient eddies are parameterized in terms of the evolving large-scale circulation. Radiative forcing and damping, planetary-scale variations in moisture and topography and Ekman dissipation are also included. Testing of this model is in progress. The results will help interpret experiments from sophisticated general-circulation models and ascertain the need for satellite observations of precipitation.

Branscome, Lee E.↗

Numerical Evaluation of Effective Thermal Conductivity of PCM with Metal Foam Incorporating Buoyancy Effects for Thermal Energy Storage

The thermal energy storage (TES) system has the capability to efficiently preserve thermal energy directly derived from the energy source, minimizing any conversion losses. Especially latent heat storage offers distinct advantages, including a substantial increase in energy storage density and minimization of temperature fluctuations within the plants. However, the phase change material (PCM) employed in latent heat storage has low thermal conductivity. Consequently, various studies are being conducted to enhance heat transfer. One approach to enhance heat transfer involves utilizing metal foam to maximize the heat transfer area. However, modeling metal foam with its intricate structure is a challenging task in numerical analysis. For this reason, ongoing research focuses on simplifying the modeling of metal foam. Nevertheless, fully encompassing all the characteristics of actual metal foam proves to be a challenging task for the simplified analytical model. The objective of this paper is to interpret the simple lattice metal foam analysis model from the perspective of behavior induced by buoyancy. When comparing the analysis results of solid PCM and liquid PCM with the same thermal conductivity under changes in porosity and gravity direction, we conducted an analysis to discern the trends in effective thermal conductivity that are overestimated due to convection. In the analysis, a constant heat flux of 10 kW and a constant surface boundary condition of 350 K were applied, and a sensitivity study regarding the mesh was conducted. The results indicate that, from the perspective of gravity in the simple lattice model, the solid analysis yields an effective thermal conductivity 29-47% higher compared to the liquid analysis. Additionally, as porosity increases, there is an observed increase of 24-33% in effective thermal conductivity.

25 ENERGY STORAGE↗

Kinetic model of product distribution over Fe catalyst for Fischer-Tropsch synthesis

A new kinetic model of the Fischer-Tropsch synthesis (FTS) is proposed to describe the non-Anderson-Schulz-Flory (ASF) product distribution. The model is based on the double-polymerization monomers hypothesis, in which the surface C{sub 2}{asterisk} species acts as a chain-growth monomer in the light-product range, while C{sub 1}{asterisk} species acts as a chain-growth monomer in the heavy-product range. The detailed kinetic model in the Langmuir-Hinshelwood-Hougen-Watson type based on the elementary reactions is derived for FTS and the water-gas-shift reaction. Kinetic model candidates are evaluated by minimization of multiresponse objective functions with a genetic algorithm approach. The model of hydrocarbon product distribution is consistent with experimental data (<C{sub 45}). The new model gives a sound interpretation of the experimental data. 49 refs., 3 figs., 5 tabs.

01 COAL, LIGNITE, AND PEAT↗

DISK CORONA INTERACTION: MECHANISM FOR THE DISK TRUNCATION AND SPECTRUM CHANGE IN LOW-LUMINOSITY ACTIVE GALACTIC NUCLEI

The truncation of an optically thick, geometrically thin accretion disk is investigated in the context of low-luminosity AGNs (LLAGNs). We generalize the disk evaporation model used in the interpretative framework of black hole X-ray binaries by including the effect of a magnetic field in accretion disks surrounding supermassive black holes. The critical transition mass accretion rate for which the disk is truncated is found to be insensitive to magnetic effects, but its inclusion leads to a smaller truncation radius in comparison to a model without its consideration. That is, a thin viscous disk is truncated for LLAGNs at an Eddington ratio less than 0.03 for a standard viscosity parameter ({alpha} = 0.3). An increase of the viscosity parameter results in a higher critical transition mass accretion rate and a correspondingly smaller truncation distance, the latter accentuated by greater magnetic energy densities in the disk. Based on these results, the truncation radii inferred from spectral fits of LLAGNs published in the literature are consistent with the disk evaporation model. The infrared emission arising from the truncated geometrically thin accretion disks may be responsible for the red bump seen in such LLAGNs.

79 ASTRONOMY AND ASTROPHYSICS↗

Computational reacting gas dynamics

In the study of high speed flows at high altitudes, such as that encountered by re-entry spacecrafts, the interaction of chemical reactions and other non-equilibrium processes in the flow field with the gas dynamics is crucial. Generally speaking, problems of this level of complexity must resort to numerical methods for solutions, using sophisticated computational fluid dynamics (CFD) codes. The difficulties introduced by reacting gas dynamics can be classified into three distinct headings: (1) the usually inadequate knowledge of the reaction rate coefficients in the non-equilibrium reaction system; (2) the vastly larger number of unknowns involved in the computation and the expected stiffness of the equations; and (3) the interpretation of the detailed reacting CFD numerical results. The research performed accepts the premise that reacting flows of practical interest in the future will in general be too complex or 'untractable' for traditional analytical developments. The power of modern computers must be exploited. However, instead of focusing solely on the construction of numerical solutions of full-model equations, attention is also directed to the 'derivation' of the simplified model from the given full-model. In other words, the present research aims to utilize computations to do tasks which have traditionally been done by skilled theoreticians: to reduce an originally complex full-model system into an approximate but otherwise equivalent simplified model system. The tacit assumption is that once the appropriate simplified model is derived, the interpretation of the detailed numerical reacting CFD numerical results will become much easier. The approach of the research is called computational singular perturbation (CSP).

Lam, S. H.↗

Using Machine Learning to Estimate Surface-Level SO2 Concentrations from Satellite-Based Measurements

Sulfur dioxide (SO2) is a criteria air pollutant due to its contributions to aerosol formation, rainfall acidification, and harm to human health. The placement of air quality monitoring sites is typically biased towards urban areas, leaving large areas with very limited monitoring data. The Ozone Monitoring Instrument (OMI) has been used to provide estimates of SO2 vertical column densities (VCDs) globally at spatial resolution of 10s of kms once per day. OMI SO2 VCDs have been previously used to estimate surface SO2 concentrations using chemical transport model (CTM) simulations. The CTMs use estimated emissions and assimilated meteorological data, and simulate the chemical and physical processes that determine the vertical profile of SO2, which can be used to derive a ratio between the surface concentrations and VCDs. These models are complex, computationally expensive, and have large uncertainties in the simulated surface-to-VCD ratio due to biases in emissions and relatively coarse resolution. Machine learning techniques are comparatively easier to use, much less computationally expensive to use after training, and can produce more accurate estimations of surface concentrations than the CTM-based method. The interpretation of machine learning models often poses challenges, and in some cases, non-physical variables unrelated to SO2 are used as predictors. In this work, we create an artificial neural network (ANN) to relate OMI retrievals and archived GEOS-FP boundary layer heights to surface SO2 concentrations from the ChinaHighAirPollutants ChinaHighSO2 dataset (CHAP; Wei et al., 2023) on a seasonal average timescale from 2013-2018. Our model only utilizes five variables that are directly relevant to the satellite retrieval, lifetime, and spatial distribution of SO2. The model was trained on 16 seasons (four of each) with independent validation (one of each season) and testing datasets (one of each season) to avoid overfitting. Our ANN generates surface SO2 concentrations that are sensitive (slope = 0.51) and consistent (r = 0.74) with the CHAP data, but are underpredicted by an average of 1.2 ppbv with a mean absolute error of 2.2 ppbv. These results are better than recent studies utilizing the CTM method. To our knowledge, this is the best performing machine learning model that only uses physical variables to predict surface SO2. Our work demonstrates that a carefully constructed, simple ML model can accurately estimate surface-based SO2 concentrations from satellite VCD measurements, and this technique has future promise to expend to newer, higher resolution satellites and other air pollutants.

SO2, air quality, OMI, machine learning↗

Thermodynamic Model for BiPO 4 (cr) and Bi(OH) 3 (am) Solubility in the Aqueous Na + –H + –H 2 PO - 4 –HPO 2- 4 –PO 3- 4 –OH - –Cl - –H 2 O System

Prior to this study no data for the solubility product of BiPO 4 (cr) or the complexation constants of Bi with phosphate were available. The solubility of BiPO 4 (cr) was studied at 23 ± 2 °C from both the over- and under-saturation directions as functions of a wide range in time (6–309 days), pH values (0–15), and phosphate concentrations (reaching as high as 1.0 mol·kg -1 ). HCl or NaOH were used to obtain a range in pH values. Steady state concentrations and equilibrium were reached in <6 days. The data were interpreted using the SIT model. These extensive data provided a solubility product value for BiPO 4 (cr) and an upper limit value for the formation of BiPO 4 (aq). Because the aqueous system in this study involved relatively high concentrations of chloride, reliable values for the complexation constants of Bi with chloride were required to accurately interpret the solubility data. Therefore as a part of this investigation, existing Bi–Cl data were critically reviewed and used to obtain values of equilibrium constants for various Bi–Cl complexes at zero ionic strength along with the values for various SIT ion interaction parameters. Predictions based on these thermodynamic quantities agreed closely with our experimental data, the chloride concentrations of which ranged as high as 0.7 mol·kg -1 . The study showed that BiPO 4 (cr) is stable at pH values <9.0. At pH values >9.0, Bi(OH) 3 (am) is the solubility controlling phase. Reliable values for the Bi(OH)3(am) solubility reactions involving Bi(OH) 3 (aq) and Bi(OH) - 4 and the formation constants of these aqueous species are also reported.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Simultaneous XTE and AXAF Observations of the Circinus Galaxy

This grant supported the analysis and interpretations of X-ray observations fo the Circinus galaxy collected with the Rossi X-ray Timing Explorer in June 2000, simultaneous to Chandra observations. The primary goals of the RXTE observations were to obtain the high-energy (> 10 keV) spectrum of the galaxy to quantify the nucleus luminosity. The latter is an essential ingredient of the models we used to interpret the results from the Chandra observations of the galaxy.

Sambruna, Rita↗

Simple holographic models of black hole evaporation

Several recent papers have shown a close relationship between entanglement wedge reconstruction and the unitarity of black hole evaporation in AdS/CFT. The analysis of these papers however has a rather puzzling feature: all calculations are done using bulk dynamics which are essentially those Hawking used to predict information loss, but applying ideas from entanglement wedge reconstruction seems to suggest a Page curve which is consistent with information conservation. Why should two different calculations in the same model give different answers for the Page curve? In this note we present a new pair of models which clarify this situation. Our first model gives a holographic illustration of unitary black hole evaporation, in which the analogue of the Hawking radiation purifies itself as expected, and this purification is reproduced by the entanglement wedge analysis. Moreover a smooth black hole interior persists until the last stages the evaporation process. Our second model gives an alternative holographic interpretation of the situation where the bulk evolution leads to information loss: unlike in the models proposed so far, this bulk information loss is correctly reproduced by the entanglement wedge analysis. This serves as an illustration that quantum extremal surfaces are in some sense kinematic: the time-dependence of the entropy they compute depends on the choice of bulk dynamics. In both models no bulk quantum corrections need to be considered: classical extremal surfaces are enough to do the job. We argue that our first model is the one which gives the right analogy for what actually happens to evaporating black holes, but we also emphasize that any complete resolution of the information problem will require an understanding of non-perturbative bulk dynamics.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Characterization and calibration of a viscoelastic simplified potential energy clock model for inorganic glasses

In this study, to analyze the stresses and strains generated during the solidification of glass-forming materials, stress and volume relaxation must be predicted accurately. Although the modeling attributes required to depict physical aging in organic glassy thermosets strongly resemble the structural relaxation in inorganic glasses, the historical modeling approaches have been distinctly different. To determine whether a common constitutive framework can be applied to both classes of materials, the nonlinear viscoelastic simplified potential energy clock (SPEC) model, developed originally for glassy thermosets, was calibrated for the Schott 8061 inorganic glass and used to analyze a number of tests. A practical methodology for material characterization and model calibration is discussed, and the structural relaxation mechanism is interpreted in the context of SPEC model constitutive equations. SPEC predictions compared to inorganic glass data collected from thermal strain measurements and creep tests demonstrate the ability to achieve engineering accuracy and make the SPEC model feasible for engineering applications involving a much broader class of glassy materials.

36 MATERIALS SCIENCE↗

Investigation of Sampling-Probe Distorted Temperature Fields with X-Ray Fluorescence Spectroscopy

Flame-sampling experiments, especially in conjunction with laminar low-pressure premixed flames, are routinely used in combustion chemistry studies to unravel the identities and quantities of key intermediates and their pathways. In many instances, however, an unambiguous interpretation of the experimental and modeling results is hampered by the uncertainties about the probe-induced, perturbed temperature profile. To overcome this limitation, two-dimensional perturbations of the temperature field caused by sampling probes with different geometries have been investigated using synchrotron-based X-ray fluorescence spectroscopy. In these experiments, which were performed at the 7-BM beamline of the Advanced Photon Source (APS) at the Argonne National Laboratory, a continuous beam of hard X-rays at 15 keV was used to excite krypton atoms that were added in a concentration of 5 vol.-% to the unburnt gas mixture and the resulting krypton fluorescence at 12.65 keV was subsequently collected. The highly spatially resolved signal was converted into the local flame temperature to obtain temperature fields at various burner-probe separations as functions of the distance to the burner surface and the radial distance from the centerline. Multiple measurements were performed with different probe geometries and because of the observed impact on the temperature profiles the results clearly revealed the need to specify the sampling probe design to enable quantitative and meaningful comparisons of modeling results with flame-sampled mole fraction data. (C) 2018 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Hansen, N.↗

Controlling the angular radiation of single emitters using dielectric patch nanoantennas

Dielectric nanoantennas have generated much interest in recent years owing to their low loss and optically induced electric and magnetic resonances. In this paper, we investigate the coupling between a single emitter and dielectric patch nanoantennas. For the coupled system involving non-spherical structures, analytical Mie theory is no longer applicable. A semi-analytical model is proposed instead to interpret the coupling mechanism and the radiation characteristics of the system. Based on the presented model, we demonstrate that the angular emission of the single emitter can be not only enhanced but also rotated using the dielectric patch nanoantennas.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

An empirical approach to model selection: weak lensing and intrinsic alignments

ABSTRACT In cosmology, we routinely choose between models to describe our data, and can incur biases due to insufficient models or lose constraining power with overly complex models. In this paper, we propose an empirical approach to model selection that explicitly balances parameter bias against model complexity. Our method uses synthetic data to calibrate the relation between bias and the χ2 difference between models. This allows us to interpret χ2 values obtained from real data (even if catalogues are blinded) and choose a model accordingly. We apply our method to the problem of intrinsic alignments – one of the most significant weak lensing systematics, and a major contributor to the error budget in modern lensing surveys. Specifically, we consider the example of the Dark Energy Survey Year 3 (DES Y3), and compare the commonly used non-linear alignment (NLA) and tidal alignment and tidal torque (TATT) models. The models are calibrated against bias in the Ωm–S8 plane. Once noise is accounted for, we find that it is possible to set a threshold Δχ2 that guarantees an analysis using NLA is unbiased at some specified level Nσ and confidence level. By contrast, we find that theoretically defined thresholds (based on, e.g. p-values for χ2) tend to be overly optimistic, and do not reliably rule out cosmological biases up to ∼1–2σ. Considering the real DES Y3 cosmic shear results, based on the reported difference in χ2 from NLA and TATT analyses, we find a roughly $30{{\ \rm per\ cent}}$ chance that were NLA to be the fiducial model, the results would be biased (in the Ωm–S8 plane) by more than 0.3σ. More broadly, the method we propose here is simple and general, and requires a relatively low level of resources. We foresee applications to future analyses as a model selection tool in many contexts.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Search for massive resonances decaying into $WW$, $WZ$, $ZZ$, $qW$, and $qZ$ with dijet final states at $\sqrt{s}=13\text{ }\text{ }\mathrm{TeV}$

Results are presented from a search in the dijet final state for new massive narrow resonances decaying to pairs of W and Z bosons or to a W/Z boson and a quark. Results are based on data recorded in proton-proton collisions at s=13 TeV with the CMS detector at the CERN LHC. The data correspond to an integrated luminosity of 35.9 fb-1. The mass range investigated extends upwards from 1.2 TeV. No excess is observed above the estimated standard model background and limits are set at 95% confidence level on cross sections, which are interpreted in terms of various models that predict gravitons, heavy spin-1 bosons, and excited quarks. In a heavy vector triplet model, W′ and Z′ resonances, with masses below 3.2 and 2.7 TeV, respectively, and spin-1 resonances with degenerate masses below 3.8 TeV are excluded at 95% confidence level. In the case of a singlet W′ resonance masses between 3.3 and 3.6 TeV can be excluded additionally. Similarly, excited quark resonances, q*, decaying to qW and qZ with masses less than 5.0 and 4.7 TeV, respectively, are excluded. In a narrow-width bulk graviton model, upper limits are set on cross sections ranging from 0.6 fb for high resonance masses above 3.6 TeV, to 36.0 fb for low resonance masses of 1.3 TeV.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A search for new phenomena in pp collisions at $\sqrt{s} = 13\,\text {TeV} $ in final states with missing transverse momentum and at least one jet using the $\alpha _{\mathrm {T}}$ variable

A search for new phenomena is performed in final states containing one or more jets and an imbalance in transverse momentum in pp collisions at a centre-of-mass energy of 13 $\,\text {TeV}$ . The analysed data sample, recorded with the CMS detector at the CERN LHC, corresponds to an integrated luminosity of 2.3 $\,\text {fb}^{-1}$ . Several kinematic variables are employed to suppress the dominant background, multijet production, as well as to discriminate between other standard model and new physics processes. The search provides sensitivity to a broad range of new-physics models that yield a stable weakly interacting massive particle. The number of observed candidate events is found to agree with the expected contributions from standard model processes, and the result is interpreted in the mass parameter space of fourteen simplified supersymmetric models that assume the pair production of gluinos or squarks and a range of decay modes. For models that assume gluino pair production, masses up to 1575 and 975 $\,\text {GeV}$ are excluded for gluinos and neutralinos, respectively. For models involving the pair production of top squarks and compressed mass spectra, top squark masses up to 400 $\,\text {GeV}$ are excluded.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗