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 361 records · Page 20

Theory and modeling of molecular modes in the NMR relaxation of fluids

Traditional theories of the nuclear magnetic resonance (NMR) autocorrelation function for intra-molecular dipole pairs assume a single-exponential decay, yet the calculated autocorrelation of realistic systems displays a rich, multi-exponential behavior, resulting in anomalous NMR relaxation dispersion (i.e., frequency dependence). We develop an approach to model and interpret the multi-exponential intra-molecular autocorrelation using simple, physical models within a rigorous statistical mechanical development that encompasses both rotational diffusion and translational diffusion in the same framework. Here, we recast the problem of evaluating the autocorrelation in terms of averaging over a diffusion propagator whose evolution is described by a Fokker–Planck equation. The time-independent part admits an eigenfunction expansion, allowing us to write the propagator as a sum over modes. Each mode has a spatial part that depends on the specified eigenfunction and a temporal part that depends on the corresponding eigenvalue (i.e., correlation time) with a simple, exponential decay. The spatial part is a probability distribution of the dipole pair, analogous to the stationary states of a quantum harmonic oscillator. Drawing inspiration from the idea of inherent structures in liquids, we interpret each of the spatial contributions as a specific molecular mode. These modes can be used to model and predict the NMR dipole–dipole relaxation dispersion of fluids by incorporating phenomena on the molecular level. We validate our statistical mechanical description of the distribution in molecular modes with molecular dynamics simulations interpreted without any relaxation models or adjustable parameters: the most important poles in the Padé–Laplace transform of the simulated autocorrelation agree with the eigenvalues predicted by the theory

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

HERSCHEL PACS AND SPIRE OBSERVATIONS OF BLAZAR PKS 1510-089: A CASE FOR TWO BLAZAR ZONES

We present the results of observations of blazar PKS 1510-089 with the Herschel Space Observatory PACS and SPIRE instruments, together with multiwavelength data from Fermi/LAT, Swift, SMARTS, and Submillimeter Array. The source was found in a quiet state, and its far-infrared spectrum is consistent with a power law with a spectral index of {alpha} {approx_equal} 0.7. Our Herschel observations were preceded by two 'orphan' gamma-ray flares. The near-infrared data reveal the high-energy cutoff in the main synchrotron component, which cannot be associated with the main gamma-ray component in a one-zone leptonic model. This is because in such a model the luminosity ratio of the external-Compton (EC) and synchrotron components is tightly related to the frequency ratio of these components, and in this particular case an unrealistically high energy density of the external radiation would be implied. Therefore, we consider a well-constrained two-zone blazar model to interpret the entire data set. In this framework, the observed infrared emission is associated with the synchrotron component produced in the hot-dust region at the supra-parsec scale, while the gamma-ray emission is associated with the EC component produced in the broad-line region at the sub-parsec scale. In addition, the optical/UV emission is associated with the accretion disk thermal emission, with the accretion disk corona likely contributing to the X-ray emission.

79 ASTRONOMY AND ASTROPHYSICS↗

Explaining System-Level Prognostics with Established Machine Learning Methods

System-level prognostics is crucial for ensuring reliability and enabling predictive maintenance in complex systems with interconnected components. This study presents a framework that integrates data-driven methods to predict the remaining useful life (RUL) of a subsystem under multiple and concurrent faults within a nuclear power plant system with explainable artificial intelligence (XAI). A nuclear power plant (NPP) operation was simulated to model the degradation behavior of NPP components, and four machine learning models—Gradient Boosting Regressor (GBR), Support Vector Regressor (SVR), Fully Connected Neural Network (FCNN), and Long Short-Term Memory (LSTM)—were evaluated for prognostics with a novel system RUL parameter. The LSTM model demonstrated potential superior repeatability, while SHAP (SHapley Additive exPlanations) for explainability provided consistent and trustworthy global explanations. In contrast, LIME (Local Interpretable Model-agnostic Explanations) offered localized interpretability but showed reduced stability for sequential data. Key findings include the interplay between component-level degradation and system-wide performance, with LSTM effectively capturing these dynamics through sequence-level predictions. The XAI techniques enhanced transparency by identifying critical features influencing model predictions and aligning with domain knowledge. Furthermore, this framework has significant implications for improving trust and understanding in predictive maintenance, particularly in safety-critical industries like nuclear energy.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

The origin of the plateau and late rebrightening in the afterglow of GRB 120326A

GRB 120326A is an unusual gamma-ray burst (GRB) that has a long plateau and a very late rebrightening in both X-ray and optical bands. The similar behavior of the optical and X-ray light curves suggests that they may share a common origin. The long plateau starts at several hundred seconds and ends at tens of thousands of seconds, and the peak time of the late rebrightening is about 30,000 s. We analyze the energy injection model by means of numerical and analytical solutions, considering both the wind environment and the interstellar medium environment for GRB afterglows. We particularly study the influence of the injection starting time, ending time, stellar wind density (or density of the circumburst environment), and injection luminosity on the shape of the afterglow light curves, respectively. In the wind model, we find that the light curve is largely affected by the parameters and that there is a 'bump' in the late stage. In the wind environment, we found that the longer the energy is injected, the more obvious the rebrightening will be. We also find that the peak time of the bump is determined by the stellar wind density. We use the late continuous injection model to interpret the unusual afterglow of GRB 120326A. The model fits the observational data well; however, we find that the timescale of the injection must be higher than 10,000 s, which implies that the timescale of the central engine activity must also be more than 10,000 s. This information can give useful constraints on the central engines of GRBs—we consider a newborn millisecond pulsar with a strong magnetic field to be the central engine. On the other hand, our results suggest that the circumburst environment of GRB 120326A is very likely a stellar wind.

79 ASTRONOMY AND ASTROPHYSICS↗

Interpretable, Extensible Linear Regression Model for Electron Density Prediction

This code allows us to generate a model to predict ground state electron densities directly from atomic structure thereby bypassing the need to perform direct DFT calculations. It is based on many body correlation descriptors defined in https://arxiv.org/abs/2004.14442 (IM release numbers: LLNL-JRNL- 808791 and LLNL-JRNL-840698). The code reads in electron densities, in CHGCAR format used by the DFT code VASP, calculates many-body correlation descriptors based on atoms surrounding a grid point and then uses linear regression to obtain the fitting parameters.

Kumar, Shashikant↗

When ancient numerical demons meet physics-informed machine learning: adjoint-based gradients for implicit differentiable modeling

Recent advances in differentiable modeling, a genre of physics-informed machine learning that trains neural networks (NNs) together with process-based equations, have shown promise in enhancing hydrological models' accuracy, interpretability, and knowledge-discovery potential. Current differentiable models are efficient for NN-based parameter regionalization, but the simple explicit numerical schemes paired with sequential calculations (operator splitting) can incur numerical errors whose impacts on models' representation power and learned parameters are not clear. Implicit schemes, however, cannot rely on automatic differentiation to calculate gradients due to potential issues of gradient vanishing and memory demand. Here we propose a “discretize-then-optimize” adjoint method to enable differentiable implicit numerical schemes for the first time for large-scale hydrological modeling. The adjoint model demonstrates comprehensively improved performance, with Kling–Gupta efficiency coefficients, peak-flow and low-flow metrics, and evapotranspiration that moderately surpass the already-competitive explicit model. Therefore, the previous sequential-calculation approach had a detrimental impact on the model's ability to represent hydrological dynamics. Furthermore, with a structural update that describes capillary rise, the adjoint model can better describe baseflow in arid regions and also produce low flows that outperform even pure machine learning methods such as long short-term memory networks. The adjoint model rectified some parameter distortions but did not alter spatial parameter distributions, demonstrating the robustness of regionalized parameterization. Despite higher computational expenses and modest improvements, the adjoint model's success removes the barrier for complex implicit schemes to enrich differentiable modeling in hydrology.

58 GEOSCIENCES↗

Orthogonal Gaussian process models

Gaussian processes models are widely adopted for nonparameteric/semi-parametric modeling. Identifiability issues occur when the mean model contains polynomials with unknown coefficients. Though resulting prediction is unaffected, this leads to poor estimation of the coefficients in the mean model, and thus the estimated mean model loses interpretability. This paper introduces a new Gaussian process model whose stochastic part is orthogonal to the mean part to address this issue. As a result, this paper also discusses applications to multi-fidelity simulations using data examples.

97 MATHEMATICS AND COMPUTING↗

Radiation from a homogeneous sphere mounted on a waveguide aperture

The results of an experimental and theoretical study of the effects of placing a dielectric sphere over the aperture of a circular waveguide are presented. A dielectric sphere excited by a Huygens' source is used as the theoretical model to interpret the experimental results. Calculations based on this model show good overall agreement with experimental patterns and directivity measurements. It is concluded that dielectric loaded antennas such as the dielectric sphere excited by a waveguide can produce directivities in excess of that obtained by a uniformly illuminated aperture of the same cross section, particularly for dielectric objects with dimensions of two wavelengths or less. Measurements of the coupling between these antennas show that they, in general, are not as closely coupled as plain apertures without dielectric object loading.

Croswell, W. F.↗

Shaded computer graphic techniques for visualizing and interpreting analytic fluid flow models

Mathematical models which predict the behavior of fluid flow in different experiments are simulated using digital computers. The simulations predict values of parameters of the fluid flow (pressure, temperature and velocity vector) at many points in the fluid. Visualization of the spatial variation in the value of these parameters is important to comprehend and check the data generated, to identify the regions of interest in the flow, and for effectively communicating information about the flow to others. The state of the art imaging techniques developed in the field of three dimensional shaded computer graphics is applied to visualization of fluid flow. Use of an imaging technique known as 'SCAN' for visualizing fluid flow, is studied and the results are presented.

Parke, F. I.↗

Capturing Complex Multivariate Time Series Interactions to Detect High-Risk Adverse Events During Flight

The reduction of aviation safety metrics below target thresholds continue to drive down the number of aviation fatalities and accidents. To meet future safety demands, sustained efforts by aviation agencies promoting safety assurance processes and systems have prompted ongoing research on identifying and mitigating in-flight risks. With the projected increase in passenger load factor and rollout of more autonomous systems into the national airspace, the need to detect high-risk events in-time or ahead-of-time is becoming increasingly crucial. New anomaly detection and precursor identification algorithms will need to scale to different airframes, levels of autonomy, and system complexity. While the pervasiveness of deep learning has resulted in the development of performant anomaly detection methods, these sophisticated models currently suffer from low end-user interpretability. Building off our previous work on identifying adverse events in multivariate flight data during descent, we propose a data-driven approach for detecting in-flight adverse events caused by the complex interplay of flight variables. Our approach utilizes ordinal patterns of important aircraft stability variables (e.g., airspeed and descent rate) to capture multivariate flight dynamics that can be used to predict the onset of unstable approaches, a high-risk adverse event that can occur during approach. Through the use of ordinal patterns, we aim to create more interpretable detection models of in-flight adverse events that can be translated to future autonomous systems without difficulty. Our analysis shows the presence of distinct ordinal pattern distributions that can be used to predict unstable approaches 1 minute ahead of time with an accuracy of 0.69 and a recall of 0.73 and 30 seconds ahead with an accuracy of 0.70 and a recall of 0.86.

Risk detection↗

Capturing Multivariate Time Series Interactions to Detect High‑Risk Instability During Approach

The reduction of aviation safety metrics below target thresholds continue to drive down the number of aviation fatalities and accidents. To meet future safety demands, sustained efforts by aviation agencies promoting safety assurance processes and systems have prompted ongoing research on identifying and mitigating in-flight risks. With the projected increase in passenger load factor and rollout of more autonomous systems into the national airspace, the need to detect high-risk events in-time or ahead-of-time is becoming increasingly crucial. New anomaly detection and precursor identification algorithms will need to scale to different airframes, levels of autonomy, and system complexity. While the pervasiveness of deep learning has resulted in the development of performant anomaly detection methods, these sophisticated models currently suffer from low end-user interpretability. Building off our previous work on identifying adverse events in multivariate flight data during descent, we propose a data-driven approach for detecting in-flight adverse events caused by the complex interplay of flight variables. Our approach utilizes ordinal patterns of important aircraft stability variables (e.g., airspeed and descent rate) to capture multivariate flight dynamics that can be used to predict the onset of unstable approaches, a high-risk adverse event that can occur during approach. Through the use of ordinal patterns, we aim to create more interpretable detection models of in-flight adverse events that can be translated to future autonomous systems without difficulty. Our analysis shows the presence of distinct ordinal pattern distributions that can be used to predict unstable approaches 1 minute ahead of time with an accuracy of 0.69 and a recall of 0.73 and 30 seconds ahead with an accuracy of 0.70 and a recall of 0.86.

Risk detection↗

Revisiting Fenton Hill Phase I reservoir creation and stimulation mechanisms through the GTO code comparison effort

A series of experiments performed at the Fenton Hill hot dry rock site after stage 2 drilling of Phase I reservoir provided intriguing field observations on the reservoir’s responses to injection and venting under various conditions. Two teams participating in the US DOE Geothermal Technologies Office (GTO)’s Code Comparison Study (CCS) used different numerical codes to model these five experiments with the objective of inferring the hydraulic stimulation mechanism involved. The codes used by the two teams are based on different numerical principles, and the assumptions made were also different, due to intrinsic limitations in the codes and the modelers’ personal interpretations of the field observations. Both sets of models were able to produce the most important field observations and both found that it was the combination of the vertical gradient of the fracture opening pressure, injection volume, and the use/absence of proppant that yielded the different outcomes of the five experiments.

Fu, Pengcheng↗

A Data-Driven Method for Modeling Creep-Fatigue Stress- Strain Behavior Using Neural ODEs

In this paper, we introduce a data-driven machine learning approach for modeling one-dimensional stress–strain behavior under cyclic loading, utilizing experimental data from the nickel-based Alloy 617. The study employs uniaxial creep–fatigue test data acquired under various loading histories and compares two distinct neural network-based ODE models. The first model, known as the black-box model, comprehensively describes the strain–stress relationship using a Neural ODE equation. To interpret this black-box model, we apply the Sparse Identification of Nonlinear Dynamical Systems (SINDy) technique, transforming the black-box model into an equation-based model using symbolic regression. The second model, the Neural flow rule model, incorporates Hooke’s Law for the linear elastic component, with the nonlinear part characterized by a Neural ODE. Both models are trained with experimental data to accurately reflect the observed stress–strain behavior. We conduct a detailed comparison with the standard Chaboche model, which includes three back stresses. Our results demonstrate that the neural network-based ODE models precisely capture the experimental creep–fatigue mechanical behavior, exceeding the standard Chaboche model’s accuracy. Furthermore, an interpretable model derived from the black-box neural ODE model through symbolic regression achieves accuracy comparable to the Chaboche model, enhancing its interpretability. The results highlight the potential of neural network-based ODE models to depict complex creep–fatigue behavior, eliminating the necessity for experts to define a specific, material-focused model form.

creep-fatigue↗