Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Explanability”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
NASA NTRS2006

Explaining Verification Conditions

The Hoare approach to program verification relies on the construction and discharge of verification conditions (VCs) but offers no support to trace, analyze, and understand the VCs themselves. We describe a systematic extension of the Hoare rules by labels so that the calculus itself can be used to build up explanations of the VCs. The labels are maintained through the different processing steps and rendered as natural language explanations. The explanations can easily be customized and can capture different aspects of the VCs; here, we focus on their structure and purpose. The approach is fully declarative and the generated explanations are based only on an analysis of the labels rather than directly on the logical meaning of the underlying VCs or their proofs. Keywords: program verification, Hoare calculus, traceability.

Deney, Ewen↗
NASA NTRS2012

The Scaling of Broadband Shock-Associated Noise with Increasing Temperature

A physical explanation for the saturation of broadband shock-associated noise (BBSAN) intensity with increasing jet stagnation temperature has eluded investigators. An explanation is proposed for this phenomenon with the use of an acoustic analogy. For this purpose the acoustic analogy of Morris and Miller is examined. To isolate the relevant physics, the scaling of BBSAN at the peak intensity level at the sideline ( = 90 degrees) observer location is examined. Scaling terms are isolated from the acoustic analogy and the result is compared using a convergent nozzle with the experiments of Bridges and Brown and using a convergent-divergent nozzle with the experiments of Kuo, McLaughlin, and Morris at four nozzle pressure ratios in increments of total temperature ratios from one to four. The equivalent source within the framework of the acoustic analogy for BBSAN is based on local field quantities at shock wave shear layer interactions. The equivalent source combined with accurate calculations of the propagation of sound through the jet shear layer, using an adjoint vector Green s function solver of the linearized Euler equations, allows for predictions that retain the scaling with respect to stagnation pressure and allows for the accurate saturation of BBSAN with increasing stagnation temperature. This is a minor change to the source model relative to the previously developed models. The full development of the scaling term is shown. The sources and vector Green s function solver are informed by steady Reynolds-Averaged Navier-Stokes solutions. These solutions are examined as a function of stagnation temperature at the first shock wave shear layer interaction. It is discovered that saturation of BBSAN with increasing jet stagnation temperature occurs due to a balance between the amplification of the sound propagation through the shear layer and the source term scaling.A physical explanation for the saturation of broadband shock-associated noise (BBSAN) intensity with increasing jet stagnation temperature has eluded investigators. An explanation is proposed for this phenomenon with the use of an acoustic analogy. For this purpose the acoustic analogy of Morris and Miller is examined. To isolate the relevant physics, the scaling of BBSAN at the peak intensity level at the sideline psi = 90 degrees) observer location is examined. Scaling terms are isolated from the acoustic analogy and the result is compared using a convergent nozzle with the experiments of Bridges and Brown and using a convergent-divergent nozzle with the experiments of Kuo, McLaughlin, and Morris at four nozzle pressure ratios in increments of total temperature ratios from one to four. The equivalent source within the framework of the acoustic analogy for BBSAN is based on local field quantities at shock wave shear layer interactions. The equivalent source combined with accurate calculations of the propagation of sound through the jet shear layer, using an adjoint vector Green s function solver of the linearized Euler equations, allows for predictions that retain the scaling with respect to stagnation pressure and allows for the accurate saturation of BBSAN with increasing stagnation temperature. This is a minor change to the source model relative to the previously developed models. The full development of the scaling term is shown. The sources and vector Green s function solver are informed by steady Reynolds-Averaged Navier-Stokes solutions. These solutions are examined as a function of stagnation temperature at the first shock wave shear layer interaction. It is discovered that saturation of BBSAN with increasing jet stagnation temperature occurs due to a balance between the amplification of the sound propagation through the shear layer and the source term scaling.

Miller, Steven A.↗
NASA NTRS2020

Improving Trust in Deep Neural Networks with Nearest Neighbors

Deep neural networks are used increasingly for perception and decision-making in UAVs. For example, they can be used to recognize objects from images and decide what actions the vehicle should take. While deep neural networks can perform very well at complex tasks, their decisions may be unintuitive to a human operator. When a human disagrees with a neural network prediction, due to the black box nature of deep neural networks, it can be unclear whether the system knows something the human does not or whether the system is malfunctioning. This uncertainty is problematic when it comes to ensuring safety. As a result, it is important to develop technologies for explaining neural network decisions for trust and safety. This paper explores a modification to the deep neural network classification layer to produce both a predicted label and an explanation to support its prediction. Specifically, at test time, we replace the final output layer of the neural network classifier by a k-nearest neighbor classifier. The nearest neighbor classifier produces 1) a predicted label through voting and 2) the nearest neighbors involved in the prediction, which represent the most similar examples from the training dataset. Because prediction and explanation are derived from the same underlying process, this approach guarantees that the explanations are always relevant to the predictions. We demonstrate the approach on a convolutional neural network for a UAV image classification task. We perform experiments using a forest trail image dataset and show empirically that the hybrid classifier can produce intuitive explanations without loss of predictive performance compared to the original neural network. We also show how the approach can be used to help identify potential issues in the network and training process.

Lee, Ritchie↗
NASA NTRSDate not supplied

Plan-Property Dependencies are Useful: A User Study

The trade-offs between different desirable plan properties –e. g. PDDL temporal plan preferences – are often difficult to understand. Recent work proposes to address this by iterative planning with explanations elucidating the dependencies between such plan properties. Users can ask questions of the form “Why does the plan you suggest not satisfy property p?”, which are answered by “Because then we would have to forego q” where ¬q is entailed by p in plan space. It has been shown that such plan-property dependencies can be computed reasonably efficiently. But is this form of explanation actually useful for users? We contribute a user study evaluating that question. We design use cases from three domains and run a large user study (N= 40 for each domain, ca. 40 minutes work time per user and domain) on the internet platform Prolific. Comparing users with vs. without access to the explanations, we find that the explanations tend to enable users to identify better trade-offs between the plan properties, indicating an improved understanding of the task.

Explainable Planning↗
DOE OSTI2021

Explaining Missing Data in Graphs: A Constraint-based Approach

Abstract: This paper introduces a constraint-based approach to clarify missing values in graphs. Our method capitalizes on a set S of graph data constraints. An explanation is a sequence of operational enforcement of S towards the recovery of interested yet missing data (e.g., attribute values, edges). We show that constraint-based approach helps us to understand not only why a value is missing, but also how to recover the missing value. We study S-explanation problem, which is to compute the optimal explanations with guarantees on the informativeness and conciseness. We show the problem is in ?P^2 for established graph data constraints such as graph keys and graph association rules. We develop an efficient bidirectional algorithm to compute optimal explanations, without enforcing S on the entire graph. We also show our algorithm can be easily extended to support graph refinement within limited time, and to explain missing answers. Using real-world graphs, we experimentally verify the effectiveness and efficiency of our algorithms.

Data Analytics↗
DOE OSTI2026

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↗
DOE OSTI2020

The Bright Side and Dark Side of Hybrid Organic–Inorganic Perovskites

The previously developed bistable amphoteric native defect (BAND) model is used for a comprehensive explanation of the unique photophysical properties and for understanding the remarkable performance of perovskites as photovoltaic materials. It is shown that the amphoteric defects in the donor (acceptor) configuration capture a fraction of photoexcited electrons (holes) dividing them into two groups: higher-energy bright and lower-energy dark electrons (holes). The spatial separation of the dark electrons and dark holes and the k-space separation of the bright and dark charge carriers reduce electron-hole recombination rates, emulating the properties of an ideal photovoltaic material with a balanced, spatially separated transport of electrons and holes. The BAND model also offers a straightforward explanation for the exceptional insensitivity of the photovoltaic performance of polycrystalline perovskite films to structural and optical inhomogeneities. The blue-shifted radiative recombination of bright electrons and holes results in a large anti-Stokes effect that provides a quantitative explanation for the spectral dependence of the laser cooling effect measured in perovskite platelets.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Hiding in plain sight: The prevalence and impact of trions and Fermi polarons in transient absorption spectroscopy experiments of 2D semiconductors

Transient absorption (TA) spectroscopy is one of the most popular experimental methods to measure the excited state lifetimes and charge carrier recombination mechanisms in two dimensional (2D) semiconductors. This fundamental information is essential for designing and optimizing the next generation of ultrathin and lightweight 2D semiconductor-based optoelectronic devices. However, the interpretation of TA spectroscopy data varies across the community. The community lacks a unifying physical explanation for how and why experimental variables such as incident light intensity, sample-substrate interactions, and/or applied bias affect TA spectral data. This Perspective (1) compares the physical chemistry TA literature to nanomaterial physics literature from a historical perspective, (2) reviews multiple physical explanations that the TA community developed to explain spectral features and experimental trends, (3) provides a unifying explanation for how and why trions—and, more generally, Fermi polarons—contribute to TA spectra, and (4) quantifies the extent to which various physical interpretations and data analysis procedures yield different timescales and mechanisms for the same set of experimental results. We highlight the importance of considering trions/Fermi polarons in TA measurements and their implications for advancing our understanding of 2D material properties.

2D materials↗
DOE OSTI2022

Evidence for the Late Arrival of Hot Jupiters in Systems with High Host-star Obliquities

It has been shown that hot Jupiters systems with massive, hot stellar primaries exhibit a wide range of stellar obliquities. On the other hand, hot Jupiter systems with low-mass, cool primaries often have stellar obliquities close to zero. Efficient tidal interactions between hot Jupiters and the convective envelopes present in lower-mass main-sequence stars have been a popular explanation for these observations. If this explanation is accurate, then aligned systems should be older than misaligned systems. Likewise, the convective envelope mass of a hot Jupiter's host star should be an effective predictor of its obliquity. We derive homogeneous stellar parameters—including convective envelope masses—for hot Jupiter host stars with high-quality sky-projected obliquity inferences. Using a thin-disk stellar population's Galactic velocity dispersion as a relative age proxy, we find that hot Jupiter host stars with larger-than-median obliquities are older than hot Jupiter host stars with smaller-than-median obliquities. The relative age difference between the two populations is larger for hot Jupiter host stars with smaller-than-median fractional convective envelope masses and is significant at the 3.6σ level. We identify stellar mass, not convective envelope mass, as the best predictor of stellar obliquity in hot Jupiter systems. The best explanation for these observations is that many hot Jupiters in misaligned systems arrived in the close proximity of their host stars long after their parent protoplanetary disks dissipated. The dependence of observed age offset on convective envelope mass suggests that tidal realignment contributes to the population of aligned hot Jupiters orbiting stars with convective envelopes.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Explainable Machine Learning for Functional Data

Black-box machine learning models are recognized as useful tools for prediction applications, but the algorithmic complexity of some models causes interpretation challenges. Explainability methods have been proposed to provide insight into these models, but there is little research focused on supervised modeling with functional data inputs. We argue that, especially in applications of high consequence, it is important to explicitly model the functional dependence in a black-box analysis to not obscure or misrepresent patterns in explanations. As such, we propose the V ariable importance E xplainable E lastic S hape A nalysis (VEESA) pipeline for training supervised machine learning models with functional inputs. The pipeline is an analysis process that includes the data preprocessing, modeling, and post-hoc explanations. The preprocessing is done using elastic functional principal components analysis, which accounts for vertical and horizontal variability in functional data and, ultimately, allows for explanations in the original data space that identify the important functional variability without bias due to correlated variables. Here, we demonstrate the pipeline on two high-consequence applications: explosives classification for national security and inkjet printer identification in forensic science. The applications exhibit the VEESA pipeline’s ability to provide an understanding of the characteristics of the functional data useful for prediction. Code for implementing the pipeline is available in the veesa R package (and supplemental python code).

Elastic Shape Analysis↗
NASA NTRS1972

Variability of plasma-line enhancement in ionospheric modification experiments.

A simple explanation for the variations of plasma-line intensity is suggested. The explanation is based on the fact that the plasma waves responsible for scattering the radar waves occur over a very limited range of heights. The explanation further makes use of the fact that the position of these height ranges of generation depends primarily on the gradient of the number density and to a lesser extent on the temperature and the orientation of the diagnostic radar beam.

Fejer, J. A.↗
NASA NTRS1975

de Sitter's theory 'melts' Europa's polar cap

In photoelectric observations of a series of occultations of the Jovian satellite Europa by the Jovian satellite Io it was observed that increasingly larger fractions of the north polar region of the former were covered at mid-event with each successive occultation. One of the possible explanations of the observed phenomena involves the existence of a bright north polar cap for Europa. The reported investigation shows that a second explanation may be more plausible. This explanation is related to the possibility that at mid-event the projected minimum distance between the centers of the satellites was smaller than predicted by Sampson's theory.

Aksnes, K.↗
NASA NTRS1979

Static stability and thermal wind in an atmosphere of variable composition Applications to Mars

Radiometric measurements of the temperature of the south polar cap of Mars in winter have yielded values significantly below the expected 148 K. One proposed explanation for this result is a substantial reduction in the CO2 content of the atmosphere and a lowering of the mean molecule weight near the surface. The meteorological consequences of this explanation are explored by deriving a criterion for vertical static stability and a thermal wind law for an atmosphere of variable composition. The atmosphere proves to be statically unstable unless the anomaly in the CO2 mixing ratio extends to heights of tens of kilometers. The effect of varying molecular weight exceeds the effect of temperature gradient, producing shears with height of reversed sign. The shears are baroclinically unstable, and this instability would eradicate the latitudinal gradient of molecular weight. This inconsistency can be resolved by invoking a reasonable elevation of the central polar cap and by imposing an adequate zonal wind. It is concluded that if the explanation requiring a change in atmospheric composition is correct, it must be accompanied by other special circumstances to make it meteorologically consistent.

Hess, S. L.↗
NASA NTRS1980

HCMM: Soil moisture in relation to geologic structure and lithology, northern California

The author has identified the following significant results. Detailed examination of Nite-IR images of intermontane basins in arid and/or semiarid areas of California discloses a ring or halo of relatively lighter greytone around the edges of each basin. Intermontane basins in the Northern Coast Range, however, do not show this thermal haloing. The topographic elevation of the haloes in arid basins shows seasonal variation, but it is present on nearly all images. A similar halo encircles many of the volcanoes on the Modoc Plateau and Southern Cascade Range. Although the halo around the arid intermontane basins can possibly be explained in relation to the location of alluvial fans (and perhaps water content of the rocks), a similar explanation cannot be made for the haloes around volcanoes or for the lack of haloes around basins in the Coast Range. Atmospheric thermal layering may be an alternative explanation; however, this explanation is also riddled with inconsistencies.

Rich, E. I.↗
NASA NTRS1982

Narrow components in the profiles of ultraviolet resonance lines - Evidence for a two-component stellar wind for O and B stars

The presence of narrow, shifted absorption features superposed on the wide P Cygni profiles of the UV resonance lines in the spectra of O and B stars is studied and possible explanations of the narrow components are discussed. Spectra from 26 stars of spectral types O4f to B1 Ib were examined for the presence of narrow components in the absorption part of the profiles, and positive evidence was detected in 17. The central velocities of the features are found to be similar for different ions of the same star, and typically equal to 0.75 times the terminal velocity of the stellar wind. The width of the narrow absorptions is about 18% of the terminal stellar wind velocity, with a column density generally not correlated with the effective temperature or mass loss rate. Possible explanations for the narrow components include peaks in the degree of ionization, a stationary stellar shell, a plateau in the stellar wind velocity law, a decelerating stellar wind, a variable mass loss rate or a two component stellar wind with low-velocity, low-density material found at a distance greater than 2 stellar radii, which explanation appears most likely.

Lamers, H. J. G. L. M.↗
NASA NTRS1986

Crew Communication as a Factor in Aviation Accidents

The crew communication process is analyzed. Planning and explanation are shown to be well-structured discourse types, described by formal rules. These formal rules are integrated with those describing the other most important discourse type within the cockpit: the command-and-control speech act chain. The latter is described as a sequence of speech acts for making requests (including orders and suggestions), for making reports, for supporting or challenging statements, and for acknowledging previous speech acts. Mitigation level, a linguistic indication of indirectness and tentativeness in speech, was an important variable in several hypotheses, i.e., the speech of subordinates is more mitigated than the speech of superiors, the speech of all crewmembers is less mitigated when they know that they are in either a problem or emergency situation, and mitigation is a factor in failures of crewmembers to initiate discussion of new topics or have suggestions ratified by the captain. Test results also show that planning and explanation are more frequently performed by captains, are done more during crew- recognized problems, and are done less during crew-recognized emergencies. The test results also indicated that planning and explanation are more frequently performed by captains than by other crewmembers, are done more during crew-recognized problems, and are done less during-recognized emergencies.

Goguen, J.↗
NASA NTRS1987

On the Holmberg asymmetry of the satellites of disk galaxies

A tidal explanation for the observation by Holmberg that the satellites of edge-on disk galaxies tend to avoid the sector within + or - 30 deg of the major axes of the disk galaxies is considered. It is shown that satellites with small orbit-to-disk inclinations are likely to become compact and consequently be left out in Holmberg's survey due to the resemblance to stars. The explanation is supported by the observation of an excess of compact galaxies near the major-axis direction of edge-on disk galaxies. The disk tidal explanation also predicts that the asymmetry should be weaker with larger satellites. It is found that the Karachentsev (1972, 1980) binary galaxy sample, where the typical companion is comparable to the primary galaxy, shows no Holmberg effect. The case of M32 as a compact satellite of the Andromeda galaxy is discussed as a nearby observationally supported example of the above processes.

Byrd, Gene G.↗
NASA NTRS1988

Knowledge representation issues for explaining plans

Explanations are recognized as an important facet of intelligent behavior. Unfortunately, expert systems are currently limited in their ability to provide useful, intelligent justifications of their results. We are currently investigating the issues involved in providing explanation facilities for expert planning systems. This investigation addresses three issues: knowledge content, knowledge representation, and explanation structure.

Prince, Mary Ellen↗