Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “classification problem”

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 667 records · Page 37
All sourcesNASA NTRSDOE OSTI
NASA NTRS2008

An Automatic Cloud Mask Algorithm Based on Time Series of MODIS Measurements

Quality of aerosol retrievals and atmospheric correction depends strongly on accuracy of the cloud mask (CM) algorithm. The heritage CM algorithms developed for AVHRR and MODIS use the latest sensor measurements of spectral reflectance and brightness temperature and perform processing at the pixel level. The algorithms are threshold-based and empirically tuned. They don't explicitly address the classical problem of cloud search, wherein the baseline clear-skies scene is defined for comparison. Here, we report on a new CM algorithm which explicitly builds and maintains a reference clear-skies image of the surface (refcm) using a time series of MODIS measurements. The new algorithm, developed as part of the Multi-Angle Implementation of Atmospheric Correction (MAIAC) algorithm for MODIS, relies on fact that clear-skies images of the same surface area have a common textural pattern, defined by the surface topography, boundaries of rivers and lakes, distribution of soils and vegetation etc. This pattern changes slowly given the daily rate of global Earth observations, whereas clouds introduce high-frequency random disturbances. Under clear skies, consecutive gridded images of the same surface area have a high covariance, whereas in presence of clouds covariance is usually low. This idea is central to initialization of refcm which is used to derive cloud mask in combination with spectral and brightness temperature tests. The refcm is continuously updated with the latest clear-skies MODIS measurements, thus adapting to seasonal and rapid surface changes. The algorithm is enhanced by an internal dynamic land-water-snow classification coupled with a surface change mask. An initial comparison shows that the new algorithm offers the potential to perform better than the MODIS MOD35 cloud mask in situations where the land surface is changing rapidly, and over Earth regions covered by snow and ice.

Lyapustin, Alexei↗
NASA NTRSDate not supplied

QuantifyML: How good is my machine learning model?

This paper presents an approach, QuantifyML, which employs model counting to assess the learnability and robustness of machine learning models. Typically the efficacy of machine learning models is determined by computing their accuracy statistically on test data sets. However, this may be misleading, if the test data is not representative of the problem that is being studied. Further, two different models may have the same accuracy on a given data set, measured statistically, but may be very different in their behavior on unseen data. Also, models with high accuracy could have poor adversarial robustness. In QuantifyML, our goal is to precisely quantify the extent to which machine learning models have learned and generalized from the given data. In QuantifyML, a trained model is translated into a C program, which is fed to the CBMC model checking tool to produce a formula in Conjunctive Normal Form (CNF), which in turn is analyzed with state-of-the-art model counters to efficiently obtain precise counts w.r.t different outputs. QuantifyML enables i) evaluating the learnability of models by comparing the counts for the outputs to ground truth, expressed as logical predicates (if available), ii) comparing the performance of different models that may be built with different machine learning algorithms (e.g., decision-trees vs. neural networks), and iii) quantifying the robustness of trained models around given inputs. Our evaluation demonstrates these applications of QuantifyML on decision trees and neural networks trained to learn relational properties of graphs, for which we know the ground truth, and to perform image classification, for which we do not have the ground truth, but we can quantify local robustness.

Deep Neural Networks↗
DOE OSTI2023

What Machine Learning Can and Cannot Do for Inertial Confinement Fusion

Machine learning methodologies have played remarkable roles in solving complex systems with large data, well-defined input–output pairs, and clearly definable goals and metrics. The methodologies are effective in image analysis, classification, and systems without long chains of logic. Recently, machine-learning methodologies have been widely applied to inertial confinement fusion (ICF) capsules and the design optimization of OMEGA (Omega Laser Facility) capsule implosion and NIF (National Ignition Facility) ignition capsules, leading to significant progress. As machine learning is being increasingly applied, concerns arise regarding its capabilities and limitations in the context of ICF. ICF is a complicated physical system that relies on physics knowledge and human judgment to guide machine learning. Additionally, the experimental database for ICF ignition is not large enough to provide credible training data. Most researchers in the field of ICF use simulations, or a mix of simulations and experimental results, instead of real data to train machine learning models and related tools. They then use the trained learning model to predict future events. This methodology can be successful, subject to a careful choice of data and simulations. However, because of the extreme sensitivity of the neutron yield to the input implosion parameters, physics-guided machine learning for ICF is extremely important and necessary, especially when the database is small, the uncertain-domain knowledge is large, and the physical capabilities of the learning models are still being developed. In this work, we identify problems in ICF that are suitable for machine learning and circumstances where machine learning is less likely to be successful. This study investigates the applications of machine learning and highlights fundamental research challenges and directions associated with machine learning in ICF.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Resilience Design Patterns: A Structured Approach to Resilience at Extreme Scale (V.2.0)

Reliability is a serious concern for future extreme-scale high-performance computing (HPC) systems. Projections based on the current generation of HPC systems and technology roadmaps suggest the prevalence of very high fault rates in future systems. The errors resulting from these faults will propagate and generate various kinds of failures, which may result in outcomes ranging from result corruptions to catastrophic application crashes. Therefore, the resilience challenge for extreme-scale HPC systems requires coordination between various hardware and software technologies that are capable of handling a broad set of fault models at accelerated fault rates. Also, due to practical limits on power consumption in future HPC systems, they are likely to embrace innovative architectures, increasing the levels of hardware and software complexities. Therefore, the techniques that seek to improve resilience must navigate the complex trade-off space between resilience and the overheads to power consumption and performance. While the HPC community has developed various resilience solutions, application-level techniques as well as system-based solutions, the solution space of HPC resilience techniques remains fragmented. There are no formal methods to integrate the various HPC resilience techniques into composite solutions, nor are there methods to holistically evaluate the adequacy and efficacy of such solutions in terms of their protection coverage, and their performance & power efficiency characteristics. Additionally, few implementations of current resilience solutions are portable to newer architectures and software environments that will be deployed on future systems. We developed a new structured approach to the management of HPC resilience using the concept of resilience-based design patterns. In general, a design pattern is a repeatable solution to a commonly occurring problem. We identified the well-known solutions that are commonly used to deal with faults, errors and failures in HPC systems. In the initial design patterns specification (version 1.0), we described the various solutions, which address specific problems in the design of resilient HPC environments, in the form of patterns. Each pattern describes a problem caused by a fault, error or failure event in an HPC environment, and then describes the core of the solution of the problem in such a way that this solution may be adapted to different systems and implemented at different layers of the system stack. The catalog of these resilience design patterns provides designers with a collection of design elements. To construct complete resilience solutions using combinations of various patterns, we defined a framework that enhances HPC designers' understanding of the important constraints and the opportunities for the design patterns to be implemented and deployed at various layers of the system stack. The design framework is also useful for establishing interfaces and mechanisms to coordinate flexible fault management across hardware and software components, as well as to consider the trade-off between performance, resilience, and power consumption when constructing a solution. The resilience design patterns specification version 1.1 included more detailed explanations of the pattern solutions, the context in which the patterns are applicable, and the implications for hardware or software design. It also provided several additional examples and detailed case studies to demonstrate the use of patterns to build realistic solutions. In version 1.2 of the specification document, we have improved the pattern descriptions, including graphical representations of the pattern components. These improvements are largely based on critical comments, feedback and suggestions received from pattern experts and readers of the previous versions of the specification. The pattern classification has been modified to further clarify the relationships between pattern categories. This version of the specification also introduces a pattern language for resilience design patterns. The pattern language presents the patterns in the catalog as a network, revealing the relations among the resilience patterns. The language provides designers with the means to explore alternative techniques for handling a specific fault model that may have different efficiency and complexity characteristics. Using the pattern language also enables the design and implementation of comprehensive resilience solutions as a set of interconnected resilience patterns that can be instantiated across layers of the system stack. The overall goal of this work is to provide hardware and software designers, as well as the users and operators of HPC systems, a systematic methodology for the design and evaluation of resilience technologies in HPC systems that keep scientific applications running to a correct solution in a timely and cost-efficient manner despite frequent faults, errors, and failures of various types. Version 2.0 expands the resilience design pattern classification and catalog to include self-stabilization patterns and reliability, availability and performance models for each structural pattern.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1998

Tone Noise and Nearfield Pressure Produced by Jet-Cavity Interaction

Cavity flow resonance can cause numerous problems in aerospace applications. While our long-term goal is to understand cavity flows well enough to devise effective cavity resonance suppression techniques, this paper describes a fundamental study of resonant tones produced by jet-cavity interaction at subsonic and supersonic speeds. Our specific jet-cavity configuration can also be used as a test bed for evaluating active and passive flow resonance control concepts. Two significant findings emerge from this study. 1) Originally, we expected that tones produced by jet-cavity interaction would resemble cavity tones or jet tones or would involve some simple combinations of each. The experimental data do not support these expectations: instead, the jet cavity interaction produce a unique set of tones. We propose simple yet and physically insightful correlations for these tones. Although the pressure patterns on the cavity floor display very complex variations with the Mach number for a length/depth = 8 cavity, the tones correspond to the acoustic modes of the cavity-independent of flow. For a length/ depth = 3 cavity, however, a surprise emerges: the pressure patterns on the cavity floor are not so complex but the tones depend significantly on the flow. Additionally, we examine the role of external feedback unique to jet-cavity interaction. 2) Previous research led us to expect that traditional classifications (open, transitional, or closed) for cavities in an infinite flight stream would be insensitive to small changes in Mach number and would depend primarily on cavity length/depth ratios. Use of the novel high resolution photoluminescent pressure sensitive paint shows that the classifications are actually quite sensitive to jet Mach number for a length/depth = 8 cavity. However, these classifications provide no guidance whatsoever for tone amplitude or frequency. Detailed experimental data and insights presented here will assist researchers who are performing numerical simulations of jet-cavity flows as a first step toward devising resonance suppression methods.

Raman, Ganesh↗
DOE OSTI2022

Assessing the Effectiveness of Generalized Likelihood Ratio Test Detector Schemes in Seismic Event Detection and the Avoidance of Nontarget Signals

Cross-correlation techniques have played a long-standing and pivotal role in seismic event monitoring. However, the performance of correlation-based detectors is challenged by nuisance seismicity, or nontarget signals. Such detections are a problem when the mission is to automatically map events to the correct source region. Using aftershocks of the 2014 $M_w$ 6.0 South Napa, California, earthquake, we demonstrate the effectiveness of utilizing a dynamic correlation processor framework in a generalized likelihood ratio test (GLRT) detector configuration to minimize nontarget detections. A GLRT maximizes a detection statistic with respect to one or more unknown parameters. In this case, the detection statistic is a template signal match against the waveform in a window sliding over a data stream, and the unknown parameter is an index variable indicating group membership of the template event or events. Detected events are assigned to the event group that yields the largest detection statistic. In this work, our results show that a GLRT detector will outperform a suite of independently operating correlation and subspace detectors in terms of having a lower nontarget detection rate at a given missed detection rate. We also show that a GLRT detector composed of a few high-rank subspace detectors has a slightly higher nontarget detection rate, but a significantly lower missed detection rate, than a GLRT detector composed of many low-rank subspace detectors. The high-rank GLRT configuration produced impressive results even with marginal data (single channel, single station, and very low time bandwidth product), which bodes well for the utility of building efficient aftershock classification systems and global monitoring systems at larger scales. However, future work is required to assess performance at the regional scale and to assess the performance of the system at detecting target events not used in the detector template creation.

58 GEOSCIENCES↗
NASA NTRS1993

A new method for recognizing quadric surfaces from range data and its application to telerobotics and automation

The problem of recognizing and positioning of objects in three-dimensional space is important for robotics and navigation applications. In recent years, digital range data, also referred to as range images or depth maps, have been available for the analysis of three-dimensional objects owing to the development of several active range finding techniques. The distinct advantage of range images is the explicitness of the surface information available. Many industrial and navigational robotics tasks will be more easily accomplished if such explicit information can be efficiently interpreted. In this research, a new technique based on analytic geometry for the recognition and description of three-dimensional quadric surfaces from range images is presented. Beginning with the explicit representation of quadrics, a set of ten coefficients are determined for various three-dimensional surfaces. For each quadric surface, a unique set of two-dimensional curves which serve as a feature set is obtained from the various angles at which the object is intersected with a plane. Based on a discriminant method, each of the curves is classified as a parabola, circle, ellipse, hyperbola, or a line. Each quadric surface is shown to be uniquely characterized by a set of these two-dimensional curves, thus allowing discrimination from the others. Before the recognition process can be implemented, the range data have to undergo a set of pre-processing operations, thereby making it more presentable to classification algorithms. One such pre-processing step is to study the effect of median filtering on raw range images. Utilizing a variety of surface curvature techniques, reliable sets of image data that approximate the shape of a quadric surface are determined. Since the initial orientation of the surfaces is unknown, a new technique is developed wherein all the rotation parameters are determined and subsequently eliminated. This approach enables us to position the quadric surfaces in a desired coordinate system. Experiments were conducted on raw range images of spheres, cylinders, and cones. Experiments were also performed on simulated data for surfaces such as hyperboloids of one and two sheets, elliptical and hyperbolic paraboloids, elliptical and hyperbolic cylinders, ellipsoids and the quadric cones. Both the real and simulated data yielded excellent results. Our approach is found to be more accurate and computationally inexpensive as compared to traditional approaches, such as the three-dimensional discriminant approach which involves evaluation of the rank of a matrix. Finally, we have proposed one other new approach, which involves the formulation of a mapping between the explicit and implicit forms of representing quadric surfaces. This approach, when fully realized, will yield a three-dimensional discriminant, which will recognize quadric surfaces based upon their component surfaces patches. This approach is faster than prior approaches and at the same time is invariant to pose and orientation of the surfaces in three-dimensional space.

Alvertos, Nicolas↗
DOE OSTI2023

Observations and Machine-Learned Models of Near-Surface Permafrost along the Koyukuk River, Alaska, USA

This dataset contains GeoTIFs (raster) and GeoPackages (vector) that map observations of near-surface permafrost and not-permafrost from a field campaign conducted near the village of Huslia, AK along the Koyukuk River and its floodplain in July 2018. These data were collected as part of a campaign to understand if and how permafrost impacts riverbank erosion. This problem cannot be assessed without knowing where permafrost exists. Permafrost was observed via frost probing (to a maximum depth of one meter), coring (to a maximum depth of two meters) and bank/bar excavations. An additional boat survey was performed wherein expert (Joel Rowland) judgment assessed the presence or absence of distinctive permafrost features (e.g., overhanging tundra mats, thermoerosional niching, ice wedges, active drainage of ice melt from soils). This dataset also contains the input features and results of two machine learning models (random forest and convolutional neural network) that extrapolate the observations to the full floodplain that may be useful for building, testing, or validating other machine-learned permafrost models. Permafrost data are provided as georasters of the same shape and geovectors (polylines/polygons) and are all projected into EPSG:32605. All data can be visualized with a GIS (QGIS, ArcGIS, etc.).

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1965

Ground testing at MSFC

The large-scale field collection of acoustic data and their analysis are described in this review. The entire sound-field program has the purpose of solving complex problems of excessive noise propagation associated with the static test firing of large space-booster powerplants. The site of data collection has been Mississippi Test Facility (MTF), where far-field sound-propagation and correlated meteorological measurements were initiated in 1962. The statistical evaluation of sound-measurement data shows that the polar distribution of sound in the springtime (at MTF) is determined mainly by the temperature field of the atmosphere. Greater deformation of sound-intensity polar distribution occurs in the autumn as a result of the influence of the wind field. The data on sound attenuation as a function of distance from its source revealed attenuation in excess of the expected magnitude. The excess was mainly between 1 and 10 kilometers from the source, with maximum attenuation (17 to 18 decibels) near 6 kilometers. Excess sound attenuation was compared with calculated sound dispersion. For all sound frequencies in all seasons the deviation between reference and empirical attenuation values formed a bell-shaped distribution on a semilogarithmic plot. The physical causes for the sound- attenuation behavior are considered to be atmospheric diffraction and scattering effects. Statistical analysis of sod-intensity extremes indicated amplification of the sound signals by focusing which could be expected along selected azimuths. Empirical amplification values showed the maximum occurring in winter (50 decibels at 22 kilometers). For the analysis of correlation between empirical sound data and local atmospheric sound- propagation conditions, #e concept of analytical classification of sound-velocity profiles has been developed. Examination of over a quarter-million individual sound-velocity profiles indicates that 33 profile types for heights up to 3 kilometers constitute the empirical possibilities of sound-propagation conditions in southeastern United States. Since the focusing qualities of each sound- velocity profile can be determined as a class characteristic, the overall probabilities of focusing occurrence can he obtained. The severity of focusing, as based upon lateral spread of focusing areas, has been determined. The data for MTF indicate focusing is most severe in winter, with the direction toward east being most affected. Sound propagation is a fluctuating phenomenon because atmospheric conditions vary continuously. A measuring system for analyzing the nonstationary sound-propagation phenomena has been established. The analysis of the acoustic fluctuation data is sup- posed to provide "dynamic” perturbation functions of sound, which are to be superimposed on the related characteristic sound-velocity-profile type. The ultimate result of the program will be the establishment of reliable contingency tables showing the relationship between atmospheric parameters and far-field sound intensities. These data will be the basis for more accurate sound-propagation forecasting for static firing tests.

STATIC TESTING↗
DOE OSTI2021

NENCI-2021. I. A large benchmark database of non-equilibrium non-covalent interactions emphasizing close intermolecular contacts

In this work, we present NENCI-2021, a benchmark database of ~8000 Non-Equilibirum Non-Covalent Interaction energies for a large and diverse selection of intermolecular complexes of biological and chemical relevance. To meet the growing demand for large and high-quality quantum mechanical data in the chemical sciences, NENCI-2021 starts with the 101 molecular dimers in the widely used S66 and S101 databases and extends the scope of these works by (i) including 40 cation–π and anion–π complexes, a fundamentally important class of non-covalent interactions that are found throughout nature and pose a substantial challenge to theory, and (ii) systematically sampling all 141 intermolecular potential energy surfaces (PESs) by simultaneously varying the intermolecular distance and intermolecular angle in each dimer. Designed with an emphasis on close contacts, the complexes in NENCI-2021 were generated by sampling seven intermolecular distances along each PES (ranging from 0.7× to 1.1× the equilibrium separation) and nine intermolecular angles per distance (five for each ion–π complex), yielding an extensive database of 7763 benchmark intermolecular interaction energies (E int ) obtained at the coupled-cluster with singles, doubles, and perturbative triples/complete basis set [CCSD(T)/CBS] level of theory. The E int values in NENCI-2021 span a total of 225.3 kcal/mol, ranging from -38.5 to +186.8 kcal/mol, with a mean (median) E int value of -1.06 kcal/mol (-2.39 kcal/mol). In addition, a wide range of intermolecular atom-pair distances are also present in NENCI-2021, where close intermolecular contacts involving atoms that are located within the so-called van der Waals envelope are prevalent—these interactions, in particular, pose an enormous challenge for molecular modeling and are observed in many important chemical and biological systems. A detailed symmetry-adapted perturbation theory (SAPT)- based energy decomposition analysis also confirms the diverse and comprehensive nature of the intermolecular binding motifs present in NENCI-2021, which now includes a significant number of primarily induction-bound dimers (e.g., cation–π complexes). NENCI-2021 thus spans all regions of the SAPT ternary diagram, thereby warranting a new four-category classification scheme that includes complexes primarily bound by electrostatics (3499), induction (700), dispersion (1372), or mixtures thereof (2192). A critical error analysis performed on a representative set of intermolecular complexes in NENCI-2021 demonstrates that the E int values provided herein have an average error of ±0.1 kcal/mol, even for complexes with strongly repulsive E int values, and maximum errors of ±0.2–0.3 kcal/mol (i.e., ~±1.0 kJ/mol) for the most challenging cases. For these reasons, we expect that NENCI-2021 will play an important role in the testing, training, and development of next-generation classical and polarizable force fields, density functional theory approximations, wavefunction theory methods, and machine learning based intra- and inter-molecular potentials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Toward Consistent High-Fidelity Quantum Learning on Unstable Devices via Efficient In-Situ Calibration

In the near-term noisy intermediate-scale quantum (NISQ) era, high noise will significantly reduce the fidelity of quantum computing. What's worse, recent works reveal that the noise on quantum devices is not stable, that is, the noise is dynamically changing over time. This leads to an imminent challenging problem: At run-time, is there a way to efficiently achieve a consistent high-fidelity quantum system on unstable devices? To study this problem, we take quantum learning (a.k.a., variational quantum algorithm) as a vehicle, which has a wide range of applications, such as combinatorial optimization and machine learning. A straightforward approach is to optimize a variational quantum circuit (VQC) with a parameter-shift approach on the target quantum device before using it; however, the optimization has an extremely high time cost, which is not practical at run-time. To address the pressing issue, in this paper, we proposed a novel quantum pulse-based noise adaptation framework, namely QuPAD. In the proposed framework, first, we identify that the CNOT gate is the fidelity bottleneck of the conventional VQC, and we employ a more robust parameterized multi-qubit gate (i.e., Rzx gate) to replace CNOT gate. Second, by benchmarking Rzx gate with different parameters, we build a fitting function for each coupling qubit pair, such that the deviation between the theoretic output of Rzx gate and its on-device output under a given pulse amplitude and duration can be efficiently predicted. On top of this, an evolutionary algorithm is devised to identify the pulse amplitude and duration of each Rzx gate (i.e., calibration) and find the quantum circuits with high fidelity. Experiments show that the runtime on quantum devices of QuPAD with 8–10 qubits is less than 15 minutes, which is up to 270 x faster than the parameter-shift approach. In addition, compared to the vanilla VQC as a baseline, QuPAD can achieve 59.33% accuracy gain on a classification task, and average 66.34% closer to ground state energy for molecular simulation.

Hu, Zhirui↗
NASA NTRS1989

Size distribution of dust grains: A problem of self-similarity

Distribution functions describing the results of natural processes frequently show the shape of power laws, e.g., mass functions of stars and molecular clouds, velocity spectrum of turbulence, size distributions of asteroids, micrometeorites and also interstellar dust grains. It is an open question whether this behavior is a result simply coming about by the chosen mathematical representation of the observational data or reflects a deep-seated principle of nature. The authors suppose the latter being the case. Using a dust model consisting of silicate and graphite grains Mathis et al. (1977) showed that the interstellar extinction curve can be represented by taking a grain radii distribution of power law type n(a) varies as a(exp -p) with 3.3 less than or equal to p less than or equal to 3.6 (example 1) as a basis. A different approach to understanding power laws like that in example 1 becomes possible by the theory of self-similar processes (scale invariance). The beta model of turbulence (Frisch et al., 1978) leads in an elementary way to the concept of the self-similarity dimension D, a special case of Mandelbrot's (1977) fractal dimension. In the frame of this beta model, it is supposed that on each stage of a cascade the system decays to N clumps and that only the portion beta N remains active further on. An important feature of this model is that the active eddies become less and less space-filling. In the following, the authors assume that grain-grain collisions are such a scale-invarient process and that the remaining grains are the inactive (frozen) clumps of the cascade. In this way, a size distribution n(a) da varies as a(exp -(D+1))da (example 2) results. It seems to be highly probable that the power law character of the size distribution of interstellar dust grains is the result of a self-similarity process. We can, however, not exclude that the process leading to the interstellar grain size distribution is not fragmentation at all. It could be, e.g., diffusion-limited growth discussed by Sander (1986), who applied the theory of fractal geometry to the classification of non-equilibrium growth processes. He received D=2.4 for diffusion-limited aggregation in 3d-space.

Henning, TH.↗
NASA NTRS1994

Does human cognition allow Human Factors (HF) certification of advanced aircrew systems?

This paper has examined the requirements of HF specification and certification within advanced or complex aircrew systems. It suggests reasons for current inadequacies in the use of HF in the design process, giving some examples in support, and suggesting an avenue towards the improvement of the HF certification process. The importance of human cognition to the operation and performance of advanced aircrew systems has been stressed. Many of the shortfalls of advanced aircrew systems must be attributed to over automated designs that show little consideration on either the mental limits or the cognitive capabilities of the human system component. Traditional approaches to system design and HF certification are set within an over physicalistic foundation. Also, traditionally it was assumed that physicalistic system functions could be attributed to either the human or the machine on a one to one basis. Moreover, any problems associated with the parallel needs, or promoting human understanding alongside system operation and direction, were generally equated in reality by the natural flexibility and adaptability of human skills. The consideration of the human component of a complex system is seen as being primarily based on manifestations of human behavior to the almost total exclusion of any appreciation of unobservable human mental and cognitive processes. The argument of this paper is that the considered functionality of any complex human-machine system must contain functions that are purely human and purely cognitive. Human-machine system reliability ultimately depends on human reliability and dependability and, therefore, on the form and frequency of cognitive processes that have to be conducted to support system performance. The greater the demand placed by an advanced aircraft system on the human component's basic knowledge processes or cognition, rather than on skill, the more insiduous the effects the human may have on that system. This paper discusses one example of an attempt to devise an improved method of specificaiton and certification with relation to the advanced aircrew system, that of the RN Merlin helicopter. The method is realized to have limitations in practice, these mainly associated with the late production of the system specification in relation to the system development process. The need for a careful appreciation of the capabilities and support needs of human cognition within the design process of a complex man machine system has been argued, especially with relation to the concept of system functionality. Unlike the physicalistic Fitts list, a new classification of system functionality is proposed, namely: (1) equipment - system equipment related; (2) cognitive - human cognition related; and (3) associated - necessary combinatin of equipment and cognitive. This paper has not proposed a method for a fuller consideration of cognition within systems design, but has suggested the need for such a method and indicated an avenue towards its development. Finally, the HF certification of advanced aircrew systems is seen as only being possible in a qualified sense until the important functions of human cognition are considered within the system design process. (This paper contains the opinions of its authors and does not necessarily refledt the standpoint of their respective organizations).

Macleod, Iain S.↗
NASA NTRS1994

Compositional Variation in Apollo 16 Impact-Melt Breccias and Inferences for the Geology and Bombardment History of the Central Highlands of the Moon

High-precision data for the concentrations of a number of lithophile and siderophile elements were obtained on multiple subsamples from 109 impact-melt rocks and breccias (mostly crystalline) from the Apollo 16 site. Compositions of nearly all Apollo 16 melt rocks fall on one of two trends of increasing Sm concentration with increasing Sc concentration. The Eastern trend (lower Sm/Sc, Mg/Fe, and Sm/Yb ratios) consists of compositional groups 3 and 4 of previous classification schemes. These melt rocks are feldspathic, poor in incompatible and siderophile elements, and appear to have provenance in the Descartes formation to the east of the site. The Western trend (higher Sm/Sc. Mg/Fe, and Sm/ Yb ratios) consists of compositional groups 1 and 2. These relatively mafic, KREEP-bearing breccias are a major component (approx.35%) of the Cayley plains west of the site and are unusual, compared to otherwise similar melt breccias from other sites, in having high concentrations of Fe-Ni metal ( 1-2 %). The metal is the carrier of the low-Ir/Au (approx. 0.3 x chondritic) siderophile-element signature that is characteristic of the Apollo 16 site. Four compositionally distinct groups (1M, 1F, 2DB, and 2NR) of Western-trend melt breccias occur that are each represented by at least six samples. Compositional group 1 or previous classification schemes (the 'poikilitic' or 'LKFM' melt breccias) can be subdivided into two groups. Group 1M (represented by six samples, including 60315) is characterized by lower Al2O3 concentrations, higher MgO and alkali concentrations, and higher Mg/Fe and Cr/Sc ratios than group 1F (represented by fifteen samples, including 65015). Group 1M also has siderophile-element concentrations averaging about twice those of group lF and Ir/Au and Ir/Ni ratios that are even lower than those of other Western-trend melt rocks (Ir/Au = 0.24 +/- 0.03. CI-normalized). At the mafic extreme of group 2 ('VHA' melt breccias), the melt lithology occurring as clasts in feldspathic fragmental breccias from North Ray crater (group 2NR) is compositionally distinct from the melt lithology ofdimict breccias from the Cayley plains (group 2DB) in having higher concentrations of Sc, Cr, and heavy rare earth elements and lower concentrations of siderophile elements. The distinct siderophile-element signature (high absolute abundances, low Ir/Au ratio) suggest that the four groups ofmafic melt breccia are all somehow related. Ratios ofsome lithophile elements also suggest that they are more closely related to each other than then, are to melt breccias from other Apoll sites. However, none of the breccia compositions can be related to any of the others by any simple process of igneous fractionation or mixing involving common lunar materials. Thus, the origin of the four groups of mafic melt breccia is enigmatic. If they were produced in only one or two impacts, then a mechanism exists for generating regimes of impact-melt breccia in a single impact that are substantially different from each other in composition. For various reasons, including the problem of delivering large volumes of four different types of melt to the Apollo 16 site, it is unlikely that any of these breccias were produced in basin-forming impacts. If they were produced in as many as four crater-forming impacts, then the unusual siderophile-element signature is difficult to explain. Possible explanations are (1) the four groups of melt breccia all contain metal from a single, earlier impact, (2) they were each formed by related metal-rich meteoroids, or (3) some common postimpact process has resulted in metal of similar composition in each of four melt pools. Within a compositional group, most intrasample and intersample variation in lithophile element concentrations is caused by differences among samples in the proportion of a component of normative anorthosite or noritic anorthosite. In most cases, this compositional variation probably reflects variation in clast abundance. For group 2DB (and probably 2NR), differences in abundance of a component of ferroan anorthosite (estimated Al2O3 approx. 32%) accounts for the compositional variation. For groups 1M and 1F, the anorthositic component is more mafic (estimated Al203 approx. 26%). Some group-2 samples may be related by a troctolitic component of varying abundance.

Korotev, Randy L.↗
DOE OSTI2024

Harnessing Machine Learning and Data Fusion for Accurate Undocumented Well Identification in Satellite Images

This study utilizes satellite data to detect undocumented oil and gas wells, which pose significant environmental concerns, including greenhouse gas emissions. Three key findings emerge from the study. Firstly, the problem of imbalanced data is addressed by recommending oversampling techniques like Rotation–GaussianBlur–Solarization data augmentation (RGS), the Synthetic Minority Over-Sampling Technique (SMOTE), or ADASYN (an extension of SMOTE) over undersampling techniques. The performance of borderline SMOTE is less effective than that of the rest of the oversampling techniques, as its performance relies heavily on the quality and distribution of data near the decision boundary. Secondly, incorporating pre-trained models trained on large-scale datasets enhances the models’ generalization ability, with models trained on one county’s dataset demonstrating high overall accuracy, recall, and F1 scores that can be extended to other areas. This transferability of models allows for wider application. Lastly, including persistent homology (PH) as an additional input improves performance for in-distribution testing but may affect the model’s generalization for out-of-distribution testing. A careful consideration of PH’s impact on overall performance and generalizability is recommended. Overall, this study provides a robust approach to identifying undocumented oil and gas wells, contributing to the acceleration of a net-zero economy and supporting environmental sustainability efforts.

SMOTE↗
NASA NTRS2010

Human Factors Research for Space Exploration: Measurement, Modeling, and Mitigation

As part of NASA's Human Research Program, the Space Human Factors Engineering Project serves as the bridge between Human Factors research and Human Spaceflight applications. Our goal is to be responsive to the operational community while addressing issues at a sufficient level of abstraction to ensure that our tools and solutions generalize beyond the point design. In this panel, representatives from four of our research domains will discuss the challenges they face in solving current problems while also enabling future capabilities. Historically, engineering-dominated organizations have tended to view good Human Factors (HF) as a desire rather than a requirement in system design and development. Our field has made significant gains in the past decade, however; the Department of Defense, for example, now recognizes Human-System Integration (HSI), of which HF is a component, as an integral part of their divisions hardware acquisition processes. And our own agency was far more accepting of HF/HSI requirements during the most recent vehicle systems definition than in any prior cycle. Nonetheless, HF subject matter experts at NASA often find themselves in catch up mode... coping with legacy systems (hardware and software) and procedures that were designed with little regard for the human element, and too often with an attitude of we can deal with any operator issues during training. Our challenge, then, is to segregate the true knowledge gaps in Space Human Factors from the prior failures to incorporate best (or even good) HF design principles. Further, we strive to extract the overarching core HF issues from the point-design-specific concerns that capture the operators (and managers) attention. Generally, our approach embraces a 3M approach to Human Factors: Measurement, Modeling, and Mitigation. Our first step is to measure human performance, to move from subjective anecdotes to objective, quantified data. Next we model the phenomenon, using appropriate methods in our field, modifying them to suit the unique aspects of the space environment. Finally, we develop technologies, tools, and procedures to mitigate the decrements in human performance and capabilities that occur in space environments. When successful, we decrease risks to crew safety and to mission success. When extremely successful (or lucky), we devise generalizable solutions that advance the state of our practice. Our panel is composed of researchers from diverse domains of our project... from different boxes, if you will, of the Human Factors Analysis and Classification System (HFACS).

Kaiser, Mary K.↗
NASA NTRS2007

High Spatial Resolution Thermal Infrared Remote Sensing Data for Analysis of the Atlanta, Georgia, Urban Heat Island Effect and Its Impacts on the Environment

The twenty-first century is the first "urban century" according to the United Nations Development Program. The focus of cities reflects awareness of the growing percentage of the world's population that lives in urban areas. In environmental terms, cities are the original producers of many of the global problems related to waste disposal, air and water pollution, and associated environmental and ecological challenges. Expansion of cities, both in population and areal extent, is a relentless process. In 2000, approximately 3 billion people representing about 40% of the global population, resided in urban areas. Urban population will continue to rise substantially over the next several decades according to UN estimates, and most of this growth will Occur in developing countries. The UN estimates that by 2025, 60% of the world's population will live in urban areas. As a consequence, the number of"megacities" (those cities with populations of 10 million inhabitants or more) will increase by 100 by 2025. Thus, there is a critical need to understand urban areas and what their impacts are on environmental, ecological and hydrologic resources, as well as on the local, regional, and even global climate. One of the more egregious side effects of urbanization is the increase in surface and air temperatures that lead to deterioration in air quality. In the United States, under the more stringent air quality guidelines established by the U.S. Environmental Protection Agency in 1997, nearly 300 counties in 34 states will not meet these new air quality standards for ground level ozone. Mitigation of the urban heat island (UHI) effect is actively being evaluated as a possible way to reduce ground ozone levels in cities and assist states in improving air quality. Foremost in the analysis of how the UHI affects air quality and other environmental factors is the use of remote sensing technology and data to characterize urban land covers in sufficient detail to quantifiably measure the impact of increased urban heating on air quality. The urban landscape impacts surface thermal energy exchanges that determine development of the UHI. This paper will illustrate how we are using high spatial remote sensing data collected over the Atlanta, Georgia metropolitan area in conjunction with other geographic information, to perform a detailed urban land cover classification and to determine the contribution of these land covers to the urban heat island effect. Also, the spatial arrangement of the land covers and the impact on urban heating from these selected patterns of development are evaluated. Additionally, this paper will show how these data are being used as inputs to improve air quality modeling for Atlanta, including potential benefits from UHI mitigation.

Quattrochi, Dale A.↗
DOE OSTI2023

Improving an Acoustic Vehicle Detector Using an Iterative Self-Supervision Procedure

In many non-canonical data science scenarios, obtaining, detecting, attributing, and annotating enough high-quality training data is the primary barrier to developing highly effective models. Moreover, in many problems that are not sufficiently defined or constrained, manually developing a training dataset can often overlook interesting phenomena that should be included. To this end, we have developed and demonstrated an iterative self-supervised learning procedure, whereby models are successfully trained and applied to new data to extract new training examples that are added to the corpus of training data. Successive generations of classifiers are then trained on this augmented corpus. Using low-frequency acoustic data collected by a network of infrasound sensors deployed around the High Flux Isotope Reactor and Radiochemical Engineering Development Center at Oak Ridge National Laboratory, we test the viability of our proposed approach to develop a powerful classifier with the goal of identifying vehicles from continuously streamed data and differentiating these from other sources of noise such as tools, people, airplanes, and wind. Using a small collection of exhaustively manually labeled data, we test several implementation details of the procedure and demonstrate its success regardless of the fidelity of the initial model used to seed the iterative procedure. Finally, we demonstrate the method’s ability to update a model to accommodate changes in the data-generating distribution encountered during long-term persistent data collection.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗