Search NASA⌕ Search

SEARCH · Search NASA

Results for “factorization machine”

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Uncovering Structure–Conductivity Relationships in Anion Exchange Membranes (AEMs) Using Interpretable Machine Learning

Anion exchange membranes (AEMs) play a vital role in the performance of water electrolyzers and fuel cells, yet their discovery and optimization remain challenging due to the complexity of structure–property relationships. In this study, we introduce a machine learning framework that leverages conditional graph neural networks (cGNNs) and descriptor-based models and a hybrid graph neural network (HGARE) to predict and interpret ionic conductivity. The descriptor-based pipeline employs principal component analysis (PCA), ablation, and SHAP analysis to identify factors governing anion conductivity, revealing electronic, topological, and compositional descriptors as key contributors. Beyond prediction, dimensionality reduction and clustering are performed by employing t-SNE and KMeans as well as SOM, which reveal distinct membranes clusters, some of which were enriched with high anion conductivity. Among graph-based approaches, the graph convolutional (GCN) achieved strong predictive performance, while the Hybrid Graph Autoencoder-Regressor Ensemble (HGARE) achieved the highest accuracy. Additionally, atom-level saliency maps from GCN provide spatial explanations for conductive behavior, revealing the importance of polarizable and flexible regions. This work contributes to the accelerated and data-driven design of high-performance AEMs.

Naghshnejad, Pegah [Department of Chemical Enginee↗
DOE OSTI2024

On the hardness of learning ground state entanglement of geometrically local Hamiltonians

Characterizing the entanglement structure of ground states of local Hamiltonians is a fundamental problem in quantum information. In this work we study the computational complexity of this problem, given the Hamiltonian as input. Our main result is that to show it is cryptographically hard to determine if the ground state of a geometrically local, polynomially gapped Hamiltonian on qudits (d=O(1)) has near-area law vs near-volume law entanglement. This improves prior work of Bouland et al. (arXiv:2311.12017) showing this for non-geometrically local Hamiltonians. In particular we show this problem is roughly factoring-hard in 1D, and LWE-hard in 2D. Our proof works by constructing a novel form of public-key pseudo-entanglement which is highly space-efficient, and combining this with a modification of Gottesman and Irani's quantum Turing machine to Hamiltonian construction. Our work suggests that the problem of learning so-called "gapless" quantum phases of matter might be intractable.

Computational Complexity (cs.CC)↗
NASA NTRS2009

A Comparison of the Unpressurized Rover and Small Pressurized Rover During a Desert Field Evaluation

To effectively explore the lunar surface, astronauts will need a transportation vehicle which can traverse all types of terrain. Currently, the National Aeronautics and Space Administration s (NASA) is investigating two lunar rover configurations to meet such a requirement. Under the Lunar Electric Rover (LER) project, a comparison study between the unpressurized rover (UPR) and the small pressurized rover (SPR) was conducted at the Black Point Lava Flow in Arizona. The objective of the study was to obtain human-in-the-loop performance data on the vehicles with respect to human-machine interfaces, vehicle impacts on crew productivity, and scientific observations. Four male participants took part in four, one-day field tests using the exact same terrain and scientific sites for an accurate comparison between vehicle configurations. Subjective data was collected using several human factors performance measures. Results indicate either vehicle configuration was generally acceptable for a lunar mission; however, the SPR configuration was preferred over the UPR configuration primarily for the SPR s ability to cause less fatigue and enabling greater crew productivity.

Litaker, Harry↗
NASA NTRS2014

Analysis and Prediction of Weather Impacted Ground Stop Operations

When the air traffic demand is expected to exceed the available airport's capacity for a short period of time, Ground Stop (GS) operations are implemented by Federal Aviation Administration (FAA) Traffic Flow Management (TFM). The GS requires departing aircraft meeting specific criteria to remain on the ground to achieve reduced demands at the constrained destination airport until the end of the GS. This paper provides a high-level overview of the statistical distributions as well as causal factors for the GSs at the major airports in the United States. The GS's character, the weather impact on GSs, GS variations with delays, and the interaction between GSs and Ground Delay Programs (GDPs) at Newark Liberty International Airport (EWR) are investigated. The machine learning methods are used to generate classification models that map the historical airport weather forecast, schedule traffic, and other airport conditions to implemented GS/GDP operations and the models are evaluated using the cross-validations. This modeling approach produced promising results as it yielded an 85% overall classification accuracy to distinguish the implemented GS days from the normal days without GS and GDP operations and a 71% accuracy to differentiate the GS and GDP implemented days from the GDP only days.

Analysis↗
NASA NTRS2007

Designing for Our Future in Space

Over the past several years, the disciplines of architecture and human factors have been increasingly recognized as specialties that have focused upon "human-centered design" in the development of spacecraft and surface habitats. These specialties have been instrumental in the conceptual design of overall spacecraft configurations and layouts, as well as habitability outfitting hardware, such as the galley, hygiene facility, sleep quarters, or the layout of displays and controls. From the human-centered perspective, this approach to design assists in the mitigation of risk when designing for an extreme environment such as space. It takes into account the human s physical and cognitive capabilities and limitations, the human s performance in the context of human space flight, the human s interaction with machines that are both physically and cognitively complex, the activities required of the human to accomplish the goals of missions, and the use of design practices that promote products to enable human activity. It is this latter aspect - the use of design practices that promote products to enable human activity - that is the focus of the approach used by the Rhode Island School of Design (RISD) in collaboration with the Habitability and Human Factors Branch (HHFB) at the NASA Johnson Space Center (JSC). During the past few years, there has been a growing recognition of the value added by utilizing industrial designers to further the conceptual development of space hardware, that when used in conjunction with architecture and human factors, provides a robust solution to the design challenge. The "Design for Extreme Environments" Studio at RISD has taken suggested design topics from the NASA JSC HHFB and asked the students to investigate solutions to these challenges. The topics have demanded that the student pay particular attention to a variety of aspects of the space environment and understand how the human responds to each. The student must then adapt the design to these responses. The studio environment has been one way to introduce these challenges, but providing for an "in-residence" opportunity at JSC has given the students a broader vision and set of experiences. The accompanying presentation highlights the studio as well as in-residence work that has been accomplished.

Connolly, Janis↗
DOE OSTI2024

Improving Neutrino Energy Reconstruction with Machine Learning

Faithful energy reconstruction is foundational for precision neutrino experiments like DUNE, but is hindered by uncertainties in our understanding of neutrino--nucleus interactions. Here, we demonstrate that dense neural networks are very effective in overcoming these uncertainties by estimating inaccessible kinematic variables based on the observable part of the final state. We find improvements in the energy resolution by up to a factor of two compared to conventional reconstruction algorithms, which translates into an improved physics performance equivalent to a 10-30% increase in the exposure.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRS1983

Engineering and fabrication cost considerations for cryogenic wind tunnel models

Design and fabrication cost drivers for cryogenic transonic wind tunnel models are defined. The major cost factors for wind tunnel models are model complexity, tolerances, surface finishes, materials, material validation, and model inspection. The cryogenic temperatures require the use of materials with relatively high fracture toughness but at the same time high strength. Some of these materials are very difficult to machine, requiring extensive machine hours which can add significantly to the manufacturing costs. Some additional engineering costs are incurred to certify the materials through mechanical tests and nondestructive evaluation techniques, which are not normally required with conventional models. When instrumentation such as accelerometers and electronically scanned pressure modules is required, temperature control of these devices needs to be incorporated into the design, which requires added effort. Additional thermal analyses and subsystem tests may be necessary, which also adds to the design costs. The largest driver to the design costs is potentially the additional static and dynamic analyses required to insure structural integrity of the model and support system.

Boykin, R. M., Jr.↗
NASA NTRS1988

Implementation and analysis of a Navier-Stokes algorithm on parallel computers

The results of the implementation of a Navier-Stokes algorithm on three parallel/vector computers are presented. The object of this research is to determine how well, or poorly, a single numerical algorithm would map onto three different architectures. The algorithm is a compact difference scheme for the solution of the incompressible, two-dimensional, time-dependent Navier-Stokes equations. The computers were chosen so as to encompass a variety of architectures. They are the following: the MPP, an SIMD machine with 16K bit serial processors; Flex/32, an MIMD machine with 20 processors; and Cray/2. The implementation of the algorithm is discussed in relation to these architectures and measures of the performance on each machine are given. The basic comparison is among SIMD instruction parallelism on the MPP, MIMD process parallelism on the Flex/32, and vectorization of a serial code on the Cray/2. Simple performance models are used to describe the performance. These models highlight the bottlenecks and limiting factors for this algorithm on these architectures. Finally, conclusions are presented.

Fatoohi, Raad A.↗
NASA NTRS1989

Ductile-regime turning of germanium and silicon

Single-point diamond turning of silicon and germanium was investigated in order to clarify the role of cutting depth in coaxing a ductile chip formation in normally brittle substances. Experiments based on the rapid withdrawal of the tool from the workpiece have shown that microfracture damage is a function of the effective depth of cut (as opposed to the nominal cutting depth). In essence, damage created by the leading edge of the tool is removed several revolutions later by lower sections of the tool edge, where the effective cutting depth is less. It appears that a truly ductile cutting response can be achieved only when the effective cutting depth, or critical chip thickness, is less than about 20 nm. Factors such as tool rake angle are significant in that they will affect the actual value of the critical chip thickness for transition from brittle to ductile response. It is concluded that the critical chip thickness is an excellent parameter for measuring the effects of machining conditions on the ductility of the cut and for designing tool-workpiece geometry in both turning and grinding.

Blake, Peter N.↗
NASA NTRS1994

Engineering design knowledge recycling in near-real-time

It is hypothesized that the capture and reuse of machine readable design records is cost beneficial. This informal engineering notebook design knowledge can be used to model the artifact and the design process. Design rationale is, in part, preserved and available for examination. Redesign cycle time is significantly reduced (Baya et al, 1992). These factors contribute to making it less costly to capture and reuse knowledge than to recreate comparable knowledge (current practice). To test the hypothesis, we have focused on validation of the concept and tools in two 'real design' projects this past year: (1) a short (8 month) turnaround project for NASA life science bioreactor researchers was done by a team of three mechanical engineering graduate students at Stanford University (in a class, ME210abc 'Mechatronic Systems Design and Methodology' taught by one of the authors, Leifer); and (2) a long range (8 to 20 year) international consortium project for NASA's Space Science program (STEP: satellite test of the equivalence principle). Design knowledge capture was supported this year by assigning the use of a Team-Design PowerBook. Design records were cataloged in near-real time. These records were used to qualitatively model the artifact design as it evolved. Dedal, an 'intelligent librarian' developed at NASA-ARC, was used to navigate and retrieve captured knowledge for reuse.

Leifer, Larry↗
DOE OSTI2024

Insights into Prismatic Loop Formation in Irradiated Fe–Cr Alloys from Hypothesis-Driven Active Learning and Causal Analysis

Neutron and electron irradiation experimental studies conducted on body-centered cubic Fe and Fe–Cr alloys have established two prismatic dislocation loop populations, which have Burgers vectors of either a/2$\langle$111$\rangle$ or a$\langle$100$\rangle$. Here, the loop formation depends on factors such as dose (D), dose rate (D rt ), temperature (T), chromium content (Cr%), and other alloying elements. Hence, it is important to understand how irradiation-induced dislocation loops evolve conditional upon the loop characteristics, such as loop density (DD), average loop size d̅, and irradiation parameters (D, D rt , T, and irradiation type), which is still an active area of research. To understand these complex structure–property relationships, machine learning (ML) is employed in a three-step approach. This includes imputing missing data with a k-nearest neighbor, generating functionalized features, and assessing feature importance with random forest classification and regression. Physics-based features are incorporated in a hypothesis-driven active learning scheme to overcome data unavailability challenges. Insights obtained from ML models (i) to categorize dislocation loop types, show the highest correlation with d̅; (ii) Log(DD), obtained through mathematical formulations involving D, Cr%, d̅, and T (e.g., Log(DD) ~ D + exp(-Cr%) + 1/d̅ and log(DD) ~ D + exp(-Cr%) + 1/T). Hypothesis-driven active learning is able to predict Log(DD) in which the experimental date is not known. Causal models verify cause–effect relationships for dislocation loop classification and irradiation factors in FeCr alloys.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1994

A psychologist's view of validating aviation systems

All systems, no matter what they are designed to do, have shortcomings that may make them less productive than was hoped during the initial development. Such shortcomings can arise at any stage of development: from conception to the end of the implementation life cycle. While systems failure and errors of a lesser magnitude can occur as a function of mechanical or software breakdown, the majority of such problems, in aviation are usually laid on the shoulders of the human operator and, to a lesser extent, on human factors. The operator bears the responsibility and blame even though, from a human factors perspective, error may have been designed into the system. Human factors is not a new concept in aviation. The name may be new, but the issues related to operators in the loop date back to the industrial revolution of the nineteenth century and certainly to the aviation build-up for World War I. During this first global confrontation, military services from all sides discovered rather quickly that poor selection and training led to drastically increased personnel losses. While hardware design became an issue later, the early efforts were primarily focused on increased care in pilot selection and on their training. This actually involved early labor-intensive simulation, using such devices as sticks and chairs mounted on rope networks which could be manually moved in response to control input. The use of selection criteria and improved training led to more viable person-machine systems. More pilots survived training and their first ten missions in the air, a rule of thumb arrived at by experience which predicted ultimate survival better than any other. This rule was to hold through World War II. At that time, personnel selection and training became very sophisticated based on previous standards. Also, many psychologists were drafted into Army Air Corps programs which were geared towards refining the human factor. However, despite the talent involved in these programs and the tremendous build-up of aviation during the war, there were still aircraft designs that were man killers (no sexism implied since all combat pilots were men). One classic design error that was identified fifty years ago was the multipointer altimeter, which could easily be misread especially by a pilot under considerable task load. It has led to flying fully operational aircraft into the terrain. The authors of the research which formally identified this problem put 'Human Errors' in quotes to express their dissatisfaction with the traditional approach to accident investigation. It traditionally places the burden of guilt on the operator. Some of these altimeters still exist in older aircraft to this day.

Stein, Earl S.↗
DOE OSTI2025

Incorporating Physical Priors into Weakly Supervised Anomaly Detection

We propose a new machine-learning-based anomaly detection strategy for comparing data with a background-only reference (a form of weak supervision). The sensitivity of previous strategies degrades significantly when the signal is too rare or there are many unhelpful features. Our prior-assisted weak supervision (PAWS) method incorporates information from a class of signal models to significantly enhance the search sensitivity of weakly supervised approaches. As long as the true signal is in the prespecified class, PAWS matches the sensitivity of a dedicated, fully supervised method without specifying the exact parameters ahead of time. On the benchmark LHC Olympics anomaly detection dataset, our mix of semisupervised and weakly supervised learning is able to extend the sensitivity over previous methods by a factor of 10 in cross section. Furthermore, if we add irrelevant (noise) dimensions to the inputs, classical methods degrade by another factor of 10 in cross section while PAWS remains insensitive to noise. This new approach could be applied in a number of scenarios and pushes the frontier of sensitivity between completely model-agnostic approaches and fully model-specific searches.

artificial neural networks↗
NASA NTRS2024

Effect of High-Speed Rotation on High-Temperature Superconducting Coils for High Efficiency Megawatt Motor

The use of a superconducting rotor in the High Efficiency Megawatt Motor (HEMM) comes with a number of challenges. The HEMM’s rotor must be designed so that the superconducting coil can tolerate the forces it will encounter during operation. These forces introduce considerable risk to the design of the HEMM. Because this risk cannot be adequately addressed through refined analysis, a set of experiments was deemed necessary to demonstrate that the HEMM’s superconducting coils can survive the stresses imparted by the centrifugal forces acting on the coil during full-speed operation of the machine. These experiments required spin testing a full-scale HEMM coil on a representative rotor structure at room temperature and sequentially higher rotation speeds and measuring the superconductivity response of the coil in liquid nitrogen (LN2) before and after each spin test. Tests were carried out to the full centripetal loading of a recent HEMM design. According to finite element analysis, this load produced stress components in the coil that exceed their design limit by a factor of 1 to 7.4. Although not yet confirmed, the current HEMM design is expected to produce even smaller stresses in the coil due to a 20 percent reduction in the coil’s mass and an increased volume of structural material. The only degradation of the coil that occurred during this testing resulted from a test rig failure and from the voltage taps, which are not part of the final coil design.

Electrified Aircraft Propulsion↗
DOE OSTI2026

Ab Initio Modeling of Aqueous Methanol Mixtures at DFT-SCAN Level Using Machine Learning Interatomic Potentials

Abstract Methanol–water mixtures find use in many applications, particularly catalytic energy conversion processes. Their importance has motivated numerous computational studies, most of which employed molecular dynamics based on classical force fields. These enable simulations of large systems on long time scales but do not reliably describe reactive dynamics involving bond breaking and bond formation. In contrast, ab initio molecular dynamics (AIMD) based on density functional theory (DFT) is generally more reliable for such applications but has a high computational cost, which discourages systematic studies of alcohol-water mixtures. To remedy this, we trained a machine learning interatomic potential capable of probing the properties of aqueous methanol mixtures at the DFT level using the SCAN functional. Our results show that SCAN qualitatively reproduces multiple key experimental features arising from the amphiphilic nature of methanol, including density, diffusion coefficients, X-ray structure factors, and Kirkwood–Buff integrals. We also find that structural correlations between water molecules are somewhat overestimated, leading to a stronger preferential association than that predicted by experiments. However, increasing the temperature by 30 K mitigates this effect and also recovers the correct mobilities of both methanol and water. These results indicate that SCAN provides an accurate description of methanol–water mixtures, making it a reliable choice for investigating the reactive dynamics in such systems.

Park, Sanghyun J. [Princeton University , , , ,]↗
NASA NTRS1993

Composite material impregnation unit

This memorandum presents an introduction to the NASA multi-purpose prepregging unit which is now installed and fully operational at the Langley Research Center in the Polymeric Materials Branch. A description of the various impregnation methods that are available to the prepregger are presented. Machine operating details and protocol are provided for its various modes of operation. These include, where appropriate, the related equations for predicting the desired prepreg specifications. Also, as the prepregger is modular in its construction, each individual section is described and discussed. Safety concerns are an important factor and a chapter has been included that highlights the major safety features. Initial experiences and observations for fiber impregnation are described. These first observations have given great insight into the areas of future work that need to be addressed. Future memorandums will focus on these individual processes and their related problems.

Wilkinson, S. P.↗
NASA NTRS1994

Whirl measurements on leakage flows in turbomachine models

The beneficial effects claimed for whirl control devices demonstrate that the dynamic behavior of rotors is influenced by the fluid whirl in shaft and balance drum seals. The present paper reports results from two series of experiments, the first on the factors affecting the whirl at the seal inlet, and the second on the variation of whirl velocity along the seal. In both cases the LDA measurement technique required the clearance between the fixed and rotating parts of the models to be substantially greater than occurs in real machines, but the results are indicative nevertheless. Experimental and theoretical results are given for the radial distribution of whirl velocity in the gap between impeller shroud and pump casing. Results of tests with modified stator surfaces are also shown. This work leads naturally into the second series of experiments where some preliminary measurements of velocity distribution in the clearance between a fixed stator and a rotating shaft are reported for a range of inlet whirl conditions.

Addlesee, A. J.↗
NASA NTRS2010

Tuning Parameters in Heuristics by Using Design of Experiments Methods

With the growing complexity of today's large scale problems, it has become more difficult to find optimal solutions by using exact mathematical methods. The need to find near-optimal solutions in an acceptable time frame requires heuristic approaches. In many cases, however, most heuristics have several parameters that need to be "tuned" before they can reach good results. The problem then turns into "finding best parameter setting" for the heuristics to solve the problems efficiently and timely. One-Factor-At-a-Time (OFAT) approach for parameter tuning neglects the interactions between parameters. Design of Experiments (DOE) tools can be instead employed to tune the parameters more effectively. In this paper, we seek the best parameter setting for a Genetic Algorithm (GA) to solve the single machine total weighted tardiness problem in which n jobs must be scheduled on a single machine without preemption, and the objective is to minimize the total weighted tardiness. Benchmark instances for the problem are available in the literature. To fine tune the GA parameters in the most efficient way, we compare multiple DOE models including 2-level (2k ) full factorial design, orthogonal array design, central composite design, D-optimal design and signal-to-noise (SIN) ratios. In each DOE method, a mathematical model is created using regression analysis, and solved to obtain the best parameter setting. After verification runs using the tuned parameter setting, the preliminary results for optimal solutions of multiple instances were found efficiently.

Arin, Arif↗