Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Learning Theory”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,639 records · Page 91

Sparse distributed memory overview

The Sparse Distributed Memory (SDM) project is investigating the theory and applications of massively parallel computing architecture, called sparse distributed memory, that will support the storage and retrieval of sensory and motor patterns characteristic of autonomous systems. The immediate objectives of the project are centered in studies of the memory itself and in the use of the memory to solve problems in speech, vision, and robotics. Investigation of methods for encoding sensory data is an important part of the research. Examples of NASA missions that may benefit from this work are Space Station, planetary rovers, and solar exploration. Sparse distributed memory offers promising technology for systems that must learn through experience and be capable of adapting to new circumstances, and for operating any large complex system requiring automatic monitoring and control. Sparse distributed memory is a massively parallel architecture motivated by efforts to understand how the human brain works. Sparse distributed memory is an associative memory, able to retrieve information from cues that only partially match patterns stored in the memory. It is able to store long temporal sequences derived from the behavior of a complex system, such as progressive records of the system's sensory data and correlated records of the system's motor controls.

Raugh, Mike↗

Cauer Prototype for Butterworth Filters: Lowpass, Highpass, Bandpass, and Bandstop Transformations

I had a need to develop a real, physical, analog filter with real RLC components. Having spent a career in theory, I had only ever designed filters for signal processing. I needed only the a and b coefficients. But converting a and b coefficients into RLC values was beyond me. Web searches returned nothing useful until I found Dr. Aaron Danner’s Designing Passive Butterworth Filters (10 - Passive Filters), part of an excellent YouTube series which introduced me to the Cauer prototype for Butterworth filters and Cauer topologies. Failing to find documentation on Cauer filter prototypes, I asked Anthropic’s Claude AI for assistance. Claude returned a few errors which I corrected. This is a comprehensive report on what I learned.

42 ENGINEERING↗

Grid Topology Discovery Algorithm Evaluation of Suitability for Utility Deployment (CRADA 606 Final Report)

This work presents the results of a field-informed demonstration aimed at evaluating the practical suitability of a topology discovery algorithm for utility environments. We demonstrated an algorithm that uses a graph-theory-informed state estimation approach for model selection. In collaboration with Survalent and Peninsula Light Co., the algorithm was applied to real feeder models and field measurements from supervisory control and data acquisition (SCADA) and advanced metering infrastructure (AMI) systems to identify the operational topology of a power distribution system. The demonstration assessed the algorithm’s performance under realistic data conditions, including sparse and noisy measurements, and examined its ability to identify the most likely network configurations. The results confirmed that the approach can effectively narrow down feasible topologies, providing operators with improved situational awareness of network status. Key lessons learned emphasize the need for systematic data validation and strategic sensor placement to enhance observability. These insights inform future deployment strategies and guide refinements for broader adoption in utility operations.

24 POWER TRANSMISSION AND DISTRIBUTION↗

AI and simulation: What can they learn from each other

Simulation and Artificial Intelligence share a fertile common ground both from a practical and from a conceptual point of view. Strengths and weaknesses of both Knowledge Based System and Modeling and Simulation are examined and three types of systems that combine the strengths of both technologies are discussed. These types of systems are a practical starting point, however, the real strengths of both technologies will be exploited only when they are combined in a common knowledge representation paradigm. From an even deeper conceptual point of view, one might even argue that the ability to reason from a set of facts (i.e., Expert System) is less representative of human reasoning than the ability to make a model of the world, change it as required, and derive conclusions about the expected behavior of world entities. This is a fundamental problem in AI, and Modeling Theory can contribute to its solution. The application of Knowledge Engineering technology to a Distributed Processing Network Simulator (DPNS) is discussed.

Colombano, Silvano P.↗

Cluster expansion by transfer learning for phase stability predictions

Recent progress towards universal machine-learned interatomic potentials holds considerable promise for materials discovery. Yet the accuracy of these potentials for predicting phase stability may still be limited. In contrast, cluster expansions provide accurate phase stability predictions but are computationally demanding to parameterize from first principles, especially for structures of low dimension or with a large number of components, such as interfaces or multimetal catalysts. We overcome this trade-off via transfer learning. Using Bayesian inference, we incorporate prior statistical knowledge from machine-learned and physics-based potentials, enabling us to sample the most informative configurations and to efficiently fit first-principles cluster expansions. Furthermore, this algorithm is tested on Pt:Ni, showing robust convergence of the mixing energies as a function of sample size with reduced statistical fluctuations.

36 MATERIALS SCIENCE↗

Learning heterogeneous reaction kinetics from X-ray videos pixel by pixel

Reaction rates at spatially heterogeneous, unstable interfaces are notoriously difficult to quantify, yet are essential in engineering many chemical systems, such as batteries and electrocatalysts. Experimental characterizations of such materials by operando microscopy produce rich image datasets, but data-driven methods to learn physics from these images are still lacking because of the complex coupling of reaction kinetics, surface chemistry and phase separation. Here we show that heterogeneous reaction kinetics can be learned from in situ scanning transmission X-ray microscopy (STXM) images of carbon-coated lithium iron phosphate (LFP) nanoparticles. Combining a large dataset of STXM images with a thermodynamically consistent electrochemical phase-field model, partial differential equation (PDE)-constrained optimization and uncertainty quantification, we extract the free-energy landscape and reaction kinetics and verify their consistency with theoretical models. We also simultaneously learn the spatial heterogeneity of the reaction rate, which closely matches the carbon-coating thickness profiles obtained through Auger electron microscopy (AEM). Across 180,000 image pixels, the mean discrepancy with the learned model is remarkably small (<7%) and comparable with experimental noise. Our results open the possibility of learning nonequilibrium material properties beyond the reach of traditional experimental methods and offer a new non-destructive technique for characterizing and optimizing heterogeneous reactive surfaces.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Integrated design of aluminum-enriched high-entropy refractory B2 alloys with synergy of high strength and ductility

Refractory high-entropy alloys (RHEAs) are promising high-temperature structural materials. Their large compositional space poses great design challenges for phase control and high strength-ductility synergy. The present research pioneers using integrated high-throughput machine learning with Monte Carlo simulations supplemented by ab initio calculations to effectively navigate phase selection and mechanical property predictions, developing single-phase ordered B2 aluminum-enriched RHEAs (Al-RHEAs) demonstrating high strength and ductility. These Al-RHEAs achieve remarkable mechanical properties, including compressive yield strengths up to 1.7 gigapascals, fracture strains exceeding 50%, and notable high-temperature strength retention. They also demonstrate a tensile yield strength of 1.0 gigapascals with a ductility of 9%, albeit with B2 ordering. Furthermore, we identify valence electron count domains for alloy ductility and brittleness with the explanation from density functional theory and provide crucial insights into elemental influence on atomic ordering and mechanical performance. The work sets forth a strategic blueprint for high-throughput alloy design and reveals fundamental principles governing the mechanical properties of advanced structural alloys.

Science & Technology - Other Topics↗

Artificial intelligence “sees” split electrons

Chemical bonds between atoms are stabilized by the exchange-correlation (xc) energy, a quantum-mechanical effect in which “social distancing” by electrons lowers their electrostatic repulsion energy. Kohn-Sham density functional theory (DFT) states that the electron density determines this xc energy, but the density functional must be approximated. Furthermore, this is usually done by satisfying exact constraints of the exact functional (making the approximation predictive), by fitting to data (making it interpolative), or both. Two exact constraints—the ensemble-based piecewise linear variation of the total energy with respect to fractional electron number and fractional electron z-component of spin —require hard-to-control nonlocality. On page 1385 of this issue, Kirkpatrick et al. have taken a big step toward more accurate predictions for chemistry through the machine learning of molecular data plus the fractional charge and spin constraints, expressed as data that a machine can learn.

74 ATOMIC AND MOLECULAR PHYSICS↗

Triangle Method for Dense ReLU Layers [SWR-25-72]

This software is an implementation of the methods for initializing and training neural networks to be more efficient per parameter, described more fully below and in the related publication: In theory, depth should make a ReLU network EXPONENTIALLY more efficient by enabling it to produce an exponential number of piecewise linear sections in its output. This reasoning is largely based on the work of mathematicians that have hand-constructed networks that make good use of depth. In practice however, even very deep ReLU networks that have been randomly initialized will behave identically to their shallow counterparts - missing an entire exponential dimension of efficiency. The triangle method is a first attempt at realizing the exponential potential of deep networks. Instead of randomly setting weights, we force pairs of neurons in each layer learn to build triangles (i.e. functions from [0,1] -> [0,1] that look like triangles). This is a very efficient pattern for generating lots of linear pieces because composing two triangular functions doubles the number of pieces with each composition. The triangle method is more than just a different initialization, it is a new paradigm of training. Instead of making direct updates to the matrix weights, we do an extra step of backpropagation to collect the derivatives of the loss function with respect to the shapes of the triangles, training them to tilt left or right. This process essentially holds the networks hand throughout the loss landscape and forces it to always use depth effectively by producing triangular shapes internally. This can produce several orders of magnitude of improvement on convex one-dimensional regression problems. Much more theoretical work is needed to realize its full potential beyond this context, but the implementation in this repository will still work in arbitrary numbers of dimensions. The file Triangle_Method.py is a generalized form of the method that will build each neuron its own custom 1-d convex activation function (with exponential efficiency). Example usage on one dimensional problems can be found in Example_Usage.ipynb and an example of using this in a real neural network can be found in Example_VGG16_CIFAR10.ipynb.

Milkert, Max [National Renewable Energy Laboratory↗

All in the Family: What Brown Dwarfs Teach Us About Extrasolar Giant Planets

As we await the first direct image of an extrasolar giant planet, we can turn to theory and the experience gained in the campaign to detect and understand brown dwarfs for guidance on what to expect. As with any new arrival to a family, there should be a strong family resemblance (one hopes) along with notable unique features and interesting peculiarities. The 300 or so known L and T dwarfs, combined with our own giant planets, already span much of the effective temperature range within which extrasolar planets will be found. Only objects with thick, easily detectable, water clouds have yet to be seen. Thus we already know much of the family. I will describe what we have learned from studying these objects, focusing on the important roles clouds and atmospheric chemistry play in affecting their atmospheres and emergent spectra. Relying on these findings and theoretical models, I'll sketch out what we can expect from extrasolar giant planets, focusing on easily detectable features. Some wild cards, of course, are to be expected. Photochemical hazes, in particular, may obscure the family traits on the faces of Jupiter's distant cousins and may make one wonder, at least momentarily, about the milkman.

Marley, M.↗

Spin-polarized high-energy scattering of charged leptons on nucleons

The proton is a composite object with spin one-half, understood to contain highly relativistic spin one-half quarks exchanging spin-one gluons, each possibly with significant orbital angular momenta. While their fundamental interactions are well described by Quantum ChromoDynamics (QCD), our standard theory of the strong interaction, nonperturbative calculations of the internal structure of the proton based directly on QCD are beginning to provide reliable results. Most of our present knowledge of the structure of the proton is based on experimental measurements interpreted within the rich framework of QCD. An area presently attracting intense interest, both experimental and theoretical, is the relationship between the spin of the proton and the spins and orbital angular momenta of its constituents. While remarkable progress has been made, especially in the last decade, the discovery and investigation of new concepts have revealed that much more remains to be learned. This progress i

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

History of HERMES III Diode to Z-Pinch Breakthrough and Beyond (learning about Pulsed Power and Z-Pinch ICF)

HERMES III and Z are two flagship accelerators of Sandia s pulsed-power program developed to generate intense ɤ-ray fields for the study of nuclear radiation effects, and to explore high energy-density physics (including the production of intense x-ray fields for Inertia Confinement Fusion [ICF]), respectively. A diode at the exit of HERMES III converts its 20-MeV electron beam into ɤ-rays. In contrast, at the center of Z, a z-pinch is used to convert its 20-MA current into an intense burst of x-rays. Here the history of how the HERMES III diode emerged from theoretical considerations to actual hardware is discussed. Next, the reverse process of how the experimental discovery of wire-array stabilization in a z-pinch, led to a better theory of wire-array implosions and its application to one of the ICF concepts on Z--the DH (Dynamic Hohlraum) is reviewed. Lastly, the report concludes with how the unexpected axial radiation asymmetry measured in the DH is understood. The first discussion illustrates the evolution of physics from theory-to-observation-to-refinement. The second two illustrate the reverse process of observation-to-theory-to-refinement. The histories are discussed through the vehicle of my research at Sandia, illustrating the unique environment Sandia provides for personal growth and development into a scientific leader.

Sanford, Thomas William Lou↗

MDO and Cross-Disciplinary Practice in R&D: A Portrait of Principles and Current Practice

For several decades, Multidisciplinary Design Optimization (MDO) has served an important role in aerospace engineering by incorporating physics based disciplinary models into integrated system or sub-system models for use in research, development, (R&D) and design. This paper examines MDO's role in facilitating the integration of the researchers from different single disciplines during R&D and early design of large-scale complex engineered systems (LaCES) such as aerospace systems. The findings in this paper are summarized from a larger study on interdisciplinary practices and perspectives that included considerable empirical data from surveys, interviews, and ethnography. The synthesized findings were derived by integrating the data with theories from organization science and engineering. The over-arching finding is that issues related to cognition, organization, and social interrelations mostly dominate interactions across disciplines. Engineering issues, such as the integration of hardware or physics-based models, are not as significant. Correspondingly, the data showed that MDO is not the primary integrator of researchers working across disciplines during R&D and early design of LaCES. Cognitive focus such as analysis versus design, organizational challenges such as incentives, and social opportunities such as personal networks often drove the human interactive practices among researchers from different disciplines. Facilitation of the inherent confusion, argument, and learning in crossdisciplinary research was identified as one of several needed elements of enabling successful research across disciplines.

Rivas McGowan, Anna-Maria↗

Organics and Ices in the Outer Solar System: Connections to the Interstellar Medium

The solar nebula, that aggregate of gas and dust that formed the birthplace of the Sun, planets and plethora of small bodies comprising the Solar System, originated in a molecular cloud that is thought to have spawned numerous additional stars, some with their own planets and attendant small bodies. The question of the chemical and physical reprocessing of the original interstellar materials in the solar nebula has challenged both theory and observations. The acquisition and analysis of samples of comet and asteroid solids, and a growing suite of in-situ and close-up analyses of relatively unaltered small Solar System bodies now adds critical new dimensions to the study of the origin and evolution of the early solar nebula. Better understanding the original composition of the material from which our solar nebula formed, and the processing that material experienced, will aid in formulations of chemistry that might occur in other solar systems. While we seek to understand the compositional history of planetary bodies in our own Solar System, we will inevitably learn more about the materials that comprise exoplanets and their surrounding systems.

SOLAR NEBULA↗

What do we learn from the CMB observations?

We give an account, at nonexpert and quantitative level, of physics behind the CMB temperature anisotropy and polarization and their peculiar features. We discuss, in particular, how cosmological parameters are determined from the CMB measurements and their combinations with other observations. We emphasize that CMB is the major source of information on the primordial density perturbations and, possibly, gravitational waves, and discuss the implication for our understanding of the extremely early Universe.

79 ASTRONOMY AND ASTROPHYSICS↗

On the geometric dependence of the molecular dipole polarizability in water: A benchmark study of higher-order electron correlation, basis set incompleteness error, core electron effects, and zero-point vibrational contributions

In this work, we investigate how geometric changes influence the static dipole polarizability ($α$) of a water molecule by explicitly computing the corresponding dipole polarizability surface (DPS) across 3125 total (1625 symmetry-unique) geometries using linear response coupled cluster theory including single, double, and triple excitations (LR-CCSDT) and the doubly augmented triple-$ζ$ basis set (d-aug-cc-pVTZ). Analytical formulae based on power series expansions of this $ab initio$ surface are generated using linear least-squares analysis and provide highly accurate estimates of this quantity as a function of molecular geometry (i.e., bond and angle variations) in a computationally tractable manner. An additional database, which consists of 25 representative molecular geometries and incorporates a more thorough treatment of both basis sets and core electron effects, is provided as a current benchmark for this quantity and the corresponding leading-order $C_6$ dispersion coefficient. This database has been utilized to assess the importance of these effects as well as the relative accuracy that can be obtained using several quantum chemical methods and a library of density functional approximations. In addition to high-level electron correlation methods (like CCSD) and our analytical least-squares formulae, we find that the SCAN0, PBE0, MN15, and B97-2 hybrid functionals yield the most accurate descriptions of the molecular polarizability tensor in H 2 O. Using first-order perturbation theory, we compute the zero-point vibrational correction to $α$ at the CCSDT/d-aug-cc-pVTZ level and find that this correction contributes approximately 3% to the isotropic ($α$ iso ) and nearly 50% to the anisotropic ($α$ aniso ) polarizability values. In doing so, we find that $α$ iso = 9.8307 bohr 3 , which is in excellent agreement with the experimental value of 9.83 ± 0.02 bohr 3 provided by Russell and Spackman. The DPS reported herein provides a benchmark-quality quantum mechanical estimate of this fundamental quantity of interest and should find extensive use in the development (and assessment) of next-generation force fields and machine-learning based approaches for modeling water in complex condensed-phase environments.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

ATLAS LTCS Vertically Challenged System Lessons Learned

Re-planning of LTCS TVAC testing and supporting RTA (Receiver Telescope Assembly) Test Plan and Procedure document preparation. The Laser Thermal Control System (LTCS) is designed to maintain the lasers onboard Advanced Topographic Laser Altimeter System (ATLAS) at their operational temperatures. In order to verify the functionality of the LTCS, a thermal balance test of the thermal hardware was performed. During the first cold start of the LTCS, the Loop Heat Pipe (LHP) was unable to control the laser mass simulators temperature. The control heaters were fully on and the loop temperature remained well below the desired setpoint. Thermal analysis of the loop did not show these results. This unpredicted behavior of the LTCS was brought up to a panel of LHP experts. Based on the testing and a review of all the data, there were multiple diagnostic performed in order to narrow down the cause. The prevailing theory is that gravity is causing oscillating flow within the loop, which artificially increased the control power needs. This resulted in a replan of the LTCS test flow and the addition of a GSE heater to allow vertical operation.

Loop Heat Pipe↗

Peri-Net-Pro: the neural processes with quantified uncertainty for crack patterns

Abstract This paper develops a deep learning tool based on neural processes (NPs) called the Peri-Net-Pro, to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified uncertainties. In particular, image classification and regression studies are conducted by means of convolutional neural networks (CNNs) and NPs. First, the amount and quality of the data are enhanced by using peridynamics to theoretically compensate for the problems of the finite element method (FEM) in generating crack pattern images. Second, case studies are conducted with the prototype microelastic brittle (PMB), linear peridynamic solid (LPS), and viscoelastic solid (VES) models obtained by using the peridynamic theory. The case studies are performed to classify the images by using CNNs and determine the suitability of the PMB, LBS, and VES models. Finally, a regression analysis is performed on the crack pattern images with NPs to predict the crack patterns. The regression analysis results confirm that the variance decreases when the number of epochs increases by using the NPs. The training results gradually improve, and the variance ranges decrease to less than 0.035. The main finding of this study is that the NPs enable accurate predictions, even with missing or insufficient training data. The results demonstrate that if the context points are set to the 10th, 100th, 300th, and 784th, the training information is deliberately omitted for the context points of the 10th, 100th, and 300th, and the predictions are different when the context points are significantly lower. However, the comparison of the results of the 100th and 784th context points shows that the predicted results are similar because of the Gaussian processes in the NPs. Therefore, if the NPs are employed for training, the missing information of the training data can be supplemented to predict the results.

Mathematics↗