Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “limited memory”

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 721 records · Page 40
All sourcesNASA NTRSDOE OSTI
NASA NTRS1979

Ensuring the relocatability of programs in the operational system DOS YeS

Specific modifications in the Disk Operational System Unified Series to insure the relocatability of programs stored permanently in the core image library is described. A self-relocating method for loading programs into the working memory with re-editing all the programs recorded in the core image library is presented. The modified linkage editor can be included in a relocation dictionary containing data about each address constant at the assembly stage at the request of the programmer. The relocation dictionary increases the dimension of the RL-phase in comparison with the dimension of this same phase when edited by the standard method, making possible the creation of multiphase program complexes. Generation and use of the modified system using Assembly language is described. An example of the use of the system is given, and limitations of the use of the relocatable programs in the modified system are outlined.

Novoseltsev, S. K.↗
NASA NTRS2007

CASPER Version 2.0

CASPER is designed to perform automated planning of interdependent activities within a system subject to requirements, constraints, and limitations on resources. In contradistinction to the traditional concept of batch planning followed by execution, CASPER implements a concept of continuous planning and replanning in response to unanticipated changes (including failures), integrated with execution. Improvements over other, similar software that have been incorporated into CASPER version 2.0 include an enhanced executable interface to facilitate integration with a wide range of execution software systems and supporting software libraries; features to support execution while reasoning about urgency, importance, and impending deadlines; features that enable accommodation to a wide range of computing environments that include various central processing units and random- access-memory capacities; and improved generic time-server and time-control features.

Chien, Steve↗
DOE OSTI2020

Evaluating the Potential and Challenges of an Uncertainty Quantification Method for Long Short–Term Memory Models for Soil Moisture Predictions

Recently, recurrent deep networks have shown promise to harness newly available satellite–sensed data for long–term soil moisture projections. However, to be useful in forecasting, deep networks must also provide uncertainty estimates. Here we evaluated Monte Carlo dropout with an input–dependent data noise term (MCD+N), an efficient uncertainty estimation framework originally developed in computer vision, for hydrologic time series predictions. MCD+N simultaneously estimates a heteroscedastic input–dependent data noise term (a trained error model attributable to observational noise) and a network weight uncertainty term (attributable to insufficiently constrained model parameters). Although MCD+N has appealing features, many heuristic approximations were employed during its derivation, and rigorous evaluations and evidence of its asserted capability to detect dissimilarity were lacking. To address this, we provided an in–depth evaluation of the scheme's potential and limitations. We showed that for reproducing soil moisture dynamics recorded by the Soil Moisture Active Passive (SMAP) mission, MCD+N indeed gave a good estimate of predictive error, provided that we tuned a hyperparameter and used a representative training data set. The input–dependent term responded strongly to observational noise, while the model term clearly acted as a detector for physiographic dissimilarity from the training data, behaving as intended. However, when the training and test data were characteristically different, the input–dependent term could be misled, undermining its reliability. Additionally, due to the data–driven nature of the model, data noise also influences network weight uncertainty, and therefore the two uncertainty terms are correlated. Altogether, this approach has promise, but care is needed to interpret the results.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2007

Development of a HTSMA-Actuated Surge Control Rod for High-Temperature Turbomachinery Applications

In recent years, a demand for compact, lightweight, solid-state actuation systems has emerged, driven in part by the needs of the aeronautics industry. However, most actuation systems used in turbomachinery require not only elevated temperature but high-force capability. As a result, shape memory alloy (SMA) based systems have worked their way to the forefront of a short list of viable options to meet such a technological challenge. Most of the effort centered on shape memory systems to date has involved binary NiTi alloys but the working temperatures required in many aeronautics applications dictate significantly higher transformation temperatures than the binary systems can provide. Hence, a high temperature shape memory alloy (HTSMA) based on NiTiPdPt, having a transformation temperature near 300 C, was developed. Various thermo-mechanical processing schemes were utilized to further improve the dimensional stability of the alloy and it was later extruded/drawn into wire form to be more compatible with envisioned applications. Mechanical testing on the finished wire form showed reasonable work output capability with excellent dimensional stability. Subsequently, the wire form of the alloy was incorporated into a benchtop system, which was shown to provide the necessary stroke requirements of approx.0.125 inches for the targeted surge-control application. Cycle times for the actuator were limited to ~4 seconds due to control and cooling constraints but this cycle time was determined to be adequate for the surge control application targeted as the primary requirement was initial actuation of a surge control rod, which could be completed in approximately one second.

Padula, Santo, II↗
DOE OSTI2024

tomoCAM : fast model-based iterative reconstruction via GPU acceleration and non-uniform fast Fourier transforms

X-ray-based computed tomography is a well established technique for determining the three-dimensional structure of an object from its two-dimensional projections. In the past few decades, there have been significant advancements in the brightness and detector technology of tomography instruments at synchrotron sources. These advancements have led to the emergence of new observations and discoveries, with improved capabilities such as faster frame rates, larger fields of view, higher resolution and higher dimensionality. These advancements have enabled the material science community to expand the scope of tomographic measurements towards increasingly in situ and in operando measurements. In these new experiments, samples can be rapidly evolving, have complex geometries and restrictions on the field of view, limiting the number of projections that can be collected. In such cases, standard filtered back-projection often results in poor quality reconstructions. Iterative reconstruction algorithms, such as model-based iterative reconstructions (MBIR), have demonstrated considerable success in producing high-quality reconstructions under such restrictions, but typically require high-performance computing resources with hundreds of compute nodes to solve the problem in a reasonable time. Here, tomoCAM , is introduced, a new GPU-accelerated implementation of model-based iterative reconstruction that leverages non-uniform fast Fourier transforms to efficiently compute Radon and back-projection operators and asynchronous memory transfers to maximize the throughput to the GPU memory. The resulting code is significantly faster than traditional MBIR codes and delivers the reconstructive improvement offered by MBIR with affordable computing time and resources. tomoCAM has a Python front-end, allowing access from Jupyter -based frameworks, providing straightforward integration into existing workflows at synchrotron facilities.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
NASA NTRS2024

Evaluation of Machine Learning and Deep Learning Algorithms for Fire Prediction in Southeast Asia

Vegetation fires are prevalent in South/Southeast Asian countries, making fire prediction crucial due to their potential environmental, economic, and social impacts. Accurate predictions of fires facilitate timely interventions, helping to mitigate uncontrolled fires that can lead to biodiversity loss and air quality issues. In this study, we utilize VIIRS satellite-derived fire data alongside six machine learning and deep learning models—Simple Persistence, Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), CNN-LSTM, and ConvLSTM—to determine the most effective fire prediction model, using Root Mean Square Error (RMSE) as the metric. Our results indicate that the CNN model is the most reliable in regions with spatial dependencies, such as Brunei, Indonesia, Malaysia, the Philippines, Timor-Leste, and Thailand. Conversely, the ConvLSTM model excels in countries with complex spatiotemporal dynamics like Laos, Myanmar, and Vietnam. The CNN-LSTM hybrid model also performed well in Cambodia, suggesting a need for a balanced approach in areas requiring both spatial and temporal feature extraction. Furthermore, simpler models like Persistence and MLP showed limitations in capturing dynamic patterns and temporal dependencies. Our findings highlight the importance of evaluating models before implementing any decision support systems (DSS) in fire management. By tailoring models to specific regional fire data, we can enhance prediction accuracy and responsiveness, ultimately improving fire risk management in Southeast Asia and beyond.

Deep learning↗
DOE OSTI2020

Implicit fast sweeping method for hyperbolic systems of conservation laws

Implicit time-accurate methods are often used to integrate stiff problems where explicit schemes impose severe time step restrictions. This paper presents an efficient numerical framework based on the Fast Sweeping Method (FSM) for solving linear and nonlinear hyperbolic systems of conservation laws. The solution at each discrete location is computed by sweeping the numerical domain in several predetermined directions that follow the causality of the characteristic families. The use of a fractional step strategy eliminates the need for a solution selection criterion while one-sided stencils limit the number of sweeps to at most 2 d for d space dimensions. This work focuses on the first-order implicit upwind method since it constitutes the building block for high-order conservative schemes. For problems where the degree of stiffness evolves over time, implicit-explicit hybridization can be accomplished with the same algorithm by simply switching the stencil at each time level. As opposed to traditional implicit solvers, the sweeping method does not require a local time linearization of the fluxes thereby preserving the nonlinear stability properties of the original implicit scheme. It also avoids the large computational and memory requirements associated with solving large block-diagonal systems of equations. Here, a series of one- and two-dimensional test cases are presented for the inviscid Burgers' equation and the reactive Euler equations. The results indicate that the implicit FSM can allow a major reduction in the number of time steps even in the presence of discontinuous solution profiles.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2022

Understanding Strong Scaling on GPUs Using Empirical Performance Saturation Size

The roofline model provides a concise overview of the maximum performance capabilities of a given computer system through a combination of peak memory bandwidth and compute performance rates. The increasing complexity of scheduling and cache in recent GPUs, however, has introduced complicated performance variability that is not captured by arithmetic intensity alone. This work examines the effect of problem size and GPU launch configurations on roofline performance for V100, A100, MI100, and MI250X graphics processing units. We introduce an extended roofline model that takes problem size into account, and find that strong scaling on GPUs can be characterized by saturation problem sizes as additional key metrics. Saturation problem sizes break up a plot of GPU performance vs. problem size into three distinct performance regimes– size-limited, cache-bound, and DRAM-bound. With our extended roofline model, we are able to provide a robust view of these performance regimes across recent GPU architectures.

Eberius, David↗
DOE OSTI2022

ASAP: Automatic Synthesis of Area-Efficient and Precision-Aware CGRAs

Coarse-grained reconfigurable accelerators (CGRAs) are a promising accelerator design choice that strikes a balance between performance and adaptability to different computing patterns across various applications domains. Designing a CGRA for a specific application domain involves enormous software/hardware engineering effort. Recent research works explore loop transformations, functional unit types, network topology, and memory size to identify optimal CGRA designs given a set of kernels from a specific application do- main. Unfortunately, the impact of functional units with different precision support has rarely been investigated. To address this gap, we propose ASAP – a hardware/software co-design framework that automatically identifies and synthesizes optimal precision-aware CGRA for a set of applications of interest. Our evaluation shows that ASAP generates specialized designs 3.2×, 4.21×, and 5.8× more efficient (in terms of performance per unit of energy or area) than non-specialized homogeneous CGRAs, for the scientific computing, embedded, and edge machine learning domains, respectively, with limited accuracy loss. Moreover, ASAP provides more efficient designs than other state-of-the-art synthesis frameworks for specialized CGRAs.

artificial intelligence↗
NASA NTRS2008

Application of Fast Multipole Methods to the NASA Fast Scattering Code

The NASA Fast Scattering Code (FSC) is a versatile noise prediction program designed to conduct aeroacoustic noise reduction studies. The equivalent source method is used to solve an exterior Helmholtz boundary value problem with an impedance type boundary condition. The solution process in FSC v2.0 requires direct manipulation of a large, dense system of linear equations, limiting the applicability of the code to small scales and/or moderate excitation frequencies. Recent advances in the use of Fast Multipole Methods (FMM) for solving scattering problems, coupled with sparse linear algebra techniques, suggest that a substantial reduction in computer resource utilization over conventional solution approaches can be obtained. Implementation of the single level FMM (SLFMM) and a variant of the Conjugate Gradient Method (CGM) into the FSC is discussed in this paper. The culmination of this effort, FSC v3.0, was used to generate solutions for three configurations of interest. Benchmarking against previously obtained simulations indicate that a twenty-fold reduction in computational memory and up to a four-fold reduction in computer time have been achieved on a single processor.

Dunn, Mark H.↗
NASA NTRS1981

Video image processor on the Spacelab 2 Solar Optical Universal Polarimeter /SL2 SOUP/

The SOUP instrument is designed to obtain diffraction-limited digital images of the sun with high photometric accuracy. The Video Processor originated from the requirement to provide onboard real-time image processing, both to reduce the telemetry rate and to provide meaningful video displays of scientific data to the payload crew. This original concept has evolved into a versatile digital processing system with a multitude of other uses in the SOUP program. The central element in the Video Processor design is a 16-bit central processing unit based on 2900 family bipolar bit-slice devices. All arithmetic, logical and I/O operations are under control of microprograms, stored in programmable read-only memory and initiated by commands from the LSI-11. Several functions of the Video Processor are described, including interface to the High Rate Multiplexer downlink, cosmetic and scientific data processing, scan conversion for crew displays, focus and exposure testing, and use as ground support equipment.

Lindgren, R. W.↗
NASA NTRS1998

Inferences About the Early Moon From Gravity and Topography

Recent spacecraft missions to the Moon have significantly improved our knowledge of the lunar gravity and topography fields, and have raised some new and old questions about the early lunar history. It has frequently been assumed that the shape of the Moon today reflects an earlier equilibrium state and that the Moon has retained some internal strength. Recent analysis indicating a superisostatic state of some lunar basins lends support to this hypothesis. On its simplest level, the present shape of the Moon is slightly flattened by 2.2 +/- 0.2 km while its gravity field, represented by an equipotential surface, is flattened only about 0.5 km. The hydrostatic component to the flattening arising from the Moon's present day rotation contributes only 7 m. This difference between the topographic shape of the MOon and the shape of its gravitational equipotential has frequently been explained as the "memory" of an earlier moon that was rotating faster and had a correspondingly larger hydrostatic flattening. To obtain this amount of hydrostatic flattening from rotation alone, and accounting for the contribution of the present-day gravity field, the Moon's rotation rate would need to be about 15x greater than at present, leading ot a period of < 2 days. Maintaining its synchronous rotation with Earth would require a radius for the Moon's orbit of approximately 9 Earth Radii. Unfortunately, our confidence in the observed lunar flattening is not as great as we would like. The uncertainty of .02 km may not properly reflect the limitations of the Clementine dataset, which did not sample poleward of latitudes 81 N and 79 S. Also, the large variation of topography +/- 8 km seen on the MOon dwarfs our estimate fo the flattening. Further the lunar south pole is on the edge of, or possibly inside the massive deep, South Pole-Aitken Basin. Thus, polar radii could be underestimated. This would yield a smaller flattening, which would imply a greater lunar rotation period and orbital radius. However, Basin compensation states and analyses of support and relaxation of topography at long wavelengths point to a lunar shape that has retained a flattening from an earlier faster rotation period.

Smith, D. E.↗
DOE OSTI2025

Physics-informed latent neural operator for real-time predictions of time-dependent parametric PDEs

Deep operator network (DeepONet) has shown significant promise as surrogate models for systems governed by partial differential equations (PDEs), enabling accurate mappings between infinite-dimensional function spaces. However, when applied to systems with high-dimensional input-output mappings arising from large numbers of spatial and temporal collocation points, these models often require heavily overparameterized networks, leading to long training times. Latent DeepONet addresses some of these challenges by introducing a two-step approach: first learning a reduced latent space using a separate model, followed by operator learning within this latent space. While efficient, this method is inherently data-driven and lacks mechanisms for incorporating physical laws, limiting its robustness and generalizability in data-scarce settings. Here, in this work, we propose PI-Latent-NO, a physics-informed latent neural operator framework that integrates governing physics directly into the learning process. Our architecture features two coupled DeepONets trained end-to-end: a Latent-DeepONet that learns a low-dimensional representation of the solution, and a Reconstruction-DeepONet that maps this latent representation back to the physical space. By embedding PDE constraints into the training via automatic differentiation, our method eliminates the need for labeled training data and ensures physics-consistent predictions. The proposed framework is both memory and compute-efficient, exhibiting near-constant scaling with problem size and demonstrating significant speedups over traditional physics-informed operator models. We validate our approach on a range of parametric PDEs, showcasing its accuracy, scalability, and suitability for real-time prediction in complex physical systems.

Latent representations↗
DOE OSTI2021

Proteo-Genomic Analysis Identifies Two Major Sites of Vulnerability on Ebolavirus Glycoprotein for Neutralizing Antibodies in Convalescent Human Plasma

Three clinically relevant ebolaviruses – Ebola (EBOV), Bundibugyo (BDBV), and Sudan (SUDV) viruses, are responsible for severe disease and occasional deadly outbreaks in Africa. The largest Ebola virus disease (EVD) epidemic to date in 2013-2016 in West Africa highlighted the urgent need for countermeasures, leading to the development and FDA approval of the Ebola virus vaccine rVSV-ZEBOV (Ervebo ® ) in 2020 and two monoclonal antibody (mAb)-based therapeutics (Inmazeb ® [atoltivimab, maftivimab, and odesivimab-ebgn] and Ebanga ® (ansuvimab-zykl) in 2020. The humoral response plays an indispensable role in ebolavirus immunity, based on studies of mAbs isolated from the antibody genes in peripheral blood circulating ebolavirus-specific human memory B cells. However, antibodies in the body are not secreted by circulating memory B cells in the blood but rather principally by plasma cells in the bone marrow. Little is known about the protective polyclonal antibody responses in convalescent plasma. Here we exploited both single-cell antibody gene sequencing and proteomic sequencing approaches to assess the composition of the ebolavirus glycoprotein (GP)-reactive antibody repertoire in the plasma of an EVD survivor. We first identified 1,512 GP-specific mAb variable gene sequences from single cells in the memory B cell compartment. Using mass spectrometric analysis of the corresponding GP-specific plasma IgG, we found that only a portion of the large B cell antibody repertoire was represented in the plasma. Molecular and functional analysis of proteomics-identified mAbs revealed recognition of epitopes in three major antigenic sites - the GP head domain, the glycan cap, and the base region, with a high prevalence of neutralizing and protective mAb specificities that targeted the base and glycan cap regions on the GP. Polyclonal plasma antibodies from the survivor reacted broadly to EBOV, BDBV, and SUDV GP, while reactivity of the potently neutralizing mAbs we identified was limited mostly to the homologous EBOV GP. Together these results reveal a restricted diversity of neutralizing humoral response in which mAbs targeting two antigenic sites on GP – glycan cap and base – play a principal role in plasma-antibody-mediated protective immunity against EVD.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2003

The Effect of Gravity on the Combustion Synthesis of Porous Biomaterials

Production of highly porous composite materials by traditional materials processing is limited by difficult processing techniques. This work investigates the use of self propagating high temperature (combustion) synthesis (SHS) to create porous tricalcium phosphate (Ca3(PO4)2), TiB-Ti, and NiTi in low and microgravity. Combustion synthesis provides the ability to use set processing parameters to engineer the required porous structure suitable for bone repair or replacement. The processing parameters include green density, particle size, gasifying agents, composition, and gravity. The advantage of the TiB-Ti system is the high level of porosity achieved together with a modulus that can be controlled by both composition (TiB-Ti) and porosity. At the same time, NiTi exhibits shape memory properties. SHS of biomaterials allows the engineering of required porosity coupled with resorbtion properties and specific mechanical properties into the composite materials to allow for a better biomaterial.

Castillo, M.↗
NASA NTRS2011

Thermal Properties of Microstrain Gauges Used for Protection of Lithium-Ion Cells of Different Designs

The purpose of this innovation is to use microstrain gauges to monitor minute changes in temperature along with material properties of the metal cans and pouches used in the construction of lithium-ion cells. The sensitivity of the microstrain gauges to extremely small changes in temperatures internal to the cells makes them a valuable asset in controlling the hazards in lithium-ion cells. The test program on lithium-ion cells included various cell configurations, including the pouch type configurations. The thermal properties of microstrain gauges have been found to contribute significantly as safety monitors in lithium-ion cells that are designed even with hard metal cases. Although the metal cans do not undergo changes in material property, even under worst-case unsafe conditions, the small changes in thermal properties observed during charge and discharge of the cell provide an observable change in resistance of the strain gauge. Under abusive or unsafe conditions, the change in the resistance is large. This large change is observed as a significant change in slope, and this can be used to prevent cells from going into a thermal runaway condition. For flexible metal cans or pouch-type lithium-ion cells, combinations of changes in material properties along with thermal changes can be used as an indication for the initiation of an unsafe condition. Lithium-ion cells have a very high energy density, no memory effect, and almost 100-percent efficiency of charge and discharge. However, due to the presence of a flammable electrolyte, along with the very high energy density and the capability of releasing oxygen from the cathode, these cells can go into a hazardous condition of venting, fire, and thermal runaway. Commercial lithium-ion cells have current and voltage monitoring devices that are used to control the charge and discharge of the batteries. Some lithium-ion cells have internal protective devices, but when used in multi-cell configurations, these protective devices either do not protect or are themselves a hazard to the cell due to their limitations. These devices do not help in cases where the cells develop high impedance that suddenly causes them to go into a thermal runaway condition. Temperature monitoring typically helps with tracking the performance of a battery. But normal thermistors or thermal sensors do not provide the accuracy needed for this and cannot track a change in internal cell temperatures until it is too late to stop a thermal runaway.

Jeevarajan, Judith↗
NASA NTRS2006

Pattern-Recognition Processor Using Holographic Photopolymer

proposed joint-transform optical correlator (JTOC) would be capable of operating as a real-time pattern-recognition processor. The key correlation-filter reading/writing medium of this JTOC would be an updateable holographic photopolymer. The high-resolution, high-speed characteristics of this photopolymer would enable pattern-recognition processing to occur at a speed three orders of magnitude greater than that of state-of-the-art digital pattern-recognition processors. There are many potential applications in biometric personal identification (e.g., using images of fingerprints and faces) and nondestructive industrial inspection. In order to appreciate the advantages of the proposed JTOC, it is necessary to understand the principle of operation of a conventional JTOC. In a conventional JTOC (shown in the upper part of the figure), a collimated laser beam passes through two side-by-side spatial light modulators (SLMs). One SLM displays a real-time input image to be recognized. The other SLM displays a reference image from a digital memory. A Fourier-transform lens is placed at its focal distance from the SLM plane, and a charge-coupled device (CCD) image detector is placed at the back focal plane of the lens for use as a square-law recorder. Processing takes place in two stages. In the first stage, the CCD records the interference pattern between the Fourier transforms of the input and reference images, and the pattern is then digitized and saved in a buffer memory. In the second stage, the reference SLM is turned off and the interference pattern is fed back to the input SLM. The interference pattern thus becomes Fourier-transformed, yielding at the CCD an image representing the joint-transform correlation between the input and reference images. This image contains a sharp correlation peak when the input and reference images are matched. The drawbacks of a conventional JTOC are the following: The CCD has low spatial resolution and is not an ideal square-law detector for the purpose of holographic recording of interference fringes. A typical state-of-the-art CCD has a pixel-pitch limited resolution of about 100 lines/mm. In contrast, the holographic photopolymer to be used in the proposed JTOC offers a resolution > 2,000 lines/mm. In addition to being disadvantageous in itself, the low resolution of the CCD causes overlap of a DC term and the desired correlation term in the output image. This overlap severely limits the correlation signal-to-noise ratio. The two-stage nature of the process limits the achievable throughput rate. A further limit is imposed by the low frame rate (typical video rates) of low- and medium-cost commercial CCDs.

Chao, Tien-Hsin↗
NASA NTRS2024

Universal Coating by 3D Hybrid Programmable Matter

Motivated by the prospect of nano-robots that assist human physiological functions at the nanoscale, we investigate the coating problem in the three-dimensional model for hybrid programmable matter. In this model, a single agent with strictly limited viewing range and the computational capability of a deterministic finite automaton can act on passive tiles by picking up a tile, moving, and placing it at some spot. The goal of the coating problem is to fill each node of some surface graph of size n with a tile. We first solve the problem on a restricted class of graphs with a single tile type, and then use constantly many tile types to encode this graph in certain surface graphs capturing the surface of 3D objects. Our algorithm requires O(n^2) steps, which is worst-case optimal compared to an agent with global knowledge and no memory restrictions.

programmable matter↗