Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “adaptive sampling”

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 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

Development of the Ultimate One-Dimensional Time to Explosion (UODTX) Apparatus

With recent advancements in experimental capabilities and worldwide improvement of digital data acquisition and control, there is sufficient justification to develop a new version of the One Dimensional Time to eXplosion (ODTX) system to allow for additional diagnostics and improved control to increase sample throughput and increase the adaptability of future experiments. The legacy ODTX system was designed to measure time to explosion (TTX) for a given temperature, which it does well, but it was not equipped to handle the recent experimental advances of Pressure-ODTX (PODTX) and Composition-ODTX (CODTX). PODTX involves fielding a pressure transducer as an additional diagnostic and applying arbitrary heating-profiles to the system anvils. CODTX furthers this concept by including in-situ gas-composition analysis via infrared (IR) absorption techniques with plans to extend CODTX to solid composition detection and in-situ mass spectrometry (MS). The legacy ODTX system was not built with these experiments in mind and performing them involves substantial modifications and downtime to the existing machine. Developing a standalone system that can perform all C/P/ODTX experiments and is adaptable to unconceived experiments would allow for the continued use of ODTX as a work-horse for Small Scale Safety Testing (SSST), but also provide a testbed for new explosive safety research. This system would ultimately be housed in the neighboring ODTX cell and would live at the High Explosive Application Facility (HEAF) until the end of its service life. HEAF, Chemistry, and Engineering would be the primary stakeholders for the system’s design as they all have experiments that they want to perform. Engineering would be responsible for the system development and would work closely with HEAF to ensure that the system is built with high explosive (HE) safety in mind. By the end of development and after product realization, HEAF/Chemistry would be the system owner, operator, and primary user. The proposed system design is highly influenced by the existing ODTX design. The system will have load frame to support the experiment, anvils to confine the explosive, and some means of applying confining pressure. The samples will be able to be heated and cooled through the confining the anvils. The biggest change from the legacy design is the ability to add diagnostics to the experiment as needed: thermocouples (TCs), resistance temperature detectors (RTDs), pressure transducers (PTs), infrared transmittance/absorbance (IR), mass spectrometry (MS), strain gages (SGs), photonic doppler velocimeter (PDV), as well as arbitrary 4-20 mV signals and +/- 10 VDC signals. The current design does not allow for flexibility in that respect and is hardwired with a few diagnostic channels. The onus will fall on designing a load frame that can support system modifications as well as a data acquisition system that can handle the multitude of signal types.

36 MATERIALS SCIENCE↗
NASA NTRS1992

Direct Adaptive Control Of An Industrial Robot

Decentralized direct adaptive control scheme for six-jointed industrial robot eliminates part of overall computational burden imposed by centralized controller and degrades performance of robot by reducing sampling rate. Control and controller-adaptation laws based on observed performance of manipulator: no need to model dynamics of robot. Adaptive controllers cope with uncertainties and variations in robot and payload.

Seraji, Homayoun↗
DOE OSTI2023

Efficient Subset Simulation using Hamiltonian Neural Network enhanced Markov Chain Monte Carlo Methods

The Monte Carlo method delivers an unbiased estimate of the probability of failure. However, the variance of the estimate depends on the number of evaluated samples. This number must be very large for estimations of a low probability of failure. If the evaluation of each sample is computationally expensive, the crude Monte Carlo simulation strategy is impracticable. Therefore, subset simulations are used to reduce the required number of evaluations. Subset simulations require a Markov Chain Monte Carlo sampler, such as the random walk Metropolis-Hastings algorithm. The algorithm, however, struggles with sampling in low-probability regions, especially if they are narrow. As a consequence, advanced Markov Chain Monte Carlo simulations have been developed. In particular, the Hamiltonian Monte Carlo method explores the target distribution rapidly. Driven by the idea of Hamiltonian dynamics, this sampler provides a non-random walk through the target distribution. The incorporation of subset simulation and Hamiltonian Monte Carlo methods has shown promising results for reliability analysis. One downside of the Hamiltonian Monte Carlo method is that gradient evaluations are computationally expensive, especially when dealing with high-dimensional problems and evaluating long trajectories. We show that integrating Hamiltonian neural networks in Hamiltonian Monte Carlo simulations significantly speeds up the sampling task. Furthermore, the enhancement of adaptive trajectory length within the Hamiltonian Monte Carlo results in the efficient proposal of the following states. Based on this recent enhancement, we provide a fast sampling strategy for subset simulations using Hamiltonian neural networks to replace the evaluation of the gradient and significantly speed up the Hamiltonian Monte Carlo simulation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Point Adversarial Self-Mining: A Simple Method for Facial Expression Recognition

In this article, we propose a simple yet effective approach, called point adversarial self mining (PASM), to improve the recognition accuracy in facial expression recognition (FER). Unlike previous works focusing on designing specific architectures or loss functions to solve this problem, PASM boosts the network capability by simulating human learning processes: providing updated learning materials and guidance from more capable teachers. Specifically, to generate new learning materials, PASM leverages a point adversarial attack method and a trained teacher network to locate the most informative position related to the target task, generating harder learning samples to refine the network. The searched position is highly adaptive since it considers both the statistical information of each sample and the teacher network capability. Other than being provided new learning materials, the student network also receives guidance from the teacher network. After the student network finishes training, the student network changes its role and acts as a teacher, generating new learning materials and providing stronger guidance to train a better student network. Here, the adaptive learning materials generation and teacher/student update can be conducted more than one time, improving the network capability iteratively. Extensive experimental results validate the efficacy of our method over the existing state of the arts for FER.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Whole-Genome Sequence of Brevibacillus borstelensis SDM, Isolated from a Sorghum-Adapted Microbial Community

The isolation of novel microbes from environmental samples continues to be a key strategy for the discovery of new metabolic capacities for the degradation and transformation of lignocellulose. We report the draft genome sequence of a new strain of Brevibacillus borstelensis isolated from a sorghum-adapted microbial community derived from a compost sample.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1987

SMI adaptive antenna arrays for weak interfering signals

The performance of adaptive antenna arrays is studied when a sample matrix inversion (SMI) algorithm is used to control array weights. It is shown that conventional SMI adaptive antennas, like other adaptive antennas, are unable to suppress weak interfering signals (below thermal noise) encountered in broadcasting satellite communication systems. To overcome this problem, the SMI algorithm is modified. In the modified algorithm, the covariance matrix is modified such that the effect of thermal noise on the weights of the adaptive array is reduced. Thus, the weights are dictated by relatively weak coherent signals. It is shown that the modified algorithm provides the desired interference protection. The use of defocused feeds as auxiliary elements of an SMI adaptive array is also discussed.

Gupta, I. J.↗
DOE OSTI2023

Correlated Trajectory Uncertainty for Adaptive Sequential Decision Making

One of the great challenges with decision making tasks on real world systems is the fact that data is sparse and acquiring additional data is expensive. In these cases, it is often crucial to make a model of the environment to assist in making decisions. At the same time, limited data means that learned models are erroneous, making it just as important to equip the model with good predictive uncertainties. In the context of learning sequential decision making policies, these uncertainties can prove useful for informing which data to collect for the greatest improvement in policy performance \citep{mehta2021experimental, mehta2022exploration} or informing the policy about unsure regions of state and action space to avoid during test time \citep{yu2020mopo}. Additionally, assuming that realistic samples of the environment can be drawn, an adaptable policy can be trained that attempts to make optimal decisions for any given possible instance of the environment \citep{ghosh2022offline, chen2021offline}. In this work, we examine the so-called ``probabilistic neural network'' (PNN) model that is ubiquitous in model-based reinforcement learning (MBRL) works. We argue that while PNN models may have good marginal uncertainties, they form a distribution of non-smooth transition functions. Not only are these samples unrealistic and may hamper adaptability, but we also assert that this leads to poor uncertainty estimates when predicting multiple step trajectory estimates. To address this issue, we propose a simple sampling method that can be implemented on top of pre-existing models.We evaluate our sampling technique on a number of environments, including a realistic nuclear fusion task, and find that, not only do smooth transition function samples produce more calibrated uncertainties, but they also lead to better downstream performance for an adaptive policy.

Offline Reinforcement Learning↗
DOE OSTI2025

Personalized and uncertainty-aware coronary hemodynamics simulations: From Bayesian estimation to improved multi-fidelity uncertainty quantification

Non-invasive simulations of coronary hemodynamics have improved clinical risk stratification and treatment outcomes for coronary artery disease, compared to relying on anatomical imaging alone. However, simulations typically use empirical approaches to distribute total coronary flow amongst the arteries in the coronary tree, which ignores patient variability, the presence of disease, and other clinical factors. Further, uncertainty in the clinical data often remains unaccounted for in the modeling pipeline. We present an end-to-end uncertainty-aware pipeline to (1) personalize coronary flow simulations by incorporating vessel-specific coronary flows as well as cardiac function; and (2) predict clinical and biomechanical quantities of interest with improved precision, while accounting for uncertainty in the clinical data. We assimilate patient-specific measurements of myocardial blood flow from clinical CT myocardial perfusion imaging to estimate branch-specific coronary artery flows. Simulated noise in the clinical data is used to estimate the joint posterior distributions of the model parameters using adaptive Markov Chain Monte Carlo sampling. Additionally, the posterior predictive distribution for the relevant quantities of interest is determined using a new approach combining multi-fidelity Monte Carlo estimation with non-linear, data-driven dimensionality reduction. This leads to improved correlations between high- and low-fidelity model outputs. Our framework accurately recapitulates clinically measured cardiac function as well as branch-specific coronary flows under measurement noise uncertainty. We observe substantial reductions in confidence intervals for estimated quantities of interest compared to single-fidelity Monte Carlo estimation and state-of-the-art multi-fidelity Monte Carlo methods. This holds especially true for quantities of interest that showed limited correlation between the low- and high-fidelity model predictions. In addition, the proposed multi-fidelity Monte Carlo estimators are significantly cheaper to compute than traditional estimators, under a specified confidence level or variance. The proposed pipeline for personalized and uncertainty-aware predictions of coronary hemodynamics is based on routine clinical measurements and recently developed techniques for CT myocardial perfusion imaging. The proposed pipeline offers significant improvements in precision and reduction in computational cost.

Bayesian parameter estimation↗
DOE OSTI2023

Demonstration of an AI-driven workflow for autonomous high-resolution scanning microscopy

Abstract Modern scanning microscopes can image materials with up to sub-atomic spatial and sub-picosecond time resolutions, but these capabilities come with large volumes of data, which can be difficult to store and analyze. We report the Fast Autonomous Scanning Toolkit (FAST) that addresses this challenge by combining a neural network, route optimization, and efficient hardware controls to enable a self-driving experiment that actively identifies and measures a sparse but representative data subset in lieu of the full dataset. FAST requires no prior information about the sample, is computationally efficient, and uses generic hardware controls with minimal experiment-specific wrapping. We test FAST in simulations and a dark-field X-ray microscopy experiment of a WSe 2 film. Our studies show that a FAST scan of <25% is sufficient to accurately image and analyze the sample. FAST is easy to adapt for any scanning microscope; its broad adoption will empower general multi-level studies of materials evolution with respect to time, temperature, or other parameters.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

A Continuous Action Space Tree search for INverse desiGn (CASTING) framework for materials discovery

Abstract Material properties share an intrinsic relationship with their structural attributes, making inverse design approaches crucial for discovering new materials with desired functionalities. Reinforcement Learning (RL) approaches are emerging as powerful inverse design tools, often functioning in discrete action spaces. This constrains their application in materials design problems, which involve continuous search spaces. Here, we introduce an RL-based framework CASTING (Continuous Action Space Tree Search for inverse design), that employs a decision tree-based Monte Carlo Tree Search (MCTS) algorithm with continuous space adaptation through modified policies and sampling. Using representative examples like Silver (Ag) for metals, Carbon (C) for covalent systems, and multicomponent systems such as graphane, boron nitride, and complex correlated oxides, we showcase its accuracy, convergence speed, and scalability in materials discovery and design. Furthermore, with the inverse design of super-hard Carbon phases, we demonstrate CASTING’s utility in discovering metastable phases tailored to user-defined target properties and preferences.

36 MATERIALS SCIENCE↗
DOE OSTI2015

THE MASS OF Kepler-93b AND THE COMPOSITION OF TERRESTRIAL PLANETS

Kepler-93b is a 1.478 ± 0.019 R {sub ⊕} planet with a 4.7 day period around a bright (V = 10.2), astroseismically characterized host star with a mass of 0.911 ± 0.033 M {sub ☉} and a radius of 0.919 ± 0.011 R {sub ☉}. Based on 86 radial velocity observations obtained with the HARPS-N spectrograph on the Telescopio Nazionale Galileo and 32 archival Keck/HIRES observations, we present a precise mass estimate of 4.02 ± 0.68 M {sub ⊕}. The corresponding high density of 6.88 ± 1.18 g cm{sup –3} is consistent with a rocky composition of primarily iron and magnesium silicate. We compare Kepler-93b to other dense planets with well-constrained parameters and find that between 1 and 6 M {sub ⊕}, all dense planets including the Earth and Venus are well-described by the same fixed ratio of iron to magnesium silicate. There are as of yet no examples of such planets with masses >6 M {sub ⊕}. All known planets in this mass regime have lower densities requiring significant fractions of volatiles or H/He gas. We also constrain the mass and period of the outer companion in the Kepler-93 system from the long-term radial velocity trend and archival adaptive optics images. As the sample of dense planets with well-constrained masses and radii continues to grow, we will be able to test whether the fixed compositional model found for the seven dense planets considered in this paper extends to the full population of 1-6 M {sub ⊕} planets.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Computation of high-frequency magnetoelastic waves in layered materials

Here, the direct calculation of magnetoelastic wave dispersion in layered media is presented using an efficient, accurate computational technique. The governing, coupled equations for elasticity and magnetism, the Navier and Landau-Lifshitz equations, respectively, are linearized to form a quadratic eigenvalue problem that determines a complex web of wave-number–frequency dispersion branches and their corresponding mode profiles. Numerical discretization of the eigenvalue problem via a spectral collocation method (SCM) is employed to determine the complete dispersion maps for both a single, finite-thickness magnetic layer and a finite magnetic-nonmagnetic double-layer. The SCM, previously used to study elastic waves in nonmagnetic media, is fast, accurate, and adaptable to a variety of sample configurations and geometries. Emphasis is placed on the extremely high-frequency regimes being accessed in ultrafast magnetism experiments. The dispersion maps and modes provide insight into how energy propagates through the coupled system, including how energy can be transferred between elastic- and magnetic-dominated waves as well as between different layers. The numerical computations for a single layer are further understood by a simplified analytical calculation in the high-frequency, exchange-dominated regime where the resonance condition required for energy exchange (an anticrossing) between quasi-elastic and quasi-magnetic dispersion branches is determined. Nonresonant interactions are shown to be well approximated by the dispersion of uncoupled elastic and magnetic waves. The methods and results provide fundamental theoretical tools to model and understand current and future magnetic devices powering spintronic innovation.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2017

Microbial dynamics in anaerobic digestion reactors for treating organic urban residues during the start-up process

Anaerobic digestion of organic residues offers economic benefits via biogas production, still methane (CH 4 ) yield relies on the development of a robust microbial consortia for adequate substrate degradation, among other factors. In this study, we monitor biogas production and changes in the microbial community composition in two semi-continuous stirred tank reactors during the setting process under mesophilic conditions (35°C) using a 16S rDNA high-throughput sequencing method. Reactors were initially inoculated with anaerobic granular sludge from a brewery wastewater treatment plant, and gradually fed organic urban residues (4·0 kg VS m -3 day -1 ) . The inocula and biomass samples showed changes related to adaptations of the community to urban organic wastes including a higher relative proportion of Clostridiales, with Ruminococcus spp. and Syntrophomonas spp. as recurrent species. Candidatus Cloacamonas spp. (Spirochaetes) also increased from ~2·2% in the inoculum to >10% in the reactor biomass. The new community consolidated the cellulose degradation and the propionate and amino acids fermentation processes. Acetoclastic methanogens were more abundant in the reactor, where Methanosaeta spp. was found as a key player. This study demonstrates a successful use of brewery treatment plant granular sludge to obtain a robust consortium for methane production from urban organic solid waste in Mexico.This study describes the selection of relevant bacteria and archaea in anaerobic digesters inoculated with anaerobic granular sludge from a brewery wastewater treatment plant. Generally, these sludge granules are used to inoculate reactors digesting organic urban wastes. Though, it is still not clearly understood how micro-organisms respond to substrate variations during the reactor start-up process. After feeding two reactors with organic urban residues, it was found that a broader potential for cellulose degradation was developed including Bacteroidetes, Firmicutes and Spirochaetes. These results clarify the bacterial processes behind new reactors establishment for treating organic wastes in urban areas.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2014

Machine Learning Toolkit for Extreme Scale

Support Vector Machines (SVM) is a popular machine learning technique, which has been applied to a wide range of domains such as science, finance, and social networks for supervised learning. MaTEx undertakes the challenge of designing a scalable parallel SVM training algorithm for large scale systems, which includes commodity multi-core machines, tightly connected supercomputers and cloud computing systems. Several techniques are proposed for improved speed and memory space usage including adaptive and aggressive elimination of samples for faster convergence , and sparse format representation of data samples. Several heuristics for earliest possible to lazy elimination of non-contributing samples are considered in MaTEx. In many cases, where an early sample elimination might result in a false positive, low overhead mechanisms for reconstruction of key data structures are proposed. The proposed algorithm and heuristics are implemented and evaluated on various publicly available datasets

Source record↗
DOE OSTI2016

Mads.jl

Mads.jl (Model analysis and decision support in Julia) is a code that streamlines the process of using data and models for analysis and decision support. It is based on another open-source code developed at LANL and written in C/C++ (MADS; http://mads.lanl.gov; LA-CC-11- 035). Mads.jl can work with external models of arbitrary complexity as well as built-in models of flow and transport in porous media. It enables a number of data- and model-based analyses including model calibration, sensitivity analysis, uncertainty quantification, and decision analysis. The code also can use a series of alternative adaptive computational techniques for Bayesian sampling, Monte Carlo, and Bayesian Information-Gap Decision Theory. The code is implemented in the Julia programming language, and has high-performance (parallel) and memory management capabilities. The code uses a series of third party modules developed by others. The code development will also include contributions to the existing third party modules written in Julia; this contributions will be important for the efficient implementation of the algorithm used by Mads.jl. The code also uses a series of LANL developed modules that are developed by Dan O'Malley; these modules will be also a part of the Mads.jl release. Mads.jl will be released under GPL V3 license. The code will be distributed as a Git repo at gitlab.com and github.com. Mads.jl manual and documentation will be posted at madsjulia.lanl.gov.

Vesselinov, Velimir↗
DOE OSTI2015

Development and testing of a prototype modular tritium and carbon-14 gas environmental monitoring system

Increasing quantities of radioactive waste are being placed into storage facilities. Many of the waste products contain organic materials which may undergo degradation leading to the release of tritium and carbon-14 species into waste containers and potentially into the environment. Monitoring for radioactive gas releases are required for environmental regulatory compliance and for radiation protection of facility workers. Research is currently being undertaken at the National Physical Laboratory (NPL) as part of a European Metrology Research Programme (EMRP) project MetroRWM to adapt and automate existing environmental sampling techniques for tritium and carbon-14 species. An innovative modular system is being developed which will lead to the introduction of an on-site small scale system capable of gas collection, liquid scintillation sample preparation and measurement. This paper outlines the evaluation of a liquid scintillation system that has been performed to date using active solutions of spiked trapping medium of similar activity concentrations to those anticipated in a waste repository. This system will operate using pre-set conditions for quench and luminescence derived from these and subsequent trials, unlike most other counters for which corrections for these phenomena are applied post measurement. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2025

Modeling Sepsis: Establishment and Validation of a 72-Hour Swine Model of Penetrating Abdominal Trauma

Background/Objectives: Fecal peritonitis following penetrating abdominal trauma is a serious condition that often results in sepsis and organ failure. The aim of our study was to develop a novel conscious porcine model of sepsis and organ dysfunction caused by multiple penetrating injuries to the small and large intestines. Methods: Twelve female Yorkshire pigs (average weight 50.6 ± 6.5 kg) were divided into two groups: Penetrating Abdominal Trauma (PAT) (n = 8) and Control (n = 4). All surgical procedures were performed under anesthesia with adequate analgesia. In the PAT group, the small and large intestines were punctured, and feces mixed with saline were introduced into the abdominal cavity to induce peritonitis. The Control group received sham surgery with only saline solution. The animals were observed in a conscious state over a period of 72 h, vital parameters were recorded, and blood samples were taken regularly. We adapted a pig-specific SOFA score and developed pig-specific SIRS criteria and NEWS2 score to assess organ function. The model was validated by independent investigators. Results: The survival rate in the PAT group was 75%, with an average survival time of 58.5 h, while all animals in the Control group survived to euthanasia. Monitoring showed pathophysiological changes, such as tachycardia, leucopenia, and thrombocytopenia, indicative of sepsis and organ dysfunction. Blinded investigators independently confirmed the model’s validity. Conclusions: A new swine model of penetrating abdominal trauma and sepsis has been successfully developed that demonstrates significant physiological and immunologic changes comparable to human sepsis. This new model provides a realistic platform for future research into sepsis, its diagnostics, and the evaluation of therapeutic strategies.

General & Internal Medicine↗
NASA NTRS1972

Kullback-Leibler information function and the sequential selection of experiments to discriminate among several linear models

The error variance of the process prior multivariate normal distributions of the parameters of the models are assumed to be specified, prior probabilities of the models being correct. A rule for termination of sampling is proposed. Upon termination, the model with the largest posterior probability is chosen as correct. If sampling is not terminated, posterior probabilities of the models and posterior distributions of the parameters are computed. An experiment was chosen to maximize the expected Kullback-Leibler information function. Monte Carlo simulation experiments were performed to investigate large and small sample behavior of the sequential adaptive procedure.

Sidik, S. M.↗