Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Variational inference”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Revealing Site Occupancy in a Complex Oxide: Terbium Iron Garnet

Complex oxide films stabilized by epitaxial growth can exhibit large populations of point defects which have important effects on their properties. The site occupancy of pulsed laser-deposited epitaxial terbium iron garnet (TbIG) films with excess terbium (Tb) is analyzed, in which the terbium:iron (Tb:Fe)ratio is 0.86 compared to the stoichiometric value of 0.6. The magnetic properties of the TbIG are sensitive to site occupancy, exhibiting a higher compensation temperature (by 90 K) and a lower Curie temperature (by 40 K) than the bulk Tb 3 Fe 5 O 12 garnet. Data derived from X-ray core-level spectroscopy, magnetometry, and molecular field coefficient modeling are consistent with occupancy of the dodecahedral sites by Tb 3+ , the octahedral sites by Fe 3+ , Tb 3+ and vacancies, and the tetrahedral sites by Fe 3+ and vacancies. Energy dispersive X-ray spectroscopy in a scanning transmission electron microscope provides direct evidence of Tb Fe antisites. A small fraction of Fe 2+ is present, and oxygen vacancies are inferred to be present to maintain charge neutrality. In conclusion, variation of the site occupancies provides a path to considerable manipulation of the magnetic properties of epitaxial iron garnet films and other complex oxides, which readily accommodate stoichiometries not found in their bulk counterparts.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Time Series Surface Temperature of Variably Inundated Sediment across 30 North American Rivers

This dataset supports a broader study examining drivers of organic matter chemistry in variably inundated hyporheic zone sediments and further linking that chemistry to biogeochemical rates. The dataset provides surficial temperature time series that can be used to infer the dynamics of inundation prior to the collection of sediments. Those inferred inundation histories can then be used to help interpret variation in the organic matter chemistry. There are related data that will be published, such as FTICR-MS data on organic matter chemistry and sediment moisture. A data package with those data is forthcoming.This dataset is comprised of two folders: (1) ECA1_iButtonData and (2) ECA1_SitePhotos. The ECA1_iButtonData folder contains: (1) file-level metadata, (2) data dictionary, (3) field metadata, (4) installation methods, (5) iButton deployment protocol, (6) readme, and (7) folder of individual time series temperature csv files for each iButton sensor deployed. The ECA1_SitePhotos folder contains site photographs taken in the field. All files are .csv, .txt, .pdf, or .jpg.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Sulphur variations in annually layered stalagmites using benchtop micro-XRF

Variation of sulphur in annually laminated stalagmites can be used to infer the impact of past volcanic activities, anthropogenic pollution, and climate change due to increased bushfire activity. The synchrotron radiation micro-X-Ray fluorescence (SR-XRF) microprobe is a powerful tool to analyse and image sulphur recorded in stalagmites with micrometre resolution. However, access to SR-XRF beamlines can be limited, so researchers must select the most promising stalagmites for imaging. Benchtop micro-XRF is an effective tool for trace elemental analysis of speleothem samples and is a candidate for routine laboratory measurement of sulphur along stalagmite laminae and screening for SR-XRF. Here, this study describes a protocol using matrix-matched standards to measure annual variations of sulphur at trace to percent level along the laminae of two Western Australian stalagmites, one of which already having been analysed using SR-XRF. Parameters that affected quantitation include X-ray tube voltage and current, spot size of the X-ray beam and stalagmite surface roughness and porosity. The use of a 20 μm X-ray spot size provides sub-annual spatial resolution that can be completed in an overnight scan. The features in a 1000 point micro-XRF analysis of sulphur along a 20 mm transect show good consistency with SR-XRF microprobe data. Micro-XRF mapping was also performed to produce chemical images on the stalagmite and compared with Raman and X-ray diffraction (XRD) to confirm that the stalagmite is exclusively calcite, with no aragonite, and that the source of sulphur in the samples was gypsum and anhydrite. Regions of very high sulphur in the micro-XRF maps were found to be artefacts due to diffraction of the incident beam but these could be efficiently removed by using a multiple point statistics approach to produce a clean image suitable for analysis of the laminae. This work shows the potential of micro-XRF for routine analysis of sulphur in stalagmites, and to streamline sample characterisation before SR-XRF imaging.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Neural network denoising of x-ray images from high-energy-density experiments

Noise is a consistent problem for x-ray transmission images of High-Energy-Density (HED) experiments because it can significantly affect the accuracy of inferring quantitative physical properties from these images. We consider experiments that use x-ray area backlighting to image a thin layer of opaque material within a physics package to observe its hydrodynamic evolution. The spatial variance of the x-ray transmission across the system due to changing opacity serves as an analog for measuring density in this evolving layer. The noise in these images adds nonphysical variations in measured intensity, which can significantly reduce the accuracy of our inferred densities, particularly at small spatial scales. Denoising these images is thus necessary to improve our quantitative analysis, but any denoising method also affects the underlying information in the image. In this paper, we present a method for denoising HED x-ray images via a deep convolutional neural network model with a modified DenseNet architecture. In our denoising framework, we estimate the noise present in the real (data) images of interest and apply the inferred noise distribution to a set of natural images. These synthetic noisy images are then used to train a neural network model to recognize and remove noise of that character. We show that our trained denoiser network significantly reduces the noise in our experimental images while retaining important physical features.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Variation Tolerant and Energy-Efficient Charge Domain Compute-in-Memory Array with Binary and Multi-Level Cell Ferroelectric FET

Here, in this work, we present a variation-tolerant and energy-efficient charge-domain Ferroelectric FET (FeFET) based Compute-in-Memory (CiM) array design that is compatible with both binary and multi-level cell memory sensing. We demonstrate that: 1) by exploiting FeFET as a nonvolatile switch, its high ON/OFF ratio in the subthreshold region can suppress the error introduced by the inaccurate ON state conductance, thus realizing robust CiM operations, unlike the current-domain CiM design where the computation results is highly sensitive to the device conductance variation; 2) by leveraging a dense dynamic random access memory (DRAM)-like 1FeFET1C cell structure, the proposed design benefits from the existing high density DRAM establishment while also significantly relaxing the capacitor retention and transistor leakage requirement; 3) the charge-domain CiM supports both binary FeFET with minimum overhead and MLC FeFET with tolerable latency for MLC state sensing, whose efficacy is validated experimentally on both cell-level and array-level; 4) the proposed CiM shows much better device variation resilience than conventional current-domain CiM, and also improves inference accuracy. Macro-level evaluation results demonstrate significantly higher energy efficiency and area efficiency compared to prior CiM works.

Duan, Jiahui [University of Notre Dame, IN (United↗
DOE OSTI2022

Probabilistic Modeling of Commercial Building Occupancy Patterns Using Location-Based Map Data: Preprint

Considering occupancy patterns is crucial to simulate buildings' energy use. Current energy models use inputs that simplify the actual diversity in occupancy into static occupancy patterns and are not able to represent the numerous variations in occupancy patterns between buildings and across different locations. Recently, inferring occupancy schedules from metered electricity consumption data was used to model occupancy in commercial buildings. However, the translation from metered data to occupancy schedules requires many assumptions that might not capture the reality, and the process is hindered by the availability of data from advanced metering infrastructure. With the development of information technologies, occupancy modeling should not be limited to traditional approaches. The prevalence of social networks and location services with real-time user feedback provides publicly accessible data via Maps Application Programming Interfaces (APIs) such as Google Maps, SafeGraph, Mapbox, Foursquare, etc. This paper presents an automated framework for modeling parametric occupancy patterns using such APIs to calibrate commercial district buildings' energy models. This process includes three main steps: data extraction and processing, parametric schedules generation, and schedules integration. We demonstrated this framework in districts where we used maps API to generate more accurate behavioral patterns for operations and electric vehicle charging events. We used these patterns to determine differences in energy use across key sociodemographic and spatial parameters. The presented method has the potential for worldwide applications. Users can utilize this framework to extract data for selected locations of interest to create more realistic behavioral patterns for commercial facilities across different districts.

building energy modeling↗
DOE OSTI2024

High-precision Galaxy Clustering Predictions from Small-volume Hydrodynamical Simulations via Control Variates

Abstract Cosmological simulations of galaxy formation are an invaluable tool for understanding galaxy formation and its impact on cosmological parameter inference from large-scale structures. However, their high computational cost is a significant obstacle for running simulations that probe cosmological volumes comparable to those analyzed by contemporary large-scale structure experiments. In this work, we explore the possibility of obtaining high-precision galaxy clustering predictions from small-volume hydrodynamical simulations such as MillenniumTNG and FLAMINGO via control variates. In this approach, the hydrodynamical full-physics simulation is paired with a matched low-resolution gravity-only simulation. By learning the galaxy–halo connection from the hydrodynamical simulation and applying it to the gravity-only counterpart, one obtains a galaxy population that closely mimics the one in the more expensive simulation. One can then construct an estimator of galaxy clustering that combines the clustering amplitudes in the small-volume hydrodynamical and gravity-only simulations with clustering amplitudes in a large-volume gravity-only simulation. Depending on the galaxy sample, clustering statistic, and scale, this galaxy clustering estimator can have an effective volume of up to around 100 times the volume of the original hydrodynamical simulation in the nonlinear regime. With this approach, we can construct galaxy clustering predictions from existing simulations that are precise enough for mock analyses of next-generation large-scale structure surveys such as the Dark Energy Spectroscopic Instrument and the Legacy Survey of Space and Time.

Doytcheva, Alexandra (ORCID:0009000111254888)↗
DOE OSTI2023

Inferring Quantum Network Topology Using Local Measurements

Statistical correlations that can be generated across the nodes in a quantum network depend crucially on its topology. However, this topological information might not be known a priori, or it may need to be verified. In this paper, we propose an efficient protocol for distinguishing and inferring the topology of a quantum network. We leverage entropic quantities-namely, the von Neumann entropy and the measured mutual information-as well as measurement covariance to uniquely characterize the topology. We show that the entropic quantities are sufficient to distinguish two networks that prepare GHZ states. Moreover, if qubit measurements are available, both entropic quantities and covariance can be used to infer the network topology without state-preparation assumptions. We show that the protocol can be entirely robust to noise and can be implemented via quantum variational optimization. Numerical experiments on both classical simulators and quantum hardware show that covariance is generally more reliable for accurately and efficiently inferring the topology, whereas entropy-based methods are often better at identifying the absence of entanglement in the low-shot regime.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Design of multi neutron-to-gamma converter array for measuring time resolved ion temperature of inertial confinement fusion implosions

The ion temperature varying during inertial confinement fusion implosions changes the amount of Doppler broadening of the fusion products, creating subtle changes in the fusion neutron pulse as it moves away from the implosion. A diagnostic design to try to measure these subtle effects is introduced—leveraging the fast time resolution of gas Cherenkov detectors along with a multi-puck array that converts a small amount of the neutron pulse into gamma-rays, one can measure multiple snapshots of the neutron pulse at intermediate distances. Further, precise measurements of the propagating neutron pulse, specifically the variation in the peak location and the skew, could be used to infer time-evolved ion temperature evolved during peak compression.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Understanding asymmetries using integrated simulations of capsule implosions in low gas-fill hohlraums at the National Ignition Facility

Current capsule implosions at the National Ignition Facility (NIF) using high-density-carbon ablators and laser energies close to 2 MJ have shown neutron yields in excess of 50 kJ. Improving on this performance requires understanding of the different degradation mechanisms. For many NIF implosions, nuclear diagnostic signatures have inferred residual hot-spot velocities that correlate with fuel areal density variations consistent with a low-order mode-1 asymmetry. A current working hypothesis attributes these asymmetries to a combination of beam to beam variations in the laser delivery and possibly coupling to target features, such as the diagnostic holes needed for x-ray imaging. Recently, a new source has been identified, thickness variation in the ablator shell. To gain better understanding and eventually mitigate the causes of <ρR> asymmetries, 3D integrated simulations using the actual delivered laser powers are needed. To capture the effect of the diagnostic holes (DHs) using direct numerical simulation would require significantly large computational resources. Instead, our 3D simulations make use of a subgrid model developed using highly-resolved 2D simulations that include several details of the DH engineering complexity. Simulations of NIF shots using hohlraums without DHs, to isolate the effect of beam-to-beam variations, reproduce fairly well the observed nuclear diagnostic signatures. Similarly, reasonable agreement between data and simulations is also obtained in the presence of diagnostic holes. To account for the remaining discrepancies a sensitivity study of ablator thickness variation showed that 1% thickness asymmetries are comparable in effect to 1% peak drive mode-1 asymmetries. Furthermore, this study identified sensitivity to variations in the imbedded doped layer (needed to shield the DT ice from the high energy x-rays generated in the hohlraum) thickness even when the inner and outer surface of the ablator are perfectly spherical.

radiation hydrodynamics↗
DOE OSTI2023

Inferring colloidal interaction from scattering by machine learning

A machine learning solution for the potential inversion problem in elastic scattering is outlined. The inversion scheme consists of two major components, a generative network featuring a variational autoencoder which extracts the targeted static two-point correlation functions from experimentally measured scattering cross sections, and a Gaussian process framework which probabilistically infers the relevant structural parameters from the inverted correlation functions. Via a case study of charged colloidal suspensions, the feasibility of this approach for quantitative study of molecular interaction is critically benchmarked and its merit over existing deterministic approaches, in terms of numerical accuracy and computationally efficiency, is demonstrated.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Implementation of a Binary Neural Network on a Passive Array of Magnetic Tunnel Junctions

The increasing scale of neural networks and their growing application space have produced demand for more energy- and memory-efficient artificial-intelligence-specific hardware. Avenues to mitigate the main issue, the von Neumann bottleneck, include in-memory and near-memory architectures, as well as algorithmic approaches. In this report we leverage the low-power and the inherently binary operation of magnetic tunnel junctions (MTJs) to demonstrate neural network hardware inference based on passive arrays of MTJs. In general, transferring a trained network model to hardware for inference is confronted by degradation in performance due to device-to-device variations, write errors, parasitic resistance, and nonidealities in the substrate. To quantify the effect of these hardware realities, we benchmark 300 unique weight matrix solutions of a two-layer perceptron to classify the Wine dataset for both classification accuracy and write fidelity. Despite device imperfections, we achieve software-equivalent accuracy of up to 95.3% with proper tuning of network parameters in 15 x 15 MTJ arrays having a range of device sizes. The success of this tuning process shows that new metrics are needed to characterize the performance and quality of networks reproduced in mixed signal hardware.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

Determinants of photochemical characteristics of the photosynthetic electron transport chain of maize

The photosynthetic electron transport chain (ETC) is the bridge that links energy harvesting during the photophysical reactions at one end and energy consumption during the biochemical reactions at the other. Its functioning is thus fundamental for the proper balance between energy supply and demand in photosynthesis. Currently, there is a lack of understanding regarding how the structural properties of the ETC are affected by nutrient availability and plant developmental stages, which is a major roadblock to comprehensive modeling of photosynthesis. Redox parameters reflect the structural controls of ETC on the photochemical reactions and electron transport. We conducted joint measurements of chlorophyll fluorescence (ChlF) and gas exchange under systematically varying environmental conditions and growth stages of maize and sampled foliar nutrient contents. We utilized the recently developed steady-state photochemical model to infer redox parameters of electron transport from these measurements. We found that the inferred values of these photochemical redox parameters varied with leaf macronutrient content. These variations may be caused either directly by these nutrients being components of protein complexes on the ETC or indirectly by their impacts on the structural integrity of the thylakoid and feedback from the biochemical reactions. Also, the redox parameters varied with plant morphology and developmental stage, reflecting seasonal changes in the structural properties of the ETC. Our findings will facilitate the parameterization and simulation of complete models of photosynthesis.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Mineralogy, fabric and deformation domains in D" across the southwestern border of the African LLSVP

Recent advances in seismic anisotropy studies that jointly use reflections and shear wave splitting have proven to place tight constraints on the plausible anisotropic and deformation scenarios in the D" region. We apply this novel methodology to a large area of the D" region beneath the South Atlantic, in proximity to and within the African large low seismic velocity province (LLSVP). This area of the mantle is characterized by a transition from fast to slow seismic velocity anomalies and it is thought to be the location of deep-seated plumes responsible for hotspot volcanism. Attempting to probe mantle composition and deformation along the LLSVP borders may provide key information on mantle dynamics. By analysing seismic phases sampling this region, we detect a D" discontinuity over a large area beneath the South Atlantic, with inferred depth ranges ~170 to ~240 km above the core–mantle boundary. We find evidence for a D" reflector within the area of the LLSVP. Shear wave splitting observations suggest that anisotropy is present in this region of the mantle, in agreement with previous studies that partially sampled this region. We model the observations considering lattice- and shape-preferred orientation of materials expected in the D" region. A regional variation of mineralogy, phase transition boundaries, and deformation direction is required to explain the data. We infer two distinct domains of mineralogy and deformation: aligned post-perovskite outside the LLSVP and aligned bridgmanite within the LLSVP. Finally, the scenario depicted by this study agrees well with the current hypotheses for the composition of the LLSVP and with the prevalence of vertical deformation directions expected to occur along the LLSVPs borders.

58 GEOSCIENCES↗
DOE OSTI2023

Theory of an Automatic Seepage Meter and Ramifications for Applications

A new approach for measuring fluxes across surface water—groundwater interfaces was recently proposed. The Automatic Seepage Meter (ASM) is equipped with a precise water level sensor and digital memory that analyzes water level time series in a vertical tube inserted into a streambed. The ability to infer flux values with high temporal resolution relies on an accurate interpretation of water level dynamics inside the tube. Here, we reduce the three-dimensional hydrodynamic problem that describes the ASM water level in a variety of field conditions to a single ordinary differential equation. This novel general analytical solution for estimating ASM responses is more comprehensive and flexible than previous approaches and is applicable to the entire range of field conditions, including steady or transient stream stages, evaporation, rainfall, and noise. For example, our analysis determines the timing of the nonmonotonic ASM response to a monotonic linear stream stage variation and explains previously used empirical parabolic approximation for estimating fluxes. We present algorithms for simultaneous inference of vertical interface flux and hydraulic conductivity values together with an example code. We quantify how the accuracy of parameter estimation depends on test duration and noise amplitude and propose how our analysis can be used to optimize field test protocols. On this basis, changing the ASM geometry by increasing the radius and decreasing tube insertion depth may enable ASM field test protocols that estimate interface flux and hydraulic conductivity faster while maintaining desired accuracy. Potential applications of joint parameter estimation are suggested.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

A Hardware and Software Co-design Framework for Energy Efficient Neuromorphic Systems

Neuromorphic systems can be realized by a variety of algorithms and architectures. A common understanding is that spiking neuromorphic designs, which encode information into spatio-temporal spiking events, are both a biologically-accurate and efficient way of processing information. However, representing the information through timing relationships induces sophisticated circuit designs in traditional CMOS-based implementations. In recent years, high-capacity resistive memory (RRAM, aka, memristor) has demonstrated great potential in mimicking synaptic behaviors. Several RRAM-based spiking neuromorphic designs exist, most of which focus on rate coding schemes. These designs simplify circuit implementations of neuron models and explore challenges such as unsatisfactory speed, resolution, and performance. As an alternative, we will explore temporal coding spiking neuromorphic systems that encode information as the relative timing of neuron activations (spikes), which have been proven to be more adaptive and energy-efficient. Developing a neuromorphic system for spiking neural network (SNN) inference and online training, however, faces some major technical challenges: (1) It lacks circuit implementation support for temporal-coding SNN to achieve satisfying power efficiency and accuracy; (2) Although existing research works have investigated memristive synapse and neuron designs for spike-timing-dependent plasticity, the non-ideal conditions in implementation, such as device variations and signal degradation, degrade online learning accuracy of large scale systems; and (3) Non-optimized, inter-layer data traffic in SNNs, leads to unnecessary data communication costs. In this project, we plan to address these challenges by a hardware and software co-design framework that incorporates solutions at the circuit, architecture, and algorithm levels. At the circuit-level, we will elaborate on the in-situ SNN processing element designs for supporting both inference and online training modes. Variation-aware schemes will be studied to improve reliability. At the architecture level, we propose a pipelined, asynchronous architecture to retain the timing resolution of spikes. At the algorithm level, we will investigate an innovative SNN training algorithm for enabling activation sparsification and reducing unnecessary data communication costs. This neuromorphic system will provide an effective solution to real-life energy-constrained applications and significantly contribute to the exploration of next-generation high-performance computing systems under the DOE context.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Machine Learning Models for Network Traffic Classification in Programmable Logic

Network traffic classification via machine learning on network packet payloads has emerged as an active area of research for network security due to the high accuracy machine learning models have achieved in classifying payloads. For effective deployment as part of network security, these machine learning models must not only classify malicious packet payloads accurately, they must also identify anomalous payloads and perform inference at speeds generally faster than 10,000 packets per second to be effective. This work explores the in- ference speeds and accuracy of several neural network models implemented in programmable logic on various field programmable gate arrays (FPGA) including the Xilinx VC1902 and Xilinx Zynq Ultrascale+. This work also presents the design and performance of both an autoencoder and variational autoencoder programmed on the FPGA for identifying anomalous packet payloads. The performance benefits of the FPGA implementation for this type of packet payload inspection driven by machine learning are compared against graphics processing unit (GPU) inference implementations run on two state-of-the-art datacenter GPU devices, the NVIDIA V100 and A100. The model accuracy difference between the FPGA and GPU implementations was found to be 4% or less while the Xilinx VC1902 outperformed both the NVIDIA V100 and A100 for inference speeds on all the models explored except the variational autoencoder.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Watersheds and stream networks viewed longitudinally: Example insights from novel spatial portrayals of watershed characteristics

Abstract Longitudinal depictions of watershed structure and characteristics, including topography, stream networks, wetlands, ground water levels, and land use, can provide watershed knowledge and understanding unavailable from standard plan view maps. Three case studies provide examples of knowledge gained by applying longitudinal views of stream networks, watershed hydrologic behavior, and land use distributions. Longitudinal views of mountain stream networks show extreme variability in the slope‐area relationships of low Strahler order streams, large discontinuities in drainage area (large parts of drainage area space are absent in networks), and large variations in network curvature. Longitudinal views of a groundwater‐dominated headwater watershed increase the inference available from limited groundwater observations and clearly reveal how groundwater connections affect the permanence of surface water features and the distribution of vadose zone storage in the landscape. Plotting land uses longitudinally illuminates and allows a quantitative analysis of how land uses are distributed relative to topographic position. Viewing watersheds and stream networks longitudinally can provide new insights into watershed forms and processes and motivate new questions and research.

60 APPLIED LIFE SCIENCES↗