Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Kernel”

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 1,459 records · Page 81
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

Separate effects irradiation testing of miniature fuel specimens

We report that qualification of new nuclear fuels is necessary for their deployment and requires a thorough understanding of fuel behavior under irradiation. Traditionally, nuclear fuels have been qualified by performing exhaustive integral tests under a limited range of prototypic conditions designed for their specific reactor application. While some integral fuel testing is essential, basic data on behavior and property evolution under irradiation can be obtained from separate effects tests. These irradiations could offer reduced cost, reduced complexity, and in the case of accelerated testing, reduced time to achieve a given burnup. Furthermore, it may be desirable to design test irradiations capable of deconvoluting the myriad effects of burnup, temperature gradients, and other factors inherent to integral irradiation tests. Oak Ridge National Laboratory has developed an experimental capability to perform separate effects irradiation testing of miniature fuel specimens in the High Flux Isotope Reactor (HFIR): the “MiniFuel” irradiation vehicle. The small size (<4 mm 3 ) of the fuel specimens simplifies the design, analysis, and post-irradiation examination. By reducing the fuel mass, the total heat generated inside the experiment vehicle can be dominated by gamma heating in the structural materials instead of fission heating in the fuel. This essentially decouples the fuel temperature from the fission rate, allowing for highly accelerated testing (3X-18X the burnup rate of a typical light water reactor for 235 U enrichments varying from 0.22 wt% to 8 wt%) and an extremely flexible experiment design that can accommodate a wide range of fuel temperatures (~100 °C to >1200 °C), compositions, enrichments, and even geometries without requiring detailed analyses for each fuel variant. In conclusion, this paper summarizes the experiment design concept, evaluates potential applications for specific fuel forms, and briefly describes the first set of experiments on uranium nitride kernels that have been assembled and are currently being irradiated in the HFIR.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2021

Modeling fission product diffusion in TRISO fuel particles with BISON

Diffusion of fission products in intact TRISO particles depends on particle geometry, fission product source rates, time, temperature, and temperature-dependent diffusion coefficients. Simulating this diffusion process requires models for source rates and diffusion coefficients, plus computation of the temperature field if not prescribed. In addition, simulation quality depends on discretization of the geometry, appropriate time stepping, and the accuracy of the solution method. In this paper, we explore the simulation of fission product diffusion in TRISO fuel particles using the finite element method via the fuel performance code Bison. Recent material model development has occurred in Bison for each material present in tri-structural isotropic (TRISO) fuel particles: the buffer, inner pyrolytic carbon, silicon carbide, and outer pyrolytic carbon layers, as well as the fuel kernel. Also, new mesh generation and fission product release fraction capabilities have been added. Diffusion capabilities are shown to converge to the correct solution via formal verification tests. A large number of code benchmarking problems are also given, with good results, showing that Bison’s computed release fractions closely match those of other software tools. Finally, a significant validation effort is detailed in which fission product release, measured as part of the AGR-1 capsule experiments, is compared to Bison outputs. Bison outputs compare very well to the experimental data and to PARFUME results.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2021

Numerical evaluation of AGR-2 fission product release

The AGR-2 experiment produced normal operation and accident condition fuel performance data for tri-structural isotropic (TRISO) particles with UCO and UO 2 kernels. Data for compacts with no failed particles and for compacts with one or more failed particles are available. To model the fission product diffusion and release from these compacts, it is important to consider the computational mesh as well as the heat conduction and mass diffusion properties of the materials in a TRISO particle. Code and solution verification studies with the Bison fuel performance code were performed to show that the code is computing correct solutions. Bison comparisons to post-irradiation examination data and PARFUME predictions were made for several sets of compacts. Furthermore, these comparisons considered silver, cesium, strontium, and krypton with and without failed particles. Bison predictions closely match those of PARFUME, while both codes generally overpredict post-irradiation examination data.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2022

Mechanistic calculation of the effective silver diffusion coefficient in polycrystalline silicon carbide: Application to silver release in AGR-1 TRISO particles

The silicon carbide (SiC) layer in tristructural isotropic (TRISO) fuel particles serves as a barrier to prevent the escape of fission products produced and not retained in the fuel kernel. The release of silver (Ag) is a concern due to the long half-life of the 110m Ag isotope. However, accurately determining the fission gas release rate requires knowing the diffusion coefficient through the SiC layer. In this study, we leverage atomistic calculations of Ag diffusivity in SiC bulk and grain boundaries (GBs) to develop a mesoscale effective Ag diffusion coefficient (D eff ) in SiC. Since GBs serve as pathways for Ag diffusion, D eff is defined as a function of temperature and microstructure variables. In particular, the size of SiC grains in the direction perpendicular to diffusion is shown to significantly affect Ag diffusion. The prediction of the mechanistic, mesoscale approach falls within one order of magnitude of empirical values. The temperature and microstructure-dependent effective Ag diffusivity in SiC is implemented in the fuel performance code Bison with a correction factor to predict Ag release from AGR-1 TRISO fuel particles. We hereby quantify the impact of SiC grain size on Ag release and improve Bison’s predictions.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2022

Microstructural heterogeneity of the buffer layer of TRISO nuclear fuel particles

Tristructural isotropic (TRISO) nuclear fuel particles contain a layered spherical shell designed to retain fission products; however, failure occurs in rare cases—commonly initiated in the porous pyrocarbon buffer layer. Achieving a comprehensive understanding of the buffer-initiated failure mechanisms requires detailed characterization of the buffer porosity and its heterogeneous distribution across multiple length scales. Here we performed FIB-SEM tomography across the buffer layer thickness (~100 µm) to produce 3D reconstructions of the buffer microstructure with 50 nm spatial resolution. We found an average overall porosity of ~14%, which does not solely account for the low density of the buffer (50% of the theoretical density). Additionally, the local porosity and its fluctuation increase from the kernel interface towards the inner pyrocarbon (IPyC) layer, which we attribute to the chemical vapor deposition process conditions during the TRISO particle fabrication. Detailed characterization of the porous microstructure—including analysis of the pore size, distribution, shape, and orientation—provides insight into the process-structure-property-performance relations of TRISO nuclear fuel particles and will inform multiscale models designed to predict the failure of TRISO particles under irradiation.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Machine learning models for volumetric swelling in uranium nitride

Machine learning methods are applied to predict the volumetric swelling rate of the nuclear fuel uranium nitride (UN) over various temperatures, irradiation conditions, and power densities. Both kernel-based methods and symbolic regression models for UN swelling are developed and compared with multiple experimental datasets. We find that the UN pellet geometry and dimensions must be taken into account to accurately model swelling behavior. Strong agreement is observed between the developed machine learning models and the data. The predictive error generated by the machine learning models improves on empirical models taken from the literature. Sensitivity analysis is performed to determine which properties such as temperature, burnup, and power density, are most important in the swelling process. We find that machine learning can be used to quickly develop accurate swelling models for nuclear materials. In conclusion, the presented results illustrate the potential of machine learning to determine volumetric swelling in UN.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2026

A mechanistic, multiscale model for predicting Pd penetration in TRISO fuels using BISON

TRistructural ISOtropic (TRISO) particles use silicon carbide (SiC) as the primary structural member and barrier against metallic fission product (FP) release. palladiums (PDs), produced by fission in the fuel kernel, can diffuse to and chemically interact with the SiC layer, degrading its structural integrity and ability to contain radioactive FPs. Existing temperature-dependent correlations for predicting Pd penetration rely on experimental data with significant scatter due to varying conditions, potentially complicating ongoing fuel qualification and licensing efforts for advanced reactors that would subject TRISO fuels to operating conditions outside of those examined in the experiments. A mechanistic model of Pd production, transport, and reaction is developed in this work to better understand and predict PD attack of SiC in TRISO particles. molecular dynamicss (MDs) simulations are utilized to calculate the Pd diffusivity in SiC grain bulk and grain boundaries. A mesoscale phase-field diffusion model, informed by the MD diffusivities, is used to develop a reduced order model (ROM) for the effect of SiC microstructure on the Pd penetration rate. The engineering scale BISON model calculates Pd production and transport, utilizing the ROM to predict penetration rates consistent with experimental data. This novel mechanistic ROM captures the effect of temperature, microstructure, and irradiation history on Pd penetration. In conclusion, these new capabilities are expected to support ongoing qualification and licensing efforts associated with near-term TRISO-fueled reactor applications.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2014

Streaming data analytics via message passing with application to graph algorithms

The need to process streaming data, which arrives continuously at high-volume in real-time, arises in a variety of contexts including data produced by experiments, collections of environmental or network sensors, and running simulations. Streaming data can also be formulated as queries or transactions which operate on a large dynamic data store, e.g. a distributed database. We describe a lightweight, portable framework named PHISH which enables a set of independent processes to compute on a stream of data in a distributed-memory parallel manner. Datums are routed between processes in patterns defined by the application. PHISH can run on top of either message-passing via MPI or sockets via ZMQ. The former means streaming computations can be run on any parallel machine which supports MPI; the latter allows them to run on a heterogeneous, geographically dispersed network of machines. We illustrate how PHISH can support streaming MapReduce operations, and describe streaming versions of three algorithms for large, sparse graph analytics: triangle enumeration, subgraph isomorphism matching, and connected component finding. Lastly, we also provide benchmark timings for MPI versus socket performance of several kernel operations useful in streaming algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Kokkos: Enabling manycore performance portability through polymorphic memory access patterns

The manycore revolution can be characterized by increasing thread counts, decreasing memory per thread, and diversity of continually evolving manycore architectures. High performance computing (HPC) applications and libraries must exploit increasingly finer levels of parallelism within their codes to sustain scalability on these devices. We found that a major obstacle to performance portability is the diverse and conflicting set of constraints on memory access patterns across devices. Contemporary portable programming models address manycore parallelism (e.g., OpenMP, OpenACC, OpenCL) but fail to address memory access patterns. The Kokkos C++ library enables applications and domain libraries to achieve performance portability on diverse manycore architectures by unifying abstractions for both fine-grain data parallelism and memory access patterns. In this paper we describe Kokkos’ abstractions, summarize its application programmer interface (API), present performance results for unit-test kernels and mini-applications, and outline an incremental strategy for migrating legacy C++ codes to Kokkos. Furthermore, the Kokkos library is under active research and development to incorporate capabilities from new generations of manycore architectures, and to address a growing list of applications and domain libraries.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Massively parallel first-principles simulation of electron dynamics in materials

Here we present a highly scalable, parallel implementation of first-principles electron dynamics coupled with molecular dynamics (MD). By using optimized kernels, network topology aware communication, and by fully distributing all terms in the time-dependent Kohn–Sham equation, we demonstrate unprecedented time to solution for disordered aluminum systems of 2000 atoms (22,000 electrons) and 5400 atoms (59,400 electrons), with wall clock time as low as 7.5 s per MD time step. Despite a significant amount of non-local communication required in every iteration, we achieved excellent strong scaling and sustained performance on the Sequoia Blue Gene/Q supercomputer at LLNL. We obtained up to 59% of the theoretical sustained peak performance on 16,384 nodes and performance of 8.75 Petaflop/s (43% of theoretical peak) on the full 98,304 node machine (1,572,864 cores). Lastly, scalable explicit electron dynamics allows for the study of phenomena beyond the reach of standard first-principles MD, in particular, materials subject to strong or rapid perturbations, such as pulsed electromagnetic radiation, particle irradiation, or strong electric currents.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Accelerating matrix-centric graph processing on GPUs through bit-level optimizations

Even though it is well known that binary values are common in graph applications (e.g., adjacency matrix), how to leverage the phenomenon for efficiency has not yet been adequately explored. This paper presents a systematic study on how to unlock the potential of the bit-level optimizations of graph computations that involve binary values. It proposes a two-level representation named Bit-Block Compressed Sparse Row (B2SR) and presents a series of optimizations to the graph operations on B2SR by the intrinsics of modern GPUs. It additionally introduces Deep Reinforcement Learning (DRL) as an efficient way to best configure the bit-level optimizations on the fly. Additionally, the DQN-based adaptive tile size selector with dedicated model training can reach 68% prediction accuracy. Evaluations on NVIDIA Pascal and Volta GPUs show that the optimizations bring up to 40× and 6555× for essential GraphBLAS kernels SpMV and SpGEMM, respectively, making GraphBLAS-based BFS accelerate up to 433×, SSSP, PR, and CC up to 35×, and TC up to 52×.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

FVCOM-plume – A Three-dimensional Lagrangian Outfall Plume Dilution and Transport Model for Dynamic Tidal Environments: Model Development

Although numerous steady-state effluent plume dilution models are in use, their application in tidal environments remains a challenge. Three-dimensional dynamic circulation models are also inadequate, often due to the lack of required resolution and simplifying assumptions. To overcome these limitations, we present FVCOM-plume—an outfall plume dilution and transport model that operates within the Finite Volume Coastal Ocean Model (FVCOM) framework. It provides simultaneous inclusion of near-field dilution and far-field plume transport processes. The near-field is based on UM3 model using Lagrangian Control Volume approach to compute buoyant plume trajectory and dilution from multiport diffusers. The far-field uses neutrally buoyant particles with point masses and the random walk method to solve unsteady advection-diffusion processes. A density kernel approach is used to compute concentrations at point locations and analyze far-field plume characteristics. The results demonstrate the ability of FVCOM-plume to simultaneously capture near-field and far-field effluent plume dynamics in tidal environments.

Plume Dilution Models↗
DOE OSTI2014

Growth kinetics of Al–Fe intermetallic compounds during annealing treatment of friction stir lap welds

In this study, we explored the growth kinetics of the Al–Fe intermetallic (IM) layer at the joint interface of the St-12/Al-5083 friction stir lap welds during post-weld annealing treatment at 350, 400 and 450 °C for 30 to 180 min. Optical microscope (OM), field emission gun scanning electron microscope (FEG-SEM) and transmission electron microscope (TEM) were employed to investigate the structure of the weld zone. The thickness and composition of the IM layers were evaluated using image analysis system and electron back-scatter diffraction (EBSD), respectively. Moreover, kernel average misorientation (KAM) analysis was performed to evaluate the level of stored energy in the as-welded state. The results showed that the growth kinetics of the IM layer was not governed by a parabolic diffusion law. Presence of the IM compounds as well as high stored energy near the joint interface of the as-welded sample was recognized to be the origin of the observed deviation from the parabolic diffusion law. - Highlights: • This work provided a new insight into growth kinetics of Al–Fe IM thickness. • The growth kinetics of IM layer was not governed by a parabolic diffusion law. • IM near the joint interface was the origin of deviation from the parabolic law. • High stored energy at joint interface was origin of deviation from parabolic law.

36 MATERIALS SCIENCE↗
DOE OSTI2016

Mechanical behavior and microstructure of Ti{sub 20}Hf{sub 20}Zr{sub 20}Ta{sub 20}Nb{sub 20} high-entropy alloy loaded under quasi-static and dynamic compression conditions

The microstructure and the mechanical behavior of equimolar Ti{sub 20}Hf{sub 20}Zr{sub 20}Ta{sub 20}Nb{sub 20} high-entropy alloy in a wide range of initial strain rates between ~ 10{sup −3} s{sup −1} and ~ 3.4 × 10{sup 3} s{sup −1} were studied. A significant increment in the yield strength with increasing strain rate was observed. The yield strength at ~ 3.4 × 10{sup 3} s{sup −1} was about 40% higher than that measured at ~ 10{sup −3} s{sup −1}. Analysis by electron backscatter diffraction shows that in the low strain rate regime (up to ~ 10 s{sup −1}) the deformation occurs mainly in evenly distributed bands, while in the dynamic regime the deformation is strongly localized in macroscopic shear bands accompanied by softening even after the onset of yielding. The Kernel Average Misorientation technique reveals a high level of lattice rotation within these bands that also carries intense shear. In addition, X-ray diffraction line profile analysis indicates that the sharp increase in the flow stress is mostly related to an increase of the dislocation density. - Highlights: • Strain rate effect on the plastic behavior of Ti{sub 20}Hf{sub 20}Zr{sub 20}Ta{sub 20}Nb{sub 20} HEA was studied. • Low strain rate regime was characterized by a continuous hardening. • At high strain rates softening occurred shortly after the onset of yielding. • Intense strain localization in shear bands occurred in the high strain rate regime. • High dislocation density explained the upturn of flow stress at high strain rates.

36 MATERIALS SCIENCE↗
DOE OSTI2016

Characterization of as-welded microstructure of heat-affected zone in modified 9Cr–1Mo–V–Nb steel weldment

Non-equilibrium microstructure of the heat-affected zone (HAZ) in the as-welded modified 9Cr–1Mo–V–Nb pipe steel (P91) weldment deposited by gas tungsten arc welding (GTAW) and flux core arc welding (FCAW) has been characterized by field-emission scanning electron microscope (FESEM) and electron backscatter diffraction (EBSD). The heterogeneous structures in the sub-layers of the as-welded HAZ are attributable to phase transformations caused by the welding thermal cycles and the local structure variations in the as-received base metal. Coarse-grained heat-affected zone (CGHAZ) has a prior austenite grain (PAG) size of 20 μm. Fine uniformly-distributed precipitates and a higher fraction of MX carbonitrides are observed in the CGHAZ. Fine-grained heat-affected zone (FGHAZ) consists of the finest grains (1.22 μm measured by EBSD, 5 μm PAG size), coarse undissolved M{sub 23}C{sub 6} carbides within the PAG boundaries and fine nucleated M{sub 23}C{sub 6} particles within the martensite laths. Inter-critical heat-affected zone (ICHAZ) consists of partially austenitized grains and over-tempered martensite laths. EBSD kernel average misorientation (KAM) map in the FGHAZ close to the ICHAZ illustrates the greatest local strain variations with a moderate normalized KAM value of 0.92°. The majority (88.1%) of the matrix grains in the CGHAZ are classified as deformed grains by EBSD grain average misorientation (GAM) evaluation. The FGHAZ close to the ICHAZ has the most recrystallized grains with an area fraction of 14.4%. The highest density variation of precipitates within grains in the FGHAZ originates from the inhomogeneous chemistry in the base metal. - Highlights: •A comprehensive characterization of the as-welded HAZ of P91 weldment is conducted. •Structural features in the each layer of the HAZ are quantified by EBSD. •Structural heterogenities in HAZ are due to welding cycle and base metal structure. •FGHAZ contains the finest grain structure and largest precipitate density variation.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Ultrafine gradient microstructure induced by severe plastic deformation under sliding contact conditions in copper

Highlights: • Sliding contact produces a gradient microstructure by severe plastic deformation. • Nanohardness is increased strongly by grain size and dislocation hardening. • The texture components in the deformed layer correspond to pure shear. • Grain size distribution, boundary misorientation and KAM were analysed. • Strain induced effects compete with crystallisation to determine the microstructure. - Abstract: Sliding contact induces severe plastic deformation (SPD) at the surface of ductile materials and induces a microstructural gradient associated to a significant increase of hardness toward the surface. This gradient allows observing all stages of grain refinement in SPD, as illustrated here by the analysis of polycrystalline electrolytic copper tested in a coaxial tribometer. Materials tested in the cold-rolled state and after annealing were characterised by high-resolution electron backscattering diffraction and nano-indentation. The incremental plastic strain produces an ultrafine microstructure in the top layers, which gradually changes to the original size in unaffected material. In cold-rolled material, an intermediate recrystallised layer is observed. The separation of the Misorientation Angle Distribution (MAD) in a low-angle portion and a high angle portion allows characterising the accumulation of strain induced misorientation, while the Kernel Average Misorientation (KAM) provides information on the evolution of substructure at the finest levels. The results point toward a process where strain-induced effects compete continuously with recrystallisation, except for the surface layer in the cold-rolled material, where dynamic recrystallisation is dominant. Combining the information from KAM and subgrain size distribution, the measured hardness can be explained as a combination of grain size and dislocation hardening.

36 MATERIALS SCIENCE↗
DOE OSTI2026

RU-net for automatic characterization of TRISO fuel cross sections

During irradiation, phenomena such as kernel swelling and buffer densification may impact the performance of tristructural isotropic (TRISO) particle fuel. Post-irradiation microscopy is often used to identify these irradiation-induced morphologic changes. However, each fuel compact generally contains thousands of TRISO particles. Manually performing the work to get statistical information on these phenomena is cumbersome and subjective. Here, to reduce the subjectivity inherent in that process and to accelerate data analysis, we used convolutional neural networks (CNNs) to automatically segment cross-sectional images of microscopic TRISO layers. CNNs are a class of machine-learning algorithms specifically designed for processing structured grid data. They have gained popularity in recent years due to their remarkable performance in various computer vision tasks, including image classification, object detection, and image segmentation. In this research, we generated a large irradiated TRISO layer dataset with more than 2,000 microscopic images of cross-sectional TRISO particles and the corresponding annotated images. Based on these annotated images, we used different CNNs to automatically segment different TRISO layers. These CNNs include RU-Net (developed in this study), as well as three existing architectures: U-Net, Residual Network (ResNet), and Attention U-Net. The preliminary results show that the model based on RU-Net performs best in terms of Intersection over Union (IoU). Using CNN models, we can expedite the analysis of TRISO particle cross sections, significantly reducing the manual labor involved and improving the objectivity of the segmentation results.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗