SEARCH · Search NASA
Results for “microstructure model”
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.
Bayesian Entropy Neural Networks for physics-aware prediction
This article addresses the need for deep learning models to integrate well-defined constraints into their outputs, driven by their application in surrogate models, learning with limited data and partial information, and scenarios requiring flexible model behavior to incorporate non-data sample information. We introduce Bayesian Entropy Neural Networks (BENN), a framework grounded in Maximum Entropy (MaxEnt) principles, designed to impose constraints on Bayesian Neural Network (BNN) predictions. BENN is capable of constraining not only the predicted values but also their derivatives and variances, ensuring a more robust and reliable model output. To achieve simultaneous uncertainty quantification and constraint satisfaction, we employ the method of multipliers approach. This allows for the concurrent estimation of neural network parameters and the Lagrangian multipliers associated with the constraints. Our experiments, spanning diverse applications such as beam deflection modeling and microstructure generation, demonstrate the effectiveness of BENN. The results highlight significant improvements over traditional BNNs and showcase competitive performance relative to contemporary constrained deep learning methods.
SHORT-LIVED RADIO BURSTS FROM THE CRAB PULSAR
Our high-time-resolution observations reveal that individual main pulses from the Crab pulsar contain one or more short-lived microbursts. Both the energy and duration of bursts measured above 1 GHz can vary dramatically in less than a millisecond. These fluctuations are too rapid to be caused by propagation through turbulence in the Crab Nebula or in the interstellar medium; they must be intrinsic to the radio emission process in the pulsar. The mean duration of a burst varies with frequency as {nu}{sup -2}, significantly different from the broadening caused by interstellar scattering. We compare the properties of the bursts to some simple models of microstructure in the radio emission region.
Improving Structural Integrity of Direct Laser-Deposited Ni – Co – Cr – Al – Y Superalloys by Alloy Modification
Abstract Ni- and Co-based M–Cr–Al–Y (M = Ni and/or Co), dual phase β and γ/γ′, β—Ni–Al, γ—Ni(Co, Cr), and γ′—Ni3–Al superalloys exhibit several beneficial high-temperature (>1000 °C) (2012 °F) strength and corrosion resistance properties. Our study investigates the feasibility of fabricating a commercially available Ni–Co–Cr–AlY alloy (tradename “Amdry 386”) using laser directed energy deposition (laser-DED). A significant challenge for laser fabrication of bulk Amdry 386 (A386) structures stems from the presence of relatively higher amounts of the β phase than γ/γ′ phases. While prior studies report laser fabrication of these materials in coating and cladding configurations, bulk structures have not been realized. To investigate these challenges, this study was developed to systematically evaluate the effects of modifying the A386 alloy by adding 10, 20, and 30% by weight of a Ni–20%Cr (Ni–Cr) alloy (mainly γ/γ′). Laser-DED-fabricated A386 coupons exhibited a metastable, rapidly solidified β-rich microstructure typical to laser processes. The A386 + Ni–Cr mixtures were processed using laser-DED, and small coupons from each composition were evaluated using SEM, XRD, and microhardness to characterize the as-processed microstructure. Thermodynamic modeling was performed to determine the phase evolution as a function of the alloy composition. The dominating phase switches from β to γ/γ’ between the A386 and A386 + 10% Ni–Cr mixtures, but the increase in structural integrity is not observed until the Ni–Cr additions reach a minimum of 20%. Our results show that the alloy chemistry can be successfully modified to improve the structural integrity of laser-processed structures.
Optimization of SOFC Electrode Microstructures with Long Term Performance Modeling and Machine Learning
17th International Symposium on Solid Oxide Fuel Cells (SOFC-XVII), Virtual, July 18-23, 2021
Modeling the Effects of Electrode Microstructural Heterogeneities on Li-Ion Battery Performance and Lifetime
Not Available
muMatScale
Code for modeling solidification microstructure based on the Cellular Automata method for binary/pseudo-binary alloys. This code is derived from the open-source software uMatIC developed at the Imperial College London https://www.imperial.ac.uk/engineering-alloys/research/software
Analyzing the effect of pressure on the properties of point defects in $\gamma\mathrm{U–Mo}$ through atomistic simulations
Uranium–molybdenum (U–Mo) alloys in monolithic fuel foil are the primary candidate for the conversion of high-performance research reactors in the USA. Monolithic fuel is utilized in a plate-type design with a zirconium diffusion barrier and aluminum cladding. These fuel types are unique in that they contain no plenum for the release of fission gases, which, in conjunction with the aluminum cladding, can lead to large stress states within the fuel. The nature of how fundamental processes of radiation damage, including the evolution of point defects, under such stresses occur is unknown. In this work, we present molecular dynamics simulations of the formation energy of point defects under applied stress. We report this work will allow for the implementation of stress-dependent microstructural evolution models of nuclear fuels, including those for both fission gas bubble growth and creep, which are critical to ensure the stable and predictable behavior of research reactor fuels.
Rolling Process Modeling Report. Finite-Element Model Validation and Parametric Study on various Rolling Process parameters
Pacific Northwest National Laboratory (PNNL) has been investigating manufacturing processes for the uranium-10% molybdenum alloy plate-type fuel for high-performance research reactors in the United States. This work supports the U.S. Department of Energy National Nuclear Security Administration’s Office of Material Management and Minimization Reactor Conversion Program. This report documents modeling results of PNNL’s efforts to perform finite-element simulations to predict roll-separating forces for various rolling mill geometries for PNNL, Babcock & Wilcox Co., Y-12 National Security Complex, Los Alamos National Laboratory, and Idaho National Laboratory. The model developed and presented in a previous report has been subjected to further validation study using new sets of experimental data generated from a rolling mill at PNNL. Simulation results of both hot rolling and cold rolling of uranium-10% molybdenum coupons have been compared with experimental results. The model was used to predict roll-separating forces at different temperatures and reductions for five rolling mills within the National Nuclear Security Administration Fuel Fabrication Capability project. This report also presents initial results of a finite-element model microstructure-based approach to study the surface roughness at the interface between zirconium and uranium-10% molybdenum.
Development of a Multi-physics Capability for Predicting Residual Stress in a GTS Reservoir.
This report documents completion of a Level 2 Milestone on the development of a multi-physics capability to predict the evolving material state through the manufacturing process of a Gas Transfer Systems (GTS) reservoir. We present details on new developments and capability improvements that address the following completion criteria: (i) validation of a microstructure evolution model, including recrystallization and strain aging, (ii) demonstration of the capability to remesh, map and transfer material state (internal state variables) and residual stress from forging to machining to welding processes, and (iii) formal V&V characterization and quantification of uncertainties of material parameters and manufacturing process parameters on residual stress.
USAMP Low-Cost Mg Sheet Component Development and Demonstration Project
The overarching objective of this USAMP project was to develop and demonstrate door panels made from magnesium (Mg) sheet with a cost penalty over conventional steel stampings of no more than $\$5.50$/kg saved. The technical approach integrated experiments with advanced computational tools based on Integrated Computational Materials Engineering (ICME) methods to develop new alloy chemistries and their thermomechanical processing that promise improved formability and lower forming temperatures. A penultimate task before finally forming the stampings was to incorporate actual microstructure into models that would enable formability simulations. This approach would, for the first time, account for individual magnesium grains moving in an anisotropic fashion unlike that for aluminum or steel that have isotropic properties upon which the current simulation tools are based. In separate activities, new coatings and lubricants to facilitate forming and improved corrosion protection and joining strategies, were developed to ensure that the door could be produced with stated product requirements. A technical cost model, which included parts production, assembly, and paint for a door specifically designed for Mg sheet, showed the cost penalty to be between $\$4.26$ to $\$6.60$/kg saved, which enveloped the project’s cost targets. The cost of the coated Mg sheet was identified as the key driver for the cost penalty. The mass of the Mg-intensive door was 7.9 kg, which was 54% less than the baseline steel door.
Low-Cost Magnesium Alloy Sheet Component Development and Demonstration Project
The overarching objective of this USAMP project was to develop and demonstrate door panels made from magnesium (Mg) sheet with a cost penalty over conventional steel stampings of no more than $5.50/kg saved. The technical approach integrated experiments with advanced computational tools based on Integrated Computational Materials Engineering (ICME) methods to develop new alloy chemistries and their thermomechanical processing that promise improved formability and lower forming temperatures. A penultimate task before finally forming the stampings was to incorporate actual microstructure into models that would enable formability simulations. This approach would, for the first time, account for individual magnesium grains moving in an anisotropic fashion unlike that for aluminum or steel that have isotropic properties upon which the current simulation tools are based. In separate activities, new coatings and lubricants to facilitate forming and improved corrosion protection and joining strategies, were developed to ensure that the door could be produced with stated product requirements. A technical cost model, which included parts production, assembly, and paint for a door specifically designed for Mg sheet, showed the cost penalty to be between 4.26 USD to 6.60 USD/kg saved, which enveloped the project’s cost targets. The cost of the coated Mg sheet was identified as the key driver for the cost penalty. The mass of the Mg-intensive door was 7.9 kg, which was 54% less than the baseline steel door.
Phase-field modeling of thin-film microstructures: A theoretical platform for understanding and designing functional nanostructured materials.
Abstract not provided.
Machine learning constitutive models of inelastic materials with microstructure.
Abstract not provided.
Machine learning constitutive models of inelastic materials with microstructure.
Abstract not provided.
Microstructure-Aware Phase-Field Models of Alloy Oxidation
Explore the source record for details and available documents.
Machine learning constitutive models of inelastic materials with microstructure
Explore the source record for details and available documents.