Search NASA⌕ Search

SEARCH · Search NASA

Results for “embedded methods”

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 307 records · Page 17
All sourcesNASA NTRSDOE OSTI
NASA NTRS2005

Closed Loop Active Flow Separation Detection and Control in a Multistage Compressor

Active closed loop flow control was successfully demonstrated on a full annulus of stator vanes in a low speed axial compressor. Two independent methods of detecting separated flow conditions on the vane suction surface were developed. The first technique detects changes in static pressure along the vane suction surface, while the second method monitors variation in the potential field of the downstream rotor. Both methods may feasibly be used in future engines employing embedded flow control technology. In response to the detection of separated conditions, injection along the suction surface of each vane was used. Injected mass flow on the suction surface of stator vanes is known to reduce separation and the resulting limitation on static pressure rise due to lowered diffusion in the vane passage. A control algorithm was developed which provided a proportional response of the injected mass flow to the degree of separation, thereby minimizing the performance penalty on the compressor system.

Bright, Michelle M.↗
DOE OSTI2025

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

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

Latent representations↗
NASA NTRS1982

Inverse transonic airfoil design methods including boundary layer and viscous interaction effects

A body-fitted grid embedment technique applicable to inviscid transonic airfoil flow field analysis was developed and verified through a series of tests. Test cases used to verify the technique show that the accuracy of the solution was increased by grid embedding. This enhancement of the solution is especially true when small supercritical zones occur which cannot be adequately described using the main grid only. Results obtained with the SKANFP full potential program are considered with regard to the massive separated flow and high lift and the undesirable unrealistic 'bump' in the vicinity of the separation point due to a mismatch between the unseparated and separated pressure distributions. Techniques used to eliminate this feature are discussed.

Carlson, L. A.↗
NASA NTRS1984

Embedded shear layer computations for increased drag reduction

One of the most promising methods of minimizing drag is the reduction of skin friction by injection of low momentum fluid into the near-wall region of turbulent boundary layer flows. This method could be made more effective by limiting the spread rate of the resulting mixing region. In order to achieve a better understanding of how this goal might be achieved, numerical investigations of the relevant fluid dynamic processes governing these regions have been conducted. A compact finite-difference algorithm has been applied to the complete form of the governing conservation equations for a two-dimensional laminar mixing layer. The ability of this computational approach to model successfully the formation and interaction of the large scale vortical structures which dominate such flow fields is verified in the present study. Parameters which affect the spread rate of the mixing region are also identified. In addition, the relative importance of viscous and momentum transport effects in the vortex interactions is determined.

Gatski, T. B.↗
NASA NTRS1972

A finite-difference method for transonic airfoil design.

This paper describes an inverse method for designing transonic airfoil sections or for modifying existing profiles. Mixed finite-difference procedures are applied to the equations of transonic small disturbance theory to determine the airfoil shape corresponding to a given surface pressure distribution. The equations are solved for the velocity components in the physical domain and flows with embedded shock waves can be calculated. To facilitate airfoil design, the method allows alternating between inverse and direct calculations to obtain a profile shape that satisfies given geometric constraints. Examples are shown of the application of the technique to improve the performance of several lifting airfoil sections. The extension of the method to three dimensions for designing supercritical wings is also indicated.

Steger, J. L.↗
NASA NTRS1998

Gel stretch method: a new method to measure constitutive properties of cardiac muscle cells

Diastolic dysfunction is an important cause of congestive heart failure; however, the basic mechanisms causing diastolic congestive heart failure are not fully understood, especially the role of the cardiac muscle cell, or cardiocyte, in this process. Before the role of the cardiocyte in this pathophysiology can be defined, methods for measuring cardiocyte constitutive properties must be developed and validated. Thus this study was designed to evaluate a new method to characterize cardiocyte constitutive properties, the gel stretch method. Cardiocytes were isolated enzymatically from normal feline hearts and embedded in a 2% agarose gel containing HEPES-Krebs buffer and laminin. This gel was cast in a shape that allowed it to be placed in a stretching device. The ends of the gel were held between a movable roller and fixed plates that acted as mandibles. Distance between the right and left mandibles was increased using a stepper motor system. The force applied to the gel was measured by a force transducer. The resultant cardiocyte strain was determined by imaging the cells with a microscope, capturing the images with a CCD camera, and measuring cardiocyte and sarcomere length changes. Cardiocyte stress was characterized with a finite-element method. These measurements of cardiocyte stress and strain were used to determine cardiocyte stiffness. Two variables affecting cardiocyte stiffness were measured, the passive elastic spring and viscous damping. The passive spring was assessed by increasing the force on the gel at 1 g/min, modeling the resultant stress vs. strain relationship as an exponential [sigma = A/k(ekepsilon - 1)]. In normal cardiocytes, A = 23.0 kN/m2 and k = 16. Viscous damping was assessed by examining the loop area between the stress vs. strain relationship during 1 g/min increases and decreases in force. Normal cardiocytes had a finite loop area = 1.39 kN/m2, indicating the presence of viscous damping. Thus the gel stretch method provided accurate measurements of cardiocyte constitutive properties. These measurements have allowed the first quantitative assessment of passive elastic spring properties and viscous damping in normal mammalian cardiocytes.

Non-NASA Center↗
NASA NTRS2015

Computational Modeling and Experimental Characterization of Martensitic Transformations in Nicoal for Self-Sensing Materials

Fundamental changes to aero-vehicle management require the utilization of automated health monitoring of vehicle structural components. A novel method is the use of self-sensing materials, which contain embedded sensory particles (SP). SPs are micron-sized pieces of shape-memory alloy that undergo transformation when the local strain reaches a prescribed threshold. The transformation is a result of a spontaneous rearrangement of the atoms in the crystal lattice under intensified stress near damaged locations, generating acoustic waves of a specific spectrum that can be detected by a suitably placed sensor. The sensitivity of the method depends on the strength of the emitted signal and its propagation through the material. To study the transition behavior of the sensory particle inside a metal matrix under load, a simulation approach based on a coupled atomistic-continuum model is used. The simulation results indicate a strong dependence of the particle's pseudoelastic response on its crystallographic orientation with respect to the loading direction and suggest possible ways of optimizing particle sensitivity. The technology of embedded sensory particles will serve as the key element in an autonomous structural health monitoring system that will constantly monitor for damage initiation in service, which will enable quick detection of unforeseen damage initiation in real-time and during onground inspections.

Wallace, T. A.↗
NASA NTRS1992

A boundary element alternating method for two-dimensional mixed-mode fracture problems

A boundary element alternating method, denoted herein as BEAM, is presented for two dimensional fracture problems. This is an iterative method which alternates between two solutions. An analytical solution for arbitrary polynomial normal and tangential pressure distributions applied to the crack faces of an embedded crack in an infinite plate is used as the fundamental solution in the alternating method. A boundary element method for an uncracked finite plate is the second solution. For problems of edge cracks a technique of utilizing finite elements with BEAM is presented to overcome the inherent singularity in boundary element stress calculation near the boundaries. Several computational aspects that make the algorithm efficient are presented. Finally, the BEAM is applied to a variety of two dimensional crack problems with different configurations and loadings to assess the validity of the method. The method gives accurate stress intensity factors with minimal computing effort.

Raju, I. S.↗
DOE OSTI2025

Explaining word embeddings with perfect fidelity: a case study in predicting research impact

The best-performing approaches for scholarly document quality prediction are based on embedding models. In addition to their performance when used in classifiers, embedding models can also provide predictions even for words that were not contained in the labelled training data for the classification model, which is important in the context of the ever-evolving research terminology. Although model-agnostic explanation methods, such as Local interpretable model-agnostic explanations, can be applied to explain machine learning classifiers trained on embedding models, these produce results with questionable correspondence to the model. We introduce a new feature importance method, Self-Model Entities Rated (SMER), for logistic regression-based classification models trained on word embeddings. We show that SMER has theoretically perfect fidelity with the explained model, as the average of logits of SMER scores for individual words (SMER explanation) exactly corresponds to the logit of the prediction of the explained model. Quantitative and qualitative evaluation is performed through five diverse experiments conducted on 50,000 research articles (papers) from the CORD-19 corpus. In conclusion, through an AOPC curve analysis, we experimentally demonstrate that SMER produces better explanations than LIME, SHAP and global tree surrogates.

Coarse-grained models↗
NASA NTRS2023

Oxidation of Silicon Particles Suspended in Mullite Matrix With Dual-Mode Vacancy Diffusion of Oxygen

Mullite bond coats filled with silicon particles have recently been proposed for the next generation of environmental barrier coatings (EBCs). The sacrificial oxidation of the silicon particles eliminates some of the environmental durability issues associated with traditional EBC designs. To investigate this novel design approach, a mathematical formulation and resulting numerical method have been developed for simulating the oxidation of silicon particles embedded in a mullite matrix. The focus is on the oxidation of the silicon in a dry oxygen environment where the oxidant is presumably oxygen. Oxygen transport from the exterior environment through the mullite is assumed to occur via two lattice defect diffusion mechanisms. As a result, the rate of oxygen transport is dictated by the self-diffusivity of the lattice defects. The numerical method is implemented using the COMSOL Multiphysics® program (COMSOL, Inc.). The method is applied to simulate the oxidation of silicon particles in a mullite pellet in recent oxidation experiments. Numerical results compare favorably with the oxidation behavior observed in the oxidation experiments. The numerical method is also used to investigate the effects of various features of the mullite-silicon particle system on the oxidation behavior.

silicon oxidation↗
DOE OSTI2026

F-Hash: Feature-Based Hash Design for Time-Varying Volume Visualization via Multi-Resolution Tesseract Encoding

Interactive time-varying volume visualization is challenging due to its complex spatiotemporal features and sheer size of the dataset. Recent works transform the original discrete time-varying volumetric data into continuous Implicit Neural Representations (INR) to address the issues of compression, rendering, and super-resolution in both spatial and temporal domains. However, training the INR takes a long time to converge, especially when handling large-scale time-varying volumetric datasets. In this work, we proposed F-Hash, a novel feature-based multi-resolution Tesseract encoding architecture to greatly enhance the convergence speed compared with existing input encoding methods for modeling time-varying volumetric data. The proposed design incorporates multi-level collision-free hash functions that map dynamic 4D multi-resolution embedding grids without bucket waste, achieving high encoding capacity with compact encoding parameters. Our encoding method is agnostic to time-varying feature detection methods, making it a unified encoding solution for feature tracking and evolution visualization. Experiments show the F-Hash achieves state-of-the-art convergence speed in training various time-varying volumetric datasets for diverse features. We also proposed an adaptive ray marching algorithm to optimize the sample streaming for faster rendering of the time-varying neural representation.

deep learning↗
NASA NTRS2002

Ceramic Inclusions In Powder Metallurgy Disk Alloys: Characterization and Modeling

Powder metallurgy alloys are increasingly used in gas turbine engines, especially as the material chosen for turbine disks. Although powder metallurgy materials have many advantages over conventionally cast and wrought alloys (higher strength, higher temperature capability, etc.), they suffer from the rare occurrence of ceramic defects (inclusions) that arise from the powder atomization process. These inclusions can have potentially large detrimental effect on the durability of individual components. An inclusion in a high stress location can act as a site for premature crack initiation and thereby considerably reduce the fatigue life. Because these inclusions are exceedingly rare, they usually don't reveal themselves in the process of characterizing the material for a particular application (the cumulative volume of the test bars in a fatigue life characterization is typically on the order of a single actual component). Ceramic inclusions have, however, been found to be the root cause of a number of catastrophic engine failures. To investigate the effect of these inclusions in detail, we have undertaken a study where a known population of ceramic particles, whose composition and morphology are designed to mimic the 'natural' inclusions, are added to the precursor powder. Surface connected inclusions have been found to have a particularly large detrimental effect on fatigue life, therefore the volume of ceramic 'seeds' added is calculated to ensure that a minimum number will occur on the surface of the fatigue test bars. Because the ceramic inclusions are irregularly shaped and have a tendency to break up in the process of extrusion and forging, a method of calculating the probability of occurrence and expected intercepted surface and embedded cross-sectional areas were needed. We have developed a Monte Carlo simulation to determine the distributions of these parameters and have verified the simulated results with observations of ceramic inclusions found in macro slices from extrusions and forgings. The ultimate goal of this study will be to use probabilistic methods to determine the reliability detriment that can be attributed to these ceramic inclusions.

Bonacuse, Pete↗
NASA NTRS1982

A natural conservative flux difference splitting for the hyperbolic systems of gasdynamics

Gasdynamic hyperbolic systems are treated by a novel conservative flux difference splitting upwind method which is applicable to explicit or implicit and iterative or direct schemes, for marching in time or space with the Euler or compressible Navier-Stokes equations. The method is able to capture sharp shocks correctly by maintaining global conservation. The embedded characteristics formulation is cast in the volumetric internal energy primitive variable, which is effective for real as well as perfect gases. Higher order upwind formulae are constructed from the distribution of pieces of the simple internal flux difference splitting, and the method reverts to first order at the appropriate points in order not to violate the domain of dependence by differencing across discontinuities. At insignificant computational overhead, switching is done by simple algebraic extensions of the truth functions that determine upwind direction for the first order scheme.

Lombard, C. K.↗
DOE OSTI2024

Formal Methods for Provably Secure Software and Firmware

This project addresses a gap observed in verifying the programming in embedded devices used in international arms control: namely verifying that embedded programming in an arms control device does exactly what it is supposed to do, no more and no less, every time without fail, and without disclosing unauthorized information accidentally or intentionally. In critical military, aerospace, and industrial safety systems this problem is sometimes addressed using formal methods (FM). This multi-year project seeks to identify formal methods toolsets useable in arms control regimes, with emphasis on applicability, ease of use, long term availability, and support.

formal methods, Arms Control Verification↗
NASA NTRS1993

surf3d: A 3-D finite-element program for the analysis of surface and corner cracks in solids subjected to mode-1 loadings

A computer program, surf3d, that uses the 3D finite-element method to calculate the stress-intensity factors for surface, corner, and embedded cracks in finite-thickness plates with and without circular holes, was developed. The cracks are assumed to be either elliptic or part eliptic in shape. The computer program uses eight-noded hexahedral elements to model the solid. The program uses a skyline storage and solver. The stress-intensity factors are evaluated using the force method, the crack-opening displacement method, and the 3-D virtual crack closure methods. In the manual the input to and the output of the surf3d program are described. This manual also demonstrates the use of the program and describes the calculation of the stress-intensity factors. Several examples with sample data files are included with the manual. To facilitate modeling of the user's crack configuration and loading, a companion program (a preprocessor program) that generates the data for the surf3d called gensurf was also developed. The gensurf program is a three dimensional mesh generator program that requires minimal input and that builds a complete data file for surf3d. The program surf3d is operational on Unix machines such as CRAY Y-MP, CRAY-2, and Convex C-220.

Raju, I. S.↗
DOE OSTI2025

Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties With Deep Learning Multi‐Member and Stochastic Parameterizations

Abstract Deep learning is a powerful tool to represent subgrid processes in climate models, but many application cases have so far used idealized settings and deterministic approaches. Here, we develop stochastic parameterizations with calibrated uncertainty quantification to learn subgrid convective and turbulent processes and surface radiative fluxes of a superparameterization embedded in an Earth System Model (ESM). We explore three methods to construct stochastic parameterizations: (a) a single Deep Neural Network (DNN) with Monte Carlo Dropout; (b) a multi‐member parameterization; and (c) a Variational Encoder Decoder with latent space perturbation. We show that the multi‐member parameterization improves the representation of convective processes, especially in the planetary boundary layer, compared to individual DNNs. The respective uncertainty quantification illustrates that methods (b) and (c) are advantageous compared to a dropout‐based DNN parameterization regarding the spread of convective processes. Hybrid simulations with our best‐performing multi‐member parameterizations remained challenging and crash within the first days. Therefore, we develop a pragmatic partial coupling strategy relying on the superparameterization for condensate emulation. Partial coupling reduces the computational efficiency of hybrid Earth‐like simulations but enables model stability over 5 months with our multi‐member parameterizations. However, our hybrid simulations exhibit biases in thermodynamic fields and differences in precipitation patterns. Despite this, the multi‐member parameterizations enable improvements in reproducing tropical extreme precipitation compared to a traditional convection parameterization. Despite these challenges, our results indicate the potential of a new generation of multi‐member machine learning parameterizations leveraging uncertainty quantification to improve the representation of stochasticity of subgrid effects.

Behrens, Gunnar [Deutsches Zentrum für Luft‐ und R↗
DOE OSTI2024

Semantic Stealth: Crafting Covert Adversarial Patches for Sentiment Classifiers Using Large Language Models

Deep learning models have been shown to be vulnerable to adversarial attacks, in which perturbations to their inputs cause the model to produce incorrect predictions. As opposed to adversarial attacks in computer vision, where small changes introduced to pixel values can drastically alter a model's output while remaining imperceptible to humans, text-based attacks are difficult to conceal due to the discrete nature of tokens. Consequently, unconstrained gradient-based attacks often produce adversarial examples that lack semantic meaning, rendering them detectable through visual inspection or perplexity filters. In contrast to methods that rely on gradient-based optimization in the embedding space, we propose an approach that leverages a Large Language Model's ability to generate grammatically correct and semantically meaningful text to craft adversarial patches that seamlessly blend in with the original input text. These patches can be used to alter the behavior of a target model, such as a text classifier. Since our approach does not rely on gradient backpropagation, it only requires access to the target model's confidence scores, making it a grey-box attack. We demonstrate the feasibility of our approach using open-source LLMs, including Intel's Neural Chat, Llama2, and Mistral-Instruct, to generate adversarial patches capable of altering the predictions of a distilBERT model fine-tuned on the IMDB reviews dataset for sentiment classification.

Roa Carvajal, Maria↗
DOE OSTI2026

Towards a Verifiable Domain-Specific Language for Hardware-Accelerated Stencils

Defining a domain-specific language (DSL) that supports vector-calculus abstractions eases the porting of partial differential equation (PDE) solvers to specialized architectures. Sufficiently high-level abstractions empower users to express universal laws with sufficient generality that the laws must always hold true within their domain of validity. A broad class of PDE solvers employs stencil-based algorithms, the target domain of Berkeley Lab's stencil accelerator chip co-design project. First released as open-source in January 2026, the Formal software framework lays a foundation for defining an embedded DSL based on composable operators that implement mimetic numerical methods -- stencil algorithms that guarantee satisfaction of discrete versions of important vector calculus theorems. The Formal DSL will be the frontend to a new class of stencil-PDE accelerators developed jointly by LBNL, UHCL, and UC Berkeley through the DOE Competitive Portfolios for Computer Science Project. This offers the potential of an order of magnitude acceleration for this important category of computational methods to serve the DOE mission. Future work on the Formal DSL will facilitate software verification via type-safe templates that enable problem-specific correctness proofs relying upon generic function theory and carefully crafted unit tests.

Rouson, Damian↗