SEARCH · Engineering Papers
Results for “multiscale features”
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.
Multiscale embedded printing of engineered human tissue and organ equivalents
Creating tissue and organ equivalents with intricate architectures and multiscale functional feature sizes is the first step toward the reconstruction of transplantable human tissues and organs. Existing embedded ink writing approaches are limited by achievable feature sizes ranging from hundreds of microns to tens of millimeters, which hinders their ability to accurately duplicate structures found in various human tissues and organs. In this study, a multiscale embedded printing (MSEP) strategy is developed, in which a stimuli-responsive yield-stress fluid is applied to facilitate the printing process. A dynamic layer height control method is developed to print the cornea with a smooth surface on the order of microns, which can effectively overcome the layered morphology in conventional extrusion-based three-dimensional bioprinting methods. Since the support bath is sensitive to temperature change, it can be easily removed after printing by tuning the ambient temperature, which facilitates the fabrication of human eyeballs with optic nerves and aortic heart valves with overhanging leaflets on the order of a few millimeters. The thermosensitivity of the support bath also enables the reconstruction of the full-scale human heart on the order of tens of centimeters by on-demand adding support bath materials during printing. Here, the proposed MSEP demonstrates broader printable functional feature sizes ranging from microns to centimeters, providing a viable and reliable technical solution for tissue and organ printing in the future.
Prediction of hydration energies of adsorbates at Pt(111) and liquid water interfaces using machine learning
Aqueous phase heterogeneous catalysis is important to various industrial processes, including biomass conversion, Fischer–Tropsch synthesis, and electrocatalysis. Accurate calculation of solvation thermodynamic properties is essential for modeling the performance of catalysts for these processes. Explicit solvation methods employing multiscale modeling, e.g., involving density functional theory and molecular dynamics have emerged for this purpose. Although accurate, these methods are computationally intensive. This study introduces machine learning (ML) models to predict solvation thermodynamics for adsorbates on a Pt(111) surface, aiming to enhance computational efficiency without compromising accuracy. In particular, ML models are developed using a combination of molecular descriptors and fingerprints and trained on previously published water–adsorbate interaction energies, energies of solvation, and free energies of solvation of adsorbates bound to Pt(111). These models achieve root mean square error values of 0.09 eV for interaction energies, 0.04 eV for energies of solvation, and 0.06 eV for free energies of solvation, demonstrating accuracy within the standard error of multiscale modeling. Feature importance analysis reveals that hydrogen bonding, van der Waals interactions, and solvent density, together with the properties of the adsorbate, are critical factors influencing solvation thermodynamics. Furthermore, these findings suggest that ML models can provide rapid and reliable predictions of solvation properties. This approach not only reduces computational costs but also offers insights into the solvation characteristics of adsorbates at Pt(111)–water interfaces.
A Polymer Canvas with the Stiffness of the Bone Matrix to Study and Control Mesenchymal Stem Cell Response
Reproducing in vitro the complex multiscale physical features of human tissues creates novel biomedical opportunities and fundamental understanding of cell–environment interfaces and interactions. While stiffness has been recognized as a key driver of cell behavior, systematic studies on the role of stiffness have been limited to values in the KPa–MPa range, significantly below the stiffness of bone. Here, in this work, a platform enabling the tuning of the stiffness of a biocompatible polymeric interface up to values characteristic of human bone is reported, which are in the GPa range, by using extremely thin polymer films on glass and cross-linking the films using ultraviolet (UV) light irradiation. It is shown that a higher stiffness is related to better adhesion, proliferation, and osteogenic differentiation, and that it is possible to switch on/off cell attachment and growth by solely tuning the stiffness of the interface, without any surface chemistry or topography modification. Since the stiffness is tuned directly by UV irradiation, this platform is ideal for rapid and simple fabrication of stiffness patterns and gradients, thus representing an innovative tool for combinatorial studies of the synergistic effect of tissue environmental cues on cell behavior, and creates new opportunities for next-generation biosensors, single-cell patterning, and lab-on-a-chip devices.
Enabling probabilistic learning on manifolds through double diffusion maps
Here, we present a generative learning framework for probabilistic sampling that extends Probabilistic Learning on Manifolds (PLoM), which is designed to generate statistically consistent realizations of a random vector in a finite-dimensional Euclidean space, informed by a (representative) set of observations. In its original form, PLoM constructs a reduced-order probabilistic model by combining three main components: (a) kernel density estimation to approximate the underlying probability measure, (b) Diffusion Maps to characterize the manifold of the data, and (c) a reduced-order Itô Stochastic Differential Equation (ISDE) to sample from the learned distribution. However, its sampling dynamics are posed in the ambient space and the retained number of reduced coordinates is chosen by projection-reconstruction error. In practice, this often (i) requires more coordinates than the data’s intrinsic dimension to achieve stable sampling and (ii) lacks a smooth, basis-independent lifting back to the data domain; moreover, standard Diffusion Maps emphasize harmonic eigenfunctions and can miss non-harmonic latent structure. We address these limitations by decoupling geometry learning from sampling: a first Diffusion Maps pass identifies non-harmonic coordinates on which we formulate a full-order ISDE directly in the latent space, while Double Diffusion Maps captures multiscale geometric features and Geometric Harmonics (GH) learns a smooth lifting map to the ambient variables that is independent of the particular diffusion basis. This hybrid design preserves the system’s dynamical richness with a compact geometric representation and enables principled out-of-sample inference. The effectiveness and robustness of the proposed method are illustrated through two numerical studies: one based on data generated from two-dimensional Hermite polynomial functions and another based on high-fidelity simulations of a detonation wave in a reactive flow.
Anisotropic Coarse-Grained Model for Conjugated Polymers: Investigations into Solution Morphologies
The optoelectronic properties of conjugated polymers are dictated by the interplay of multiscale structural features, including intrachain dihedrals, interchain pi-pi stacking, and the complex mesoscale morphology. While much is known about the structures of polymers with isotropically interacting monomers, little is known about polymers with strongly anisotropic backbone monomers, a class of materials to which conjugated polymers belong. Fundamental understanding is further complicated when the semiflexible and molecularly heterogeneous nature of conjugated polymers is taken into account. We present an anisotropic coarse-grained (CG) model for conjugated polymers that incorporates key molecular features (monomer anisotropy, intermonomer dihedrals, and side chains). Finally, the model is employed to characterize the single-chain conformational properties of conjugated polymers, revealing a rich temperature-dependent conformational landscape. These studies provide a critical link between CG molecular descriptors and conformational ordering in conjugated polymers.
MS-POFT: multiscale phase-orientation guided feature transform for multi-modal image matching
Not Available
Large-Scale Classification of Urban Structural Units From Remote Sensing Imagery
Remote sensing in combination with deep learning has become instrumental for efficiently and accurately classifying land-use and land-cover across large geographic areas. These technologies have also been successful in characterizing urban environments in terms of their structural units, structure types, or morphological regions. In these approaches, an urban area is partitioned into regions that exhibit homogeneous physical characteristics. However, existing approaches are typically limited to a single city, use inconsistent typologies, and lack scalability and generalization capacity. In this article, we propose an urban structural units categorization scheme and demonstrate its utility by applying it to 13 cities. Inspired by the lack of scalability and generalization capacity in urban structural units mapping, we extend the reach of deep learning and conduct a set of classification experiments in all 13 cities. These experiments offer insights into the strengths and limitations of deep neural networks for classifying urban structural units over diverse geographic regions and on heterogeneous collections of satellite imagery. The efficacy of the proposed deep learning approach is compared to a baseline method of multiscale image features and support vector machines. Our validation on five cities shows that better performance is achieved with deep neural networks. Additionally, we evaluate the impact of input size, model depth, and spatial pyramid pooling to assess the generalization capacity of deep neural networks.
Modeling-Based Assessment of Deep Seismic Potential Induced by Geologic Carbon Storage
Induced seismicity is an inherent risk associated with geologic carbon storage (GCS) in deep rock formations that could contain undetected faults prone to failure. Modeling-based risk assessment has been implemented to quantify the potential of injection-induced seismicity, but typically simplified multiscale geologic features or neglected multiphysics coupled mechanisms because of the uncertainty in field data and computational cost of field-scale simulations, which may limit the reliable prediction of seismic hazard caused by industrial-scale CO 2 storage. The degree of lateral continuity of the stratigraphic interbedding below the reservoir and depth-dependent fault permeability can enhance or inhibit pore-pressure diffusion and corresponding poroelastic stressing along a basement fault. In conclusion, this study presents a rigorous modeling scheme with optimal geological and operational parameters needed to be considered in seismic monitoring and mitigation strategies for safe GCS.
Model-Based, Multiscale Self-Tuning Controller Developed for Active Combustion Control
New challenges concerning system health-monitoring and life-extending robust controls for the Ultra-Efficient Engine Technology Project, as well as other advanced engine and power system concepts at NASA and elsewhere, have renewed the control community s interest in smart, model-based methods. In particular, these challenges have further motivated efforts at the NASA Glenn Research Center to exploit the versatility and superiority of the dynamic features extraction of multiscale analysis for controls--such as with "wavelets" and "wavelet filter-banks.' The accomplishments reported herein pertain to the active suppression of combustion instabilities in liquid-fuel combustors via fuel modulation. The fundamentals and initial success of this innovation were reported for a unique demonstration of active combustion control (a research collaboration of NASA Glenn with Pratt & Whitney and the United Technologies Research Center, UTRC). This demonstration, conducted in 2002 at UTRC on the NASA single nozzle rig (SNR) combustor, was the first known suppression of high-frequency instability with a liquid-fueled combustor. The SNR is based on a high-powered military engine combustor that exhibited well-known instabilities.
Dissolvable Template Nanoimprint Lithography: A Facile and Versatile Nanoscale Replication Technique
Nanoimprinting lithography (NIL) is a next-generation nanofabrication method, capable of replicating nanostructures from original master surfaces. In this work, we develop highly scalable, simple, and nondestructive NIL using a dissolvable template. Termed dissolvable template nanoimprinting lithography (DT-NIL), our method utilizes an economic thermoplastic resin to fabricate nanoimprinting templates, which can be easily dissolved in simple organic solvents. Additionally, we used the DT-NIL method to replicate cicada wings which have surface nanofeatures of ~100 nm in height. The master, template, and replica surfaces showed a >~94% similarity based on the measured diameter and height of the nanofeatures. The versatility of DT-NIL was also demonstrated with the replication of re-entrant, multiscale, and hierarchical features on fly wings, as well as hard silicon wafer-based artificial nanostructures. The DT-NIL method can be performed under ambient conditions with inexpensive materials and equipment. Our work opens the door to opportunities for economical and high-throughput nanofabrication processes.
CEGANN: Crystal Edge Graph Attention Neural Network for multiscale classification of materials environment
Abstract We introduce Crystal Edge Graph Attention Neural Network (CEGANN) workflow that uses graph attention-based architecture to learn unique feature representations and perform classification of materials across multiple scales (from atomic to mesoscale) and diverse classes ranging from metals, oxides, non-metals to hierarchical materials such as zeolites and semi-ordered mesophases. CEGANN can classify based on a global, structure-level representation such as space group and dimensionality (e.g., bulk, 2D, clusters, etc.). Using representative materials such as polycrystals and zeolites, we demonstrate its transferability in performing local atom-level classification tasks, such as grain boundary identification and other heterointerfaces. CEGANN classifies in (thermal) noisy dynamical environments as demonstrated for representative zeolite nucleation and growth from an amorphous mixture. Finally, we use CEGANN to classify multicomponent systems with thermal noise and compositional diversity. Overall, our approach is material agnostic and allows for multiscale feature classification ranging from atomic-scale crystals to heterointerfaces to microscale grain boundaries.
Dynamic Bubbling Balanced Proactive CO 2 Capture and Reduction on a Triple-Phase Interface Nanoporous Electrocatalyst
The formation and preservation of the active phase of the catalysts at the triple-phase interface during CO 2 capture and reduction is essential for improving the conversion efficiency of CO 2 electroreduction toward value-added chemicals and fuels under operational conditions. Designing such ideal catalysts that can mitigate parasitic hydrogen generation and prevent active phase degradation during the CO 2 reduction reaction (CO 2 RR), however, remains a significant challenge. Herein, we developed an interfacial engineering strategy to build a new SnO x catalyst by invoking multiscale approaches. This catalyst features a hierarchically nanoporous structure coated with an organic F-monolayer that modifies the triple-phase interface in aqueous electrolytes, substantially reducing competing hydrogen generation (less than 5%) and enhancing CO 2 RR selectivity (~90%). This rationally designed triple-phase interface overcomes the issue of limited CO 2 solubility in aqueous electrolytes via proactive CO 2 capture and reduction. Concurrently, we utilized pulsed square-wave potentials to dynamically recover the active phase for the CO 2 RR to regulate the production of C1 products such as formate and carbon monoxide (CO). This protocol ensures profoundly enhanced CO 2 RR selectivity (~90%) compared with constant potential (~70%) applied at -0.8 V (V vs RHE). We further achieved a mechanistic understanding of the CO 2 capture and reduction processes under pulsed square-wave potentials via in situ Raman spectroscopy, thereby observing the potential-dependent intensity of Raman vibrational modes of the active phase and CO 2 RR intermediates. Finally, this work will inspire material design strategies by leveraging triple-phase interface engineering for emerging electrochemical processes, as technology moves toward electrification and decarbonization.
Building confidence in models for complex barrier systems for radionuclides
The modeling and simulation of the Cement-clay Interaction-Diffusion field (CI-D) experiment at the Mont Terri site in Switzerland presented here demonstrates that it is possible to capture the multiscale physical and chemical features of natural and engineered barrier systems for radionuclides. The simulations are successfully carried out with the newly developed CrunchODiTi high-performance computing software that accounts for multiple continua, including a continuum representing the electrical double layer (EDL) developed along negatively charged clay particles in clay rock. The simulation also accounts for both the complex three-dimensional (3D) geometry, expected as the norm in a geological waste repository, and the anisotropy of the geological formation. In addition, the high resolution of the model makes it possible to include "skin effects" developed at the interface between highly reactive materials, in this case between the high pH cement and the circumneutral but electrostatic Opalinus Clay. The successful history matching with the field experiment demonstrates that the distinct geochemical and physical properties of the cement and the Opalinus Clay in the CI-D experiment can be accounted for. Such analyses are essential for developing a defensible safety case for the underground storage of radioactive waste.
Tracking magnetogram proper motions by multiscale regularization
Long uninterrupted sequences of solar magnetograms from the global oscillations network group (GONG) network and from the solar and heliospheric observatory (SOHO) satellite will provide the opportunity to study the proper motions of magnetic features. The possible use of multiscale regularization, a scale-recursive estimation technique which begins with a prior model of how state variables and their statistical properties propagate over scale. Short magnetogram sequences are analyzed with the multiscale regularization algorithm as applied to optical flow. This algorithm is found to be efficient, provides results for all the spatial scales spanned by the data and provides error estimates for the solutions. It is found that the algorithm is less sensitive to evolutionary changes than correlation tracking.
Numerical validation of scaling laws for stratified turbulence
Recent theoretical progress using multiscale asymptotic analysis has revealed various possible regimes of stratified turbulence. Notably, buoyancy transport can either be dominated by advection or diffusion, depending on the effective Péclet number of the flow. Two types of asymptotic models have been proposed, which yield measurably different predictions for the characteristic vertical velocity and length scale of the turbulent eddies in both diffusive and non-diffusive regimes. The first, termed a ‘single-scale model’, is designed to describe flow structures having large horizontal and small vertical scales, while the second, termed a ‘multiscale model’, additionally incorporates flow features with small horizontal scales, and reduces to the single-scale model in their absence. By comparing predicted vertical velocity scaling laws with direct numerical simulation data, we show that the multiscale model correctly captures the properties of strongly stratified turbulence within regions dominated by small-scale isotropic motions, whose volume fraction decreases as the stratification increases. Meanwhile its single-scale reduction accurately describes the more orderly, layer-like, quiescent flow outside those regions.
Integrating Ultra-Coarse-Grained Protein Models into Accessible Workflows for Multiscale Molecular Dynamics
To capture protein conformational transitions using molecular dynamics (MD), several simulation resolutions covering different spatial and temporal scales are typically needed. All-atom (AA) simulations provide fine resolution, but are computationally infeasible for large systems over longer durations. Coarse-grained (CG) and ultra-coarse-grained (UCG) models have a lower resolution and computational cost while still being able to conserve essential protein features. Prior work on a Multiscale Machinelearned Modeling Infrastructure (MuMMI) combined both AA and CG simulations to study RAS-RAF protein interactions, leveraging CG models for longer time scales and using AA to investigate unusual conformations in greater detail. However, MuMMI is still resource-intensive, and this study aims to maximize exploration of the protein conformational space while reducing computational cost. In this paper, we build on prior work that integrates UCG models based on heterogeneous elastic network modeling (hENM) into the MuMMI workflow. We demonstrate that UCG models enable accurate sampling of protein conformations, focusing on simulating RAS-RAF protein interactions. Using higher-resolution CG Martini simulation data, we can automatically refine intramolecular interactions in UCG models. We present a scalable Python package that uses fluctuations observed in higher-resolution CG Martini simulations to estimate bond coefficients of the UCG model. We built novel machine learning-based backmapping methods to recover more detailed CG Martini structures from UCG structures, using diffusion models to learn the mapping between scales. Finally, we present UCG-mini-MuMMI, an accessible and less compute-intensive version of MuMMI as a resource for the scientific community. Incorporating UCG models into MD studies is applicable to a broad range of systems and proteins, and our study offers insights into the advantages and limitations of these methods.
Deep Learning without Global Optimization by Random Fourier Neural Networks
Here we introduce a new training algorithm for deep neural networks that utilize random complex exponential activation functions. Our approach employs a Markov chain Monte Carlo sampling procedure to iteratively train network layers, avoiding global and gradient-based optimization while maintaining error control. It consistently attains the theoretical approximation rate for residual networks with complex exponential activation functions, determined by network complexity. Additionally, it enables efficient learning of multiscale and high-frequency features, producing interpretable parameter distributions. Despite using sinusoidal basis functions, we do not observe Gibbs phenomena in approximating discontinuous target functions.