Search NASA⌕ Search

SEARCH · Search NASA

Results for “architectural”

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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Tailored Silicone Network Architecture for Ultimate Mechanical Reinforcement

Hydrosilylation cured silicone elastomers are subject to reaction inefficiency, leading to incomplete and non-uniform crosslink networks, restricting the potential of mechanical reinforcement. This work investigates pre-synthesized, functional PDMS architectures as additives to improve ultimate mechanical performance relative to conventional single-step curing. Three custom, functional structures were prepared: a partially crosslinked PDMS scaffold (Structure A), a bottle-brush PDMS (Structure B), and a star-shaped PDMS derived from an MQ resin (Structure C). Rheological characterization was used to identify the ultimate design space and proper stoichiometric ratio for Structure A, and confirm successful formation of all structures for suitable incorporation into a base silicone formulation at 30wt%. Mechanical tests indicated that all three structures increased in ultimate tensile strength relative to their single-step counterparts, with Structure A providing additional improvements to toughness (432 vs. 258 kJ/m3) and ultimate elongation (158 vs. 115%). Furthermore, Structure B remained very soft in the unfilled state, while Structure C provided hardness (23 vs. 18 Shore A) and stiffness (780 vs. 420 kPa Young’s modulus) increases. In silica filled systems, Structure A retained increased strength but reduced elongation, while Structure B indicated strong reinforcement in terms of strength, toughness, and stiffness. Thermal analysis on the cure profiles of these materials suggested that pre-formation of network architectures enable a more complete reaction than a single-step process (15.9 vs. 15.1 J/g). Ultimately, these results indicate that tailoring PDMS architecture before the final cure can improve ultimate mechanical properties via improved network development in silicone elastomers. Furthermore, this work offers a promising strategy for designing higher-performance, more tunable silicone formulations.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Distributed Grid-Sensing Service Layer: A System Architecture Blueprint

In this report, a scalable, decentralized, sensor-oriented communication blueprint is introduced. The proposed architecture has been specifically designed to support the mass deployment of sensors and also enable data consumers to access the underlying data streams using a distributed systems approach. With the proposed approach, sensors and data clients can be decoupled from the physical communication infrastructure and migrated into a modern, software-defined infrastructure ecosystem that can be configured to suit the end application’s requirements. In its current iteration, this blueprint defines the interactions, processes, and expected outcomes that each individual component within the architecture must support the overall ecosystem’s needs. Within this work, such an assembly of services is called the data bus; it represents the heart of the distributed grid sensing services architecture, a multiyear project that aims to design, implement, and test a reference sensor-data dissemination network that can support the needs of the extended grid state. We expect this blueprint to guide our future implementation efforts

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting

Advancing the capabilities of earthquake nowcasting, the real-time forecasting of seismic activities, remains crucial for reducing casualties. This multifaceted challenge has recently gained attention within the deep learning domain, facilitated by the availability of extensive earthquake datasets. Despite significant advancements, the existing literature on earthquake nowcasting lacks comprehensive evaluations of pre-trained foundation models and modern deep learning architectures; each focuses on a different aspect of data, such as spatial relationships, temporal patterns, and multi-scale dependencies. This paper addresses the mentioned gap by analyzing different architectures and introducing two innovative approaches called Multi Foundation Quake and GNNCoder. We formulate earthquake nowcasting as a time series forecasting problem for the next 14 days within 0.1-degree spatial bins in Southern California. Earthquake time series are generated using the logarithm energy released by quakes, spanning 1986 to 2024. Our comprehensive evaluations demonstrate that our introduced models outperform other custom architectures by effectively capturing temporal-spatial relationships inherent in seismic data. The performance of existing foundation models varies significantly based on the pre-training datasets, emphasizing the need for careful dataset selection. However, we introduce a novel method, Multi Foundation Quake, that achieves the best overall performance by combining a bespoke pattern with Foundation model results handled as auxiliary streams.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Enabling Parallel Performance and Portability of Solid Mechanics Simulations Across CPU and GPU Architectures

Efficiently simulating solid mechanics is vital across various engineering applications. As constitutive models grow more complex and simulations scale up in size, harnessing the capabilities of modern computer architectures has become essential for achieving timely results. This paper presents advancements in running parallel simulations of solid mechanics on multi-core CPUs and GPUs using a single-code implementation. This portability is made possible by the C++ matrix and array (MATAR) library, which interfaces with the C++ Kokkos library, enabling the selection of fine-grained parallelism backends (e.g., CUDA, HIP, OpenMP, pthreads, etc.) at compile time. MATAR simplifies the transition from Fortran to C++ and Kokkos, making it easier to modernize legacy solid mechanics codes. We applied this approach to modernize a suite of constitutive models and to demonstrate substantial performance improvements across different computer architectures. This paper includes comparative performance studies using multi-core CPUs along with AMD and NVIDIA GPUs. Results are presented using a hypoelastic–plastic model, a crystal plasticity model, and the viscoplastic self-consistent generalized material model (VPSC-GMM). The results underscore the potential of using the MATAR library and modern computer architectures to accelerate solid mechanics simulations.

Morgan, Nathaniel (ORCID:0000000276118449)↗
DOE OSTI2020

Vertically Aligned Single-Crystalline CoFe 2 O 4 Nanobrush Architectures with High Magnetization and Tailored Magnetic Anisotropy

Micrometer-tall vertically aligned single-crystalline CoFe 2 O 4 nanobrush architectures with extraordinarily large aspect ratio have been achieved by the precise control of a kinetic and thermodynamic non-equilibrium pulsed laser epitaxy process. Direct observations by scanning transmission electron microscopy reveal that the nanobrush crystal is mostly defect-free by nature, and epitaxially connected to the substrate through a continuous 2D interface layer. In contrast, periodic dislocations and lattice defects such as anti-phase boundaries and twin boundaries are frequently observed in the 2D interface layer, suggesting that interface misfit strain relaxation under a non-equilibrium growth condition plays a critical role in the self-assembly of such artificial architectures. Further, magnetic property measurements have found that the nanobrushes exhibit a saturation magnetization value of 6.16 μB/f.u., which is much higher than the bulk value. The discovery not only enables insights into an effective route for fabricating unconventional high-quality nanostructures, but also demonstrates a novel magnetic architecture with potential applications in nanomagnetic devices.

36 MATERIALS SCIENCE↗
DOE OSTI2021

MicroPulse DIAL (MPD) – a diode-laser-based lidar architecture for quantitative atmospheric profiling

Abstract. Continuous water vapor and temperature profiles are critically needed for improved understanding of the lower atmosphere and potential advances in weather forecasting skill. Ground-based, national-scale profiling networks are part of a suite of instruments to provide such observations; however, the technological method must be cost-effective and quantitative. We have been developing an active remote sensing technology based on a diode-laser-based lidar technology to address this observational need. Narrowband, high-spectral-fidelity diode lasers enable accurate and calibration-free measurements requiring a minimal set of assumptions based on direct absorption (Beer–Lambert law) and a ratio of two signals. These well-proven quantitative methods are known as differential absorption lidar (DIAL) and high-spectral-resolution lidar (HSRL). This diode-laser-based architecture, characterized by less powerful laser transmitters than those historically used for atmospheric studies, can be made eye-safe and robust. Nevertheless, it also requires solar background suppression techniques such as narrow-field-of-view receivers with an ultra-narrow bandpass to observe individual photons backscattered from the atmosphere. We discuss this diode-laser-based lidar architecture's latest generation and analyze how it addresses a national-scale profiling network's need to provide continuous thermodynamic observations. The work presented focuses on general architecture changes that pertain to both the water vapor and the temperature profiling capabilities of the MicroPulse DIAL (MPD). However, the specific subcomponent testing and instrument validation presented are for the water vapor measurements only. A fiber-coupled seed laser transmitter optimization is performed and shown to meet all of the requirements for the DIAL technique. Further improvements – such as a fiber-coupled near-range receiver, the ability to perform quality control via automatic receiver scanning, advanced multi-channel scalar capabilities, and advanced processing techniques – are discussed. These new developments increase narrowband DIAL technology readiness and are shown to allow higher-quality water vapor measurements closer to the surface via preliminary intercomparisons within the MPD network itself and with radiosondes.

Spuler, Scott M.↗
DOE OSTI2021

Architecture and Performance of Perlmutter’s 35 PB ClusterStor E1000 All-Flash File System

NERSC's newest system, Perlmutter, features a 35 PB all-flash Lustre file system built on HPE Cray ClusterStor E1000. We present its architecture, early performance figures, and performance considerations unique to this architecture. We demonstrate the performance of E1000 OSSes through low-level Lustre tests that achieve over 90% of the theoretical bandwidth of the SSDs at the OST and LNet levels. We also show end-to-end performance for both traditional dimensions of I/O performance (peak bulk-synchronous bandwidth) and non-optimal workloads endemic to production computing (small, incoherent I/Os at random offsets) and compare them to NERSC's previous system, Cori, to illustrate that Perlmutter achieves the performance of a burst buffer and the resilience of a scratch file system. Finally, we discuss performance considerations unique to all-flash Lustre and present ways in which users and HPC facilities can adjust their I/O patterns and operations to make optimal use of such architectures.

Lockwood, Glenn K↗
DOE OSTI2023

Surrogate Model Based Optimization for Finding Robust Deep Learning Model Architectures

Deep Learning (DL) models are increasingly used throughout the sciences. However, their performance and usefulness depend greatly on their architecture which is defined by hyperparameters such as the number of nodes, layers, the learning rate, etc. Tuning these hyperparameters is time-consuming because evaluating their performance requires a lengthy training step. Stochastic optimizers used in training lead to performance variability and potentially prediction reliability issues. In this talk, we will describe an automated optimization method based on surrogate models and active learning strategies for tuning DL model architectures. We take into account the prediction variability with the goal to identify architectures that make reliable and robust predictions. We demonstrate our developments on an application arising in particle physics.

deep learning↗
DOE OSTI2024

Scaling and performance portability of the particle-in-cell scheme for plasma physics applications through mini-apps targeting exascale architectures

We perform a scaling and performance portability study of the particle-in-cell scheme for plasma physics applications through a set of mini-apps we name "Alpine", which can make use of exascale computing capabilities. The mini-apps are based on Independent Parallel Particle Layer, a framework that is designed around performance portable and dimension independent particles and fields. We benchmark the simulations with varying parameters such as grid resolutions (5123 to 20483) and number of simulation particles (109 to 1011) with the following mini-apps: weak and strong Landau damping, bump-on-tail and two-stream instabilities, and the dynamics of an electron bunch in a charge-neutral Penning trap. We show strong and weak scaling and analyze the performance of different components on several pre-exascale architectures such as Piz-Daint, Cori, Summit and Perlmutter. While the scaling and portability study helps identify the performance critical components of the particle-in-cell scheme in the current state-of-the-art computing architectures, the mini-apps by themselves can be used to develop new algorithms and optimize their high performance implementations targeting exascale architectures.

Muralikrishnan, Sriramkrishnan↗
DOE OSTI2026

Multi-Agent Control Planes for Quantum Networks: A Scalable Architecture for Autonomous Quantum Internet Management

Quantum networks are expected to enable distributed quantum computing, secure communication, and global entanglement distribution. However, operating such networks presents significant challenges, including stochastic quantum processes, fragile entanglement resources, dynamic topology, and cross-layer control requirements. Current quantum network control architectures largely rely on centralized or hierarchical controllers inspired by classical software-defined networking (SDN). While effective for small testbeds, these approaches face scalability, latency, and reliability limitations as quantum networks grow. This paper proposes a multi-agent control plane architecture for quantum networks. In this design, intelligent software agents operate at quantum nodes, repeaters, and orchestration layers, collectively managing entanglement generation, routing, purification, and scheduling. The distributed intelligence of the agent system allows the network to adapt dynamically to quantum hardware variability and environmental noise. We argue that multi-agent systems provide significant advantages over centralized control approaches, including scalability, resilience, local autonomy, and real-time adaptation. The paper discusses architectural design principles, agent coordination mechanisms, and research challenges in deploying multi-agent control planes for the emerging quantum Internet.

Alnajjar, Anees [ORNL] (ORCID:0000000237101601)↗
DOE OSTI2026

Architectural scaling tradeoffs in modular 3D bosonic quantum processors

We propose a modular three-dimensional bosonic quantum processor built from repeatable coupled-cavity modules linked by configurable interconnect networks. Using hardware-motivated graph-theoretic measures, we compare nearest-neighbor, hub-based, and hybrid architectures in terms of interconnect count, communication distance, resource concentration, and implementation complexity. Rather than identifying a universally optimal topology, our analysis shows how these architectures redistribute the costs of scaling, including wiring and port requirements, nonlocal communication distance, exposure to shared resources, routing bottlenecks, and scheduling overhead. Case studies of a \(3\times3\) processor and a larger hierarchical architecture further distinguish finite-size performance from asymptotic scaling. The resulting framework provides a systematic basis for evaluating modular three-dimensional bosonic processors and for identifying the device-level parameters required for quantitative hardware design.

Zhu, Shaojiang [Fermilab] (ORCID:0000000293180092)↗
DOE OSTI2025

Reference Architecture Discussion [Slides]

NREL is developing reference architectures - industry-validated and verified models - to provide a common operational view and emulation environment for analyzing risk to the energy sector. In this presentation, we overview our work in developing these architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Unveiling stimulation fluid-driven alterations in shale pore architecture through combined interpretation of TD-NMR and multi-component gas adsorption

The hydraulic fracturing stimulation process can alter petrophysical and flow properties of the shale matrix in unconventional reservoirs. These alterations can result from interactions with native and exogenous fluids, and can be detrimental to reservoir productivity. Different experimental procedures have been proposed to study reactive processes and thereby understand the aforementioned interactions. Most published works do not consider the rock initial saturation with formation fluid, or the effect of organic matter (OM). To gain insights into these effects, two reactive experimental sets, differing in experimental conditions, were conducted. In the first set, at 125°C and 45 MPa, a stimulation fluid surrogate was put in contact with Baxter Shale samples, previously saturated and equilibrated with synthetic formation fluid. A baseline consisted of exposure to formation fluid exclusively. In the second set, at 0.08 MPa and 25°C, contact with stimulation fluid took place as in the first set. Two additional assays in this second set either skipped the initial formation fluid saturation or removed the OM, respectively. Petrophysical properties analyses, focused on the pore architecture, relied on the combined interpretation of gas adsorption and Time-Domain Nuclear Magnetic Resonance (TD-NMR). Results show that the presence of the formation fluid significantly restricts pore space accessibility to the stimulation fluid and mitigates the pore-architecture alteration to the macro-pore region. Mineral dissolution and precipitation dominate over surface or wettability alteration. Next, results suggest that the OM restricts accessibility to some pores. Ultimately, the equilibration process with synthetic formation fluid drives mainly surface alteration, unhindered by OM.

58 GEOSCIENCES↗
DOE OSTI2021

Micro-mechanical investigation of the thermo-mechanical properties of micro-architectured tungsten coatings

The thermo-mechanical response of micro-architectured tungsten coatings is characterized in the temperature range of 293 to 673 K using both in situ micro-compression experiments inside a scanning electron microscope (SEM) as well as image-based crystal plasticity finite element method (CPFEM) simulations. The experiments were conducted on micropillar-like specimens that were focus ion beam milled into the coatings, while the simulations were conducted on columnar-grained micropillar simulation cells constructed based on the statistics of the coating’s microstructure. The experimental results show that the stress–strain response and deformation mode exhibit a strong temperature-dependence. At room temperature, catastrophic failure is observed shortly after yield and is manifested in the form of intergranular fracture and buckling of individual columnar grains. With increasing temperature, this catastrophic failure is gradually suppressed and the material exhibits a steadier strain hardening response at 693 K. The CPFEM simulations are also shown to be in good agreement with the experimental results, and these simulations indicate that the material response is strongly influenced by the local crystallographic anisotropy and microstructure inhomogeneity. Furthermore, the simulations capture the underlying mechanisms that control the temperature-dependent transition in deformation mode. The current results highlight that the micro-architectured microstructure offers a great combination of excellent mechanical strength and structural integrity at elevated temperatures, which is of importance for high temperature applications.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Impact of electrode porosity architecture on electrochemical performances of 1 mm-thick LiFePO4 binder-free Li-ion electrodes fabricated by Spark Plasma Sintering

Thick electrodes with high active material loadings have been intensively studied over the last couple of decades in pursuit of achieving high energy density systems. To optimize and enhance the electrochemical performance of such electrodes, one has to control the pore morphology by, for example, varying the pore size and shape, and the level of porosity. In the present work, the fabrication of thick binder-free LiFePO4 (LFP) electrodes with two different pore sizes (12 and 20 mu m) and porosities (21 vol% and 44 vol%) using Spark Plasma Sintering (SPS) and templating approach is reported. The well-controlled porous architecture inside the thick electrodes is realized by fine-tuning experimental parameters. The impact of porosity architecture on electrochemical performance is quantified and correlated with the 3D tortuosity values determined from both micro-computed tomography and electrochemical impedance-based experimental methods. Based on the micro-computed tomography data analysis, estimated tortuosity values along X, Y, and Z axes reveal an anisotropic effect perpendicularly to the SPS compression axis (Z-direction). This is particularly profoundly observed in the samples with larger pores (20 mu m). The correlation between morphological properties and the rate capability performance is established indicating that the capacity loss happens mainly due to the Li-ion diffusion limitations.

Cathode material↗
DOE OSTI2026

Generalist multimodal AI: A review of architectures, challenges and opportunities

Multimodal models are expected to be a critical component to future advances in artificial intelligence. Here, this field is starting to grow rapidly with a surge of new design elements motivated by the success of foundation models in natural language processing (NLP) and vision. It is widely hoped that further extending the foundation models to multiple modalities (e.g., text, image, video, sensor, time series, graph, etc.) will ultimately lead to generalist multimodal models, i.e. one model across different data modalities and tasks. However, there is little research that systematically analyzes recent multimodal models (particularly the ones that work beyond text and vision) with respect to the underling architecture proposed. Therefore, this work provides a fresh perspective on generalist multimodal models (GMMs) via a novel architecture and training configuration specific taxonomy. This includes factors such as Unifiability, Modularity, and Adaptability that are pertinent and essential to the wide adoption and application of GMMs. The review further highlights key challenges and prospects for the field and guide the researchers into the new advancements.

Artificial intelligence (AI)↗
DOE OSTI2023

Quantifying uncertainty for deep learning based forecasting and flow-reconstruction using neural architecture search ensembles

Classical problems in computational physics such as data-driven forecasting and signal reconstruction from sparse sensors have recently seen an explosion in deep neural network (DNN) based algorithmic approaches. However, most DNN models do not provide uncertainty estimates, which are crucial for establishing the trustworthiness of these techniques in downstream decision making tasks and scenarios. In recent years, ensemble-based methods have achieved significant success for the uncertainty quantification in DNNs on a number of benchmark problems. However, their performance on real-world applications remains under-explored. In this work, we present an automated approach to DNN discovery and demonstrate how this may also be utilized for ensemble-based uncertainty quantification. Specifically, we propose the use of a scalable neural and hyperparameter architecture search for discovering an ensemble of DNN models for complex dynamical systems. We highlight how the proposed method not only discovers high-performing neural network ensembles for our tasks, but also quantifies uncertainty seamlessly. This is achieved by using genetic algorithms and Bayesian optimization for sampling the search space of neural network architectures and hyperparameters. Subsequently, a model selection approach is used to identify candidate models for an ensemble set construction. Afterwards, a variance decomposition approach is used to estimate the uncertainty of the predictions from the ensemble. We demonstrate the feasibility of this framework for two tasks — forecasting from historical data and flow reconstruction from sparse sensors for the sea-surface temperature. In conclusion, we demonstrate superior performance from the ensemble in contrast with individual high-performing models and other benchmarks.

Deep ensembles↗
DOE OSTI2020

Porous architecture and thermal properties of thermal barrier coatings deposited by suspension plasma spray

Besides the intrinsic low heat transfer capability of material, the thermal insulation property of thermal barrier coatings (TBCs) also relies on their microstructures. For better understanding the relationship between process parameters, porous architecture, and thermal properties of coatings, YSZ coatings were firstly manufactured by suspension plasma spray (SPS). Afterwards their total porosities were characterized by using the technique of X-ray transmission, the nano/submicro pores in those coatings were detected with Ultra-Small Angle X-ray Scattering (USAXS), and the thermal properties of coatings were measured using the laser flash method. The results indicated that: i) the porous architecture of SPS coatings can be tailored by adapting process parameters. ii) increasing total porosity is an effective means for reducing the heat transfer capability of SPS coatings. iii) nano/submicro pores have higher influence sensitivity on thermal properties than that of larger pores. Increasing the content of nano-submicro pores is therefore more effective for improving the thermal insulation property of SPS coatings. The potential of SPS technique for fabricating TBCs was discussed as well.

36 MATERIALS SCIENCE↗