Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “ARCHITECTURE”

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 649 records · Page 36
All sourcesNASA NTRSDOE OSTI
DOE OSTI2012

A Service Oriented Architecture for Exploring High Performance Distributed Power Models

Power grids are increasingly incorporating high quality, high throughput sensor devices inside power distribution networks. These devices are driving an unprecedented increase in the volume and rate of available information. The real-time requirements for handling this data are beyond the capacity of conventional power models running in central utilities. Hence, we are exploring distributed power models deployed at the regional scale. The connection of these models for a larger geographic region is supported by a distributed system architecture. This architecture is built in a service oriented style, whereby distributed power models running on high performance clusters are exposed as services. Each service is semantically annotated and therefore can be discovered through a service catalog and composed into workflows. The overall architecture has been implemented as an integrated workflow environment useful for power researchers to explore newly developed distributed power models.

Service oriented architecture, high performance co↗
DOE OSTI2023

Online evolutionary neural architecture search for multivariate non-stationary time series forecasting

Time series forecasting (TSF) is one of the most important tasks in data science. TSF models are usually pre-trained with historical data and then applied on future unseen datapoints. However, real-world time series data is usually non-stationary and models trained offline usually face problems from data drift. Models trained and designed in an offline fashion can not quickly adapt to changes quickly or be deployed in real-time. To address these issues, this work presents the Online NeuroEvolution-based Neural Architecture Search (ONE-NAS) algorithm, which is a novel neural architecture search method capable of automatically designing and dynamically training recurrent neural networks (RNNs) for online forecasting tasks. Without any pre-training, ONE-NAS utilizes populations of RNNs that are continuously updated with new network structures and weights in response to new multivariate input data. ONE-NAS is tested on real-world, large-scale multivariate wind turbine data as well as the univariate Dow Jones Industrial Average (DJIA) dataset. These results demonstrate that ONE-NAS outperforms traditional statistical time series forecasting methods, including online linear regression, fixed long short-term memory (LSTM) and gated recurrent unit (GRU) models trained online, as well as state-of-the-art, online ARIMA strategies. Additionally, results show that utilizing multiple populations of RNNs which are periodically repopulated provide significant performance improvements, allowing this online neural network architecture design and training to be successful.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Geochemical reactions and alteration of pore architecture in saturated shale after injection of stimulation fluid

Pore architecture regulates fluid flow between unconventional shale reservoir and hydraulically-induced fractures. Imbibition of stimulation fluid may change this architecture and alter hydrocarbon flow. Hydrothermal experiments were conducted at reservoir conditions (125 °C, 45 MPa) to test two hypotheses: 1) Shale, not stimulation fluid, dominates the geochemistry of an unconventional reservoir containing formation water; and 2) Mineral dissolution and precipitation induced by stimulation fluid is transient and manifests across micro-, meso-, and macro-scales of pore architecture. Experiments replicated a shut-in well in the Cretaceous Baxter Shale, Green River Basin, Wyoming USA. Stimulation fluid was injected into one experiment after formation water (I = 0.35 mol/kg, pH = 6.4) and core reacted for 48 days. This novel approach equilibrated rock and formation water and saturated pores with formation water before introducing stimulation fluid. The second experiment served as a control. Trends of aqueous calcium, silica and aluminum in the injection experiment suggest transient dissolution of calcite and feldspar and/or quartz as well as clay and barite precipitation; mineralogic evidence was limited to calcite dissolution and barite precipitation. The rock maintained reducing conditions (Eh = +0.08 to -0.16 V) despite injection of oxidizing stimulation fluid (Eh = 1.1 V). pH of the stimulation fluid-formation water mixture evolved from 2.3 to pre-injection values (~6) within 24 h. The results indicate that mineral dissolution and precipitation manifest in macropores with barely detectable alteration to micro- and mesopores. Formation water or organic matter in pores may have inhibited access of stimulation fluid to micro- and mesopores

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Controlled 3D nanoparticle deposition by drying of colloidal suspension in designed thin micro-porous architectures

Nanoparticle deposition by drying of colloidal suspension in thin micro-porous architectures has attracted a lot of attention in scientific research as well as industrial applications. However, the underlying mechanisms of such three-dimensional (3D) deposition are not yet fully revealed due to the complexity of the co-occurring processes of two-phase fluid flows, phase change and mass transport. Consequently, the control of 3D nanoparticle deposition remains a challenge. We use a combined experimental and numerical approach to achieve controlled 3D nanoparticle deposition by drying of colloidal suspension in two pillar-based thin micro-porous architectures. By the design of pillar layout, rectangular-spiral and circular-spiral deposition configurations are obtained globally. By varying the surface wettability, vertically symmetric and sloped nanoparticle depositions can be achieved locally. While the numerical modeling reveals the mechanisms of liquid internal flow, as well as the impact of local drying rate on nanoparticle transport, accumulation and final deposition, the experimental results of deposition configurations validate the controlling strategies. This combined experimental and numerical work provides a framework to achieve desired 3D nanoparticle deposition in thin micro-porous architectures, with a thorough understanding of the underlying mechanisms of two-phase fluid flows, phase change and mass transport.

36 MATERIALS SCIENCE↗
DOE OSTI2021

A scalable algorithm for the optimization of neural network architectures

In this work, we propose a new scalable method to optimize the architecture of an artificial neural network. The proposed algorithm, called Greedy Search for Neural Network Architecture, aims to determine a neural network with minimal number of layers that is at least as performant as neural networks of the same structure identified by other hyperparameter search algorithms in terms of accuracy and computational cost. Numerical results performed on benchmark datasets show that, for these datasets, our method outperforms state-of-the-art hyperparameter optimization algorithms in terms of attainable predictive performance by the selected neural network architecture, and time-to-solution for the hyperparameter optimization to complete.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Soil pore architecture and rhizosphere legacy define N 2 O production in root detritusphere

Root detritusphere is one of the most important sources of N 2 O, however, understanding of how N 2 O emission from the detritusphere is influenced by soil properties remains elusive. Here, we evaluated the effects of pore architecture and soil moisture on N 2 O emission during the decomposition of in-situ grown roots of switchgrass, an important bioenergy crop. We combined dual isotope labeling ( 15 C and 15 N) with zymography to gain insights into the location of the microbial N 2 O production in soils with contrasting pore architectures. In the studied soil, the effect of soil pore architecture on N 2 O emissions was 6 times greater than that of soil moisture. Soil dominated by > 30 μm Ø pores (i.e., large-pore soil) had higher chitinase activity than the soil dominated by < 10 μm Ø pores (i.e., small-pore soil), especially near the decomposing roots. The chitinase activity on the decomposing roots was positively correlated with emission of root-derived N 2 O, indicating that N released from root decomposition was an important source of N 2 O. Greater N 2 O and N2 emission was induced by switchgrass roots in soils dominated by the large- compared to the small-pore soils. Here, the microenvironment developed near decomposing roots of the large-pore soil also resulted in positive N 2 O priming. Our study challenged the traditional view on soil moisture as the main factor of N 2 O production. Production and emission of N 2 O was most intensive in microbial activity hotspots (i.e., rhizosphere legacy) in the large pores, where decomposed roots release mineral N as the main N 2 O source.

13C Pulse labeling↗
DOE OSTI2021

Nonlinear Architectures Can Alter the Dynamics of Polymer–Nanoparticle Composites

Polymer nanocomposites exhibit remarkable physical properties that are attractive for many applications. These systems have been so far investigated by using linear polymer chains; the role of polymer matrix architecture on local dynamics, bulk rheology and nanoparticle motion remains unexplored. Here, using quasielastic neutron scattering, bulk rheology and x-ray photon correlation spectroscopy, we investigated nanocomposites with spherical silica nanoparticles well dispersed in poly (ethylene oxide) matrices having different architectures (specifically linear, stars and hyperbranched). The results reveal that the mechanical reinforcement of the nanocomposites with the nonlinear polymers can be altered by orders of magnitude with respect to the conventional nanocomposite with the linear polymer. Additionally, the polymer compactness and inter-penetrability are found to play crucial roles in determining their bulk rheology. At microscopic level, average segmental dynamics is remarkably slowed down by the attractive NPs in the matrices of high degree of branching whereas no significant effect is observed in linear polymer matrix at the same NP loading. In addition, the nanoscale dynamics of particles in the compact nonlinear matrices exhibit strong decoupling from the bulk viscoelasticity, allowing their fast relaxation even at ≈ 30% by volume. These results provide an experimental evidence that macromolecular architecture is a powerful new tool for tuning the bulk rheological properties as well as the nanoscale dynamics of PNCs without the need for changing polymer molecular weight, nanoparticle size, shape, loading or dispersion state.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Conserved Macromolecular Architecture of Poplar Secondary Cell Walls Revealed by ssNMR and Atomistic Modeling

The macromolecular architecture of plant secondary cell walls governs wood's mechanical and biochemical properties, yet its natural intra-species variability remains poorly characterized. Here, we combined 13C solid-state NMR (ssNMR), multivariate statistical analysis, and molecular modeling to profile nanoscale structure across 13 genetically diverse Populus trichocarpa genotypes grown in 13C-enriched atmospheres. SsNMR-derived phenotypes spanning composition, structure, mobility, and inter-polymer proximities reveal a conserved architecture, with a subtle yet coordinated variation organizing into dominant structural and secondary mobility axes. A representative atomistic model captures these features and reproduces experimental metrics. Molecular dynamics simulations support a weak but consistent positive correlation between cellulose abundance and crystalline-like order, with interior cellulose chains enriched in tg (trans-gauche) conformations without expanding crystalline cores. Together, experiment and simulation reveal a genetically buffered, broadly conserved nanoscale architecture across genotypes, where subtle fine-tuning of cellulose bundling and matrix packing balances mechanical performance with biological function.

09 BIOMASS FUELS↗
DOE OSTI2012

A 3D Porous Architecture of Si/graphene Nanocomposite as High-performance Anode Materials for Li-ion Batteries

A 3D porous architecture of Si/graphene nanocomposite has been rationally designed and constructed through a series of controlled chemical processes. In contrast to random mixture of Si nanoparticles and graphene nanosheets, the porous nanoarchitectured composite has superior electrochemical stability because the Si nanoparticles are firmly riveted on the graphene nanosheets through a thin SiO{sub x} layer. The 3D graphene network enhances electrical conductivity, and improves rate performance, demonstrating a superior rate capability over the 2D nanostructure. This 3D porous architecture can deliver a reversible capacity of {approx}900 mA h g{sup -1} with very little fading when the charge rates change from 100 mA g{sup -1} to 1 A g{sup -1}. Furthermore, the 3D nanoarchitechture of Si/graphene can be cycled at extremely high Li{sup +} extraction rates, such as 5 A g{sup -1} and 10 A g{sup -1}, for over than 100 times. Both the highly conductive graphene network and porous architecture are considered to contribute to the remarkable rate capability and cycling stability, thereby pointing to a new synthesis route to improving the electrochemical performances of the Si-based anode materials for advanced Li-ion batteries.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2025

Neural architecture codesign for fast physics applications

We develop a pipeline to streamline neural architecture codesign for physics applications to reduce the need for ML expertise when designing models for novel tasks. Our method employs neural architecture search and network compression in a two-stage approach to discover hardware efficient models. This approach consists of a global search stage that explores a wide range of architectures while considering hardware constraints, followed by a local search stage that fine-tunes and compresses the most promising candidates. We exceed performance on various tasks and show further speedup through model compression techniques such as quantization-aware-training and neural network pruning. We synthesize the optimal models to high level synthesis code for FPGA deployment with the hls4ml library. Additionally, our hierarchical search space provides greater flexibility in optimization, which can easily extend to other tasks and domains. We demonstrate this with two case studies: Bragg peak finding in materials science and jet classification in high energy physics, achieving models with improved accuracy, smaller latencies, or reduced resource utilization relative to the baseline models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Novel Module Architecture for Lower CapEx and Improved Recyclability for c-Si PV Modules

Photovoltaic (PV) energy production is currently increasing at a rate at which recycling is becoming necessary. A novel module architecture has been demonstrated that has potential for high value recycling for c-Si PV. This architecture eliminates the vacuum lamination process and cross-linked en capsulants. Functioning prototypes of c-Si have been fabricated for stress testing in collaboration with NREL. These modules are being tested against traditionally manufactured modules. Based on preliminary results, this module architecture is a potentially viable solution for improving the manufacturing cost and recyclability of PV modules while retaining module performance.

computer architecture↗
DOE OSTI2012

GridOPTICS(TM): A Design for Plug-and-Play Smart Grid Software Architecture

As the smart grid becomes reality, software architectures for integrating legacy systems with new innovative approaches for grid management are needed. These architectures must exhibit flexibility, extensibility, interoperability and scalability. In this position paper, we describe our preliminary work to design such an architecture, known as GridOPTICS, that will enable the deployment and integration of new software tools in smart grid operations. Our preliminary design is based upon use cases from PNNL’s Future Power Grid Initiative, which is a developing a collection of advanced software technologies for smart grid management and control. We describe the motivations for GridOPTICS, and the preliminary design that we are currently prototyping for several distinct use cases.

Smart grid↗
DOE OSTI2024

Altering the substitution and cross‐linking of glucuronoarabinoxylans affects cell wall architecture in Brachypodium distachyon

Summary The Poaceae family of plants provides cereal crops that are critical for human and animal nutrition, and also, they are an important source of biomass. Interacting plant cell wall components give rise to recalcitrance to digestion; thus, understanding the wall molecular architecture is important to improve biomass properties. Xylan is the main hemicellulose in grass cell walls. Recently, we reported structural variation in grass xylans, suggesting functional specialisation and distinct interactions with cellulose and lignin. Here, we investigated the functions of these xylans by perturbing the biosynthesis of specific xylan types. We generated CRISPR/Cas9 knockout mutants in Brachypodium distachyon XAX1 and GUX2 genes involved in xylan substitution. Using carbohydrate gel electrophoresis, we identified biochemical changes in different xylan types. Saccharification, cryo‐SEM, subcritical water extraction and ssNMR were used to study wall architecture. Bd XAX1A and Bd GUX2 enzymes modify different types of grass xylan. Brachypodium mutant walls are likely more porous, suggesting the xylan substitutions directed by both Bd XAX1A and GUX2 enzymes influence xylan‐xylan and/or xylan–lignin interactions. Since xylan substitutions influence wall architecture and digestibility, our findings open new avenues to improve cereals for food and to use grass biomass for feed and the production of bioenergy and biomaterials.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

The architecture of resilience: a genome assembly of Myrothamnus flabellifolia sheds light on desiccation tolerance and sex determination

Myrothamnus flabellifolia is a dioecious resurrection plant endemic to southern Africa that has become an important model for understanding desiccation tolerance. Despite its ecological and medicinal significance, genomic and transcriptomic resources for the species are limited. We generated a chromosome-level, haplotype-resolved reference genome assembly and annotation for M. flabellifolia and conducted transcriptomic profiling across a natural dehydration–rehydration time course in the field. Genome architecture and sex determination were characterized, and co-expression network and cis-regulatory element (CRE) enrichment analyses were used to investigate dynamic responses to desiccation. The 1.28-Gb genome exhibits unusually consistent chromatin architecture with unique chromosome organization across highly divergent haplotypes. We identified an XY sexual system with a small sex-determining region on Chromosome 8. Transcriptomic responses varied with dehydration severity, pointing to early suppression of growth, progressive activation of protective mechanisms, and subsequent return to homeostasis upon rehydration. Late embryogenesis abundant and early light-induced protein transcripts were dynamically regulated and showed enrichment of abscisic acid and stress-responsive CREs pointing toward conserved responses. Together, this study provides foundational resources for understanding the genomic architecture and reproductive biology of M. flabellifolia and offers new insights into the mechanisms of desiccation tolerance.

chromosome structure↗
DOE OSTI2021

Incorporating Valve Switching Losses Into a Static Optimal Control Algorithm for the Hybrid Hydraulic-Electric Architecture (HHEA)

The Hybrid Hydraulic-Electric Architecture (HHEA) has, in recent years, been proposed as an energy efficient alternative to conventional load-sensing architectures in mobile machines such as excavators and wheel-loaders. HHEA leverages the advantages of hydraulic power and electric power to eliminate throttling valves while also improving the energy and control performance of the system. The architecture utilizes a set of common pressure rails to provide a majority of power and and a small electric motor driven pump to modulate this power to meet the exact demand. Previous work has developed a computationally efficient Lagrange Multiplier approach for determining the optimal pressure rail selections that minimizes the energy losses in the system. The static model used considers only the energy use for each pressure rail selection but not the losses associated with the valves during the transition. This paper presents an approach to include the switching losses in the model and in the optimization procedure. To capture the switching losses, switching events between different rails and at various input and output flow rates were simulated with consideration of valve spool dynamics. A parameterized model that summarizes the losses is then obtained, allowing switching losses to be added to the previous energy analysis. The performance of the switching loss model was compared with reference data obtained from a high-fidelity simulation model. To incorporate the switching losses into optimal control algorithm, an efficient dynamic programming approach that prevents frequent switching is adopted in place of the Lagrange multiplier approach. The overall effect of switching losses on the energy consumption and optimal control decisions is presented. In general, switching losses contribute to about 9–10% of input energy.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2018

System Architecture Directions for Post-SoC/32-bit Networked Sensors

The emergence of low-power 32-bit Systems-on-Chip (SoCs), which integrate a 32-bit MCU, radio, and flash, presents an opportunity to re-examine design points and trade-offs at all levels of the system architecture of networked sensors. To this end, we develop a post-SoC/32-bit design point called Hamilton, showing that using integrated components enables a ~$7 core and shifts hardware modularity to design time. We study the interaction between hardware and embedded operating systems, identifying that (1) post-SoC motes provide lower idle current (5.9 μA) than traditional 16-bit motes, (2) 32-bit MCUs are a major energy consumer (e.g., tick increases idle current >50 times), comparable to radios, and (3) thread-based concurrency is viable, requiring only 8.3 μs of context switch time. We design a system architecture, based on a tickless multithreading operating system, with cooperative/adaptive clocking, advanced sensor abstraction, and preemptive packet processing. Its efficient MCU control improves concurrency with ~30% less energy consumption. Together, these developments set the system architecture for networked sensors in a new direction.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2019

MetaStrider: Architectures for Scalable Memory-centric Reduction of Sparse Data Streams

Reduction is an operation performed on the values of two or more key-value pairs that share the same key. Reduction of sparse data streams finds application in a wide variety of domains such as data and graph analytics, cybersecurity, machine learning, and HPC applications. However, these applications exhibit low locality of reference, rendering traditional architectures and data representations inefficient. This article presents MetaStrider, a significant algorithmic and architectural enhancement to the state-of-the-art, SuperStrider. Furthermore, these enhancements enable a variety of parallel, memory-centric architectures that we propose, resulting in demonstrated performance that scales near-linearly with available memory-level parallelism.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

XploreNAS : Explore Adversarially Robust and Hardware-efficient Neural Architectures for Non-ideal Xbars

Compute In-Memory platforms such as memristive crossbars are gaining focus as they facilitate acceleration of Deep Neural Networks (DNNs) with high area and compute efficiencies. However, the intrinsic non-idealities associated with the analog nature of computing in crossbars limits the performance of the deployed DNNs. Furthermore, DNNs are shown to be vulnerable to adversarial attacks leading to severe security threats in their large-scale deployment. Thus, finding adversarially robust DNN architectures for non-ideal crossbars is critical to the safe and secure deployment of DNNs on the edge. This work proposes a two-phase algorithm-hardware co-optimization approach called XploreNAS that searches for hardware efficient and adversarially robust neural architectures for non-ideal crossbar platforms. We use the one-shot Neural Architecture Search approach to train a large Supernet with crossbar-awareness and sample adversarially robust Subnets therefrom, maintaining competitive hardware efficiency. Our experiments on crossbars with benchmark datasets (SVHN, CIFAR10, CIFAR100) show up to ~8–16% improvement in the adversarial robustness of the searched Subnets against a baseline ResNet-18 model subjected to crossbar-aware adversarial training. We benchmark our robust Subnets for Energy-Delay-Area-Products (EDAPs) using the Neurosim tool and find that with additional hardware efficiency–driven optimizations, the Subnets attain ~1.5–1.6× lower EDAPs than ResNet-18 baseline.

97 MATHEMATICS AND COMPUTING↗