Structure Prediction from Neutron Scattering Profiles: A Data Sciences Approach
Explore the source record for details and available documents.
Computer Science↗
SEARCH · Engineering Papers
Results for “network performance evaluation”
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.
Connect the Dots: In Situ 4-D Seismic Monitoring of CO 2 Storage With Spatio-Temporal CNNs
4-D seismic imaging has been widely used in CO 2 sequestration projects to monitor the fluid flow in the volumetric subsurface region that is not sampled by wells. Ideally, real-time monitoring and near-future forecasting would provide site operators with great insights to understand the dynamics of the subsurface reservoir and assess any potential risks. However, due to obstacles such as high deployment cost, availability of acquisition equipment, exclusion zones around surface structures, only very sparse seismic imaging data can be obtained during monitoring. That leads to an unavoidable and growing knowledge gap over time. The operator needs to understand the fluid flow throughout the project lifetime and the seismic data are only available at a limited number of times. This is insufficient for understanding reservoir behavior. To overcome those challenges, we have developed spatio-temporal neural-network-based models that can produce high-fidelity interpolated or extrapolated images effectively and efficiently. Specifically, our models are built on an autoencoder, and incorporate the long short-term memory (LSTM) structure with a new loss function regularized by optical flow. We validate the performance of our models using real 4-D post-stack seismic imaging data acquired at the Sleipner CO 2 sequestration field. We employ two different strategies in evaluating our models. Numerically, we compare our models with different baseline approaches using classic pixel-based metrics. We also conduct a blind survey and collect a total of 20 responses from domain experts to evaluate the quality of data generated by our models. Finally, via both numerical and expert evaluation, we conclude that our models can produce high-quality 2-D/3-D seismic imaging data at a reasonable cost, offering the possibility of real-time monitoring or even near-future forecasting of the CO 2 storage reservoir.
4-D seismic imaging↗
Integrative Modeling and Analysis of Fungal Central Carbon Metabolism
Over a thousand fungal genomes have been sequenced, yet manually curated genome-scale metabolic models (GEMs) are available for only a limited number of species. Moreover, these models have often been developed independently, leading to inconsistencies in namespaces, compartment definitions, and pathway representations that hinder comparative analysis, the systematic reuse of prior curation efforts, and the integration of consolidated metabolic knowledge. Here, we present the Consolidated Fungal Core Metabolism Model (CFCMM), constructed by integrating thirteen published fungal models spanning Ascomycota, Mucoromycota, and both Crabtree-positive and Crabtree-negative yeasts. We harmonized metabolites and reactions into a non-redundant shared ModelSEED ontological space, standardized compartmentalization, and refined gene–protein–reaction (GPR) rules. Using pathway-level visualization and systematic gap detection, we further improved the integrated network through literature-guided curation to correct stoichiometry, stereospecificity, and pathway architecture. Orthologous protein family reconstruction and functional annotation workflows were used to validate and inform GPR associations, with particular emphasis on ambiguous enzyme superfamilies and membrane-associated components. Using the resulting CFCMM, we built high-quality central carbon core models for each fungus and performed flux balance analysis to quantify ATP-yield variation under aerobic and anaerobic conditions, explicitly evaluating scenarios driven by differences in electron transport chain (ETC) composition. Simulations reproduced the expected fermentative yield of approximately 2 mmol ATP per mmol glucose under anaerobic conditions and separated the thirteen fungi into two bioenergetic groups under aerobic respiration based on Complex I status, with predicted yields of approximately 30 versus 22 mmol ATP per mmol glucose. Forcing flux through the alternative oxidase bypass further reduced ATP yields to approximately 12 and 4 mmol ATP per mmol glucose in Complex I-containing and Complex I-lacking fungi, respectively. Collectively, this work provides a manually curated, ModelSEED-consistent, and extensible fungal core metabolic template, deployed in DOE KBase as a resource for automated reconstruction of central carbon core models from any sequenced fungal genome. In addition, the CFCMM provides modular components for developing GEMs with more accurate energy predictions and enables robust comparative analyses of fungal bioenergetics and core metabolic diversity
59 BASIC BIOLOGICAL SCIENCES↗
Demonstrate new plasticity models for doped UO 2 that capture dislocation mechanisms
In light water reactors, fuel vendors are investigating the use of dopants to modify the properties of UO 2 pellets, with the goal of improving pellet-cladding mechanical interactions during operation. Dopants are expected to ‘soften’ the pellets; that is, the doped pellets have higher plastic deformation than conventional UO 2 . This leads to a reduction in the severity of mechanical pellet-cladding interactions, helping to reduce the hoop strain on the cladding. By minimizing the strain exerted by the pellet on the cladding, it is anticipated that cladding performance under accident conditions can be enhanced (i.e., lowering the risk of burst during a LOCA). Dopants such as chromium (Cr) promote grain growth during pellet fabrication, leading to larger grains; therefore, understanding the link between chemistry, microstructure and mechanical deformation (enhanced creep rates) behavior of UO 2 is critical to helping operators further substantiate the benefits of doping UO 2 . Historically, the nuclear energy industry has relied on empirical models to make assessments of performance. Compared to empirical models, mechanistic physics-based models provide benefits, such as, fewer data points for validation and better extrapolation where experimental data is scarce or non-existent. In this report, Bayesian inference techniques have been applied to a previously developed lower length-scale-informed diffusional creep model. The objective is to i) infer lower-length-scale parameter distributions from available experiment and then ii) determine the uncertainties in the measurable quantity (in this case creep rates) after propagating the inferred lower length scale parameter uncertainties. The approach requires many evaluations of the model, which becomes computationally insurmountable; therefore, a neural-network model is trained to data obtained by sampling the full model over the most important parameters. This neural-network is then used in the Bayesian inference approach to determine probability distributions in the parameter values that represent the uncertainty in the model given what is known from the experiments (posterior). A significant reduction compared to conservative initial (prior) uncertainties is achieved through inference against the experimental data, demonstrating the efficacy of this approach. Furthermore, by accounting for uncertainties in the experimental conditions and sample non-stoichiometry, it is possible to resolve apparent discrepancies in experimental measurements within a self-consistent grain boundary (Coble) creep model that is sensitive to chemistry. This work has been written up and submitted to Nuclear Technology for a special issue on accelerated fuel qualification (AFQ). This uncertainty quantification (UQ) work not only improves the diffusional model, while accounting for uncertainty, but also establishes a framework which can readily be applied to the mechanistic models of dislocation deformation developed in this study. The most likely values from the Bayesian analysis are incorporated into our UO 2 diffusional creep model and a lower length scale-informed irradiation UO 2 creep mechanistic model to generate a dataset. This dataset has been provided to our INL collaborators for training an artificial neural network surrogate model, which will be implemented in the BISON fuel performance code to assess how the results differ from those currently obtained using a fully empirical model and that of using the nominal (uncalibrated) atomic scale parameters in our mechanistic model. Plastic deformation (creep and glide) in UO 2 is a complex phenomenon, governed by multiple underlying processes such as local defect concentrations, applied stresses, and microstructural characteristics. Consequently, there is a need for a meso-scale model with polycrystalline resolution capable of extrapolating to large grain sizes applicable to doped UO 2 , where data is limited and the model can help bridge the knowledge gap. By integrating atomistic data into the polycrystal LApx code, it becomes possible to predict dislocation climb and glide plasticity that simple analytical models cannot accurately represent. The application of atomic-scale data within LApx demonstrated the importance of climb and glide mechanisms in reproducing high-stress UO 2 behavior. Behaviors such as this are crucial to capture and implement in BISON, as parts of the fuel pellet can reach temperatures where glide can occur before pellet cracking. This model which captures dislocation based mechanisms for UO 2 is then used to stand up the doped model accounting for larger grain sizes. It was found that larger grain sizes can lead to enhanced deformation rates in the glide regime, and therefore can help with the pellet cladding mechanical interaction. Therefore if the fuel pellet reaches conditions (stress/temperature) where glide is active, the enhanced creep rates for larger grains in the glide regime (doped UO 2 ) can help with pellet cladding mechanical interactions. Plastic deformation in UO 2 involves multiple mechanisms, including diffusional creep, dislocation climb, and glide. This milestone contains two parts: (1) UQ of a pre-existing lower length scale informed mechanistic diffusional creep model, and (2) development of a new LApx based model for dislocation-mediated creep mechanisms in UO 2 , with application to large-grain doped UO 2 .
11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
Geomechanical and Hydrogeological Evaluation of a Shallow Hydraulic Fracture at the Devine Fracture Pilot Site, Medina County, Texas
UT-Austin’s Devine Fracture Pilot Site, 50 miles southwest of San Antonio, Texas, has been targeted for a comprehensive, multidisciplinary development of fracture diagnostic techniques that are cross-validated by ground-truth data acquisition near a recently created, 175-ft-deep, horizontal hydraulic fracture (Ahmadian et al. 2018 Demonstration of proof of concept of electromagnetic geophysical methods for high resolution illumination of induced fracture networks. In Proceedings of the SPE Hydraulic Fracturing Technology Conference and Exhibition, The Woodlands, Texas, USA, 23–25 January 2018. SPE-189858-MS.). To evaluate the fracture diagnostic methods at this site, we conducted injection tests with a predefined volumetric flow-rate profile, resembling a diagnostic fracture injection test on September 2020. Subsequently, we developed hydrogeological and geomechanical models based on flow-rate and bottomhole-pressure measurements. History-matching efforts using a simplified layer-cake hydrogeological model resulted in the field-scale formation permeability of 9.87 × 10 –15 m 2 (10 mD) and Darcy-scale fracture permeability. The analysis of the bottomhole pressure and injection-rate history showed that (1) the newly created horizontal fracture was closed adjacent to the injection well pre-injection and (2) the initial pump-pressure increase at a nominal volumetric injection rate led to near-well fracture reopening, fluid conductivity increase, and abrupt injection-rate increase. To overcome hydrogeological-model limitations of predicting fracture reopening throughout injection, we extended the modeling to a finite-element, poroelastic analysis of horizontal-fracture growth using a cohesive-zone model. Using this fracture-reopening model, we improved the history match of the transient-pressure response during the experiment by adjusting the hydromechanical properties. Furthermore, post-injection pressure transient analyses helped reduce uncertainty in the overburden-stress gradient, and the initial hydraulic-fracturing simulation verified the plausibility of achieving the surveyed propped fracture area.
02 PETROLEUM↗
A leaf-level spectral library to support high-throughput plant phenotyping: predictive accuracy and model transfer
Abstract Leaf-level hyperspectral reflectance has become an effective tool for high-throughput phenotyping of plant leaf traits due to its rapid, low-cost, multi-sensing, and non-destructive nature. However, collecting samples for model calibration can still be expensive, and models show poor transferability among different datasets. This study had three specific objectives: first, to assemble a large library of leaf hyperspectral data (n=2460) from maize and sorghum; second, to evaluate two machine-learning approaches to estimate nine leaf properties (chlorophyll, thickness, water content, nitrogen, phosphorus, potassium, calcium, magnesium, and sulfur); and third, to investigate the usefulness of this spectral library for predicting external datasets (n=445) including soybean and camelina using extra-weighted spiking. Internal cross-validation showed satisfactory performance of the spectral library to estimate all nine traits (mean R2=0.688), with partial least-squares regression outperforming deep neural network models. Models calibrated solely using the spectral library showed degraded performance on external datasets (mean R2=0.159 for camelina, 0.337 for soybean). Models improved significantly when a small portion of external samples (n=20) was added to the library via extra-weighted spiking (mean R2=0.574 for camelina, 0.536 for soybean). The leaf-level spectral library greatly benefits plant physiological and biochemical phenotyping, whilst extra-weight spiking improves model transferability and extends its utility.
59 BASIC BIOLOGICAL SCIENCES↗
Acceleration of Graph Neural Network-Based Prediction Models in Chemistry via Co-Design Optimization on Intelligence Processing Units
Atomic structure prediction and associated property calculations are the bedrock of chemical physics. Since high-fidelity ab initio modeling techniques for computing the structure and properties can be prohibitively expensive, this motivates the development of machine-learning (ML) models that make these predictions more efficiently. Training graph neural networks over large atomistic databases introduces unique computational challenges such as the need to process millions of small graphs with variable size and support communication patterns that are distinct from learning over large graphs such as social networks. We demonstrate a novel hardware-software co-design approach to scale up the training of atomistic graph neural networks (GNN) for structure and property prediction. First, to eliminate redundant computation and memory associated with alternative padding techniques and to improve throughput via minimizing communication, we formulate the effective coalescing of the batches of variable-size atomistic graphs as the bin packing problem and introduce a hardware-agnostic algorithm to pack these batches. In addition, we propose hardware-specific optimizations including a planner and vectorization for the gather-scatter operations targeted for Graphcore’s Intelligence Processing Unit (IPU), as well as model-specific optimizations such as merged communication collectives and optimized softplus. Putting these all together, we demonstrate the effectiveness of the proposed co-design approach by providing an implementation of a well-established atomistic GNN on the Graphcore IPUs. We evaluate the training performance on multiple atomistic graph databases with varying degrees of graph counts, sizes and sparsity. Here, we demonstrate that such a co-design approach can reduce the training time of atomistic GNNs and can improve the performance by up to 1.5× compared to the baseline implementation of the model on the IPUs. Additionally, we compare our IPU implementation with a Nvidia GPU-based implementation and show that our atomistic GNN implementation on the IPUs can run 1.8× faster on average compared to the execution time on the GPUs.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Evaluating disease surveillance strategies for early outbreak detection in contact networks with varying community structure
Disease surveillance systems allow public health agencies to respond to emerging diseases before they become widespread. Developing such systems requires identifying optimal ways to monitor in the context of an epidemic outbreak; this problem is known as sensor selection. Contact networks represent the dynamics of interaction in a population and are used to model how a disease spreads in a population and to explore strategies of sensor selection. We evaluated five sensor selection strategies on their ability to provide an early warning of a COVID-like outbreak in synthetic contact networks encapsulated in four network scenarios. Three of these scenarios assessed different aspects of community structure. The fourth scenario employed a contact network representing the population and interactions of 6.8 million people in New York City, constructed from an agent-based simulation using census and transportation data. This scenario exemplifies how sensor selection strategies may perform in a real-world, urban context. Our findings suggest that the choice of the optimal strategy depends heavily on the community structure of the network. Strategies that select highly connected nodes or maximize network coverage are the optimal surveillance strategy for outbreak detection in many network community structures. However, a naive implementation of these strategies may fail to provide an early warning at all—including in the New York City scenario. Moreover, these methods are impractical for real-world use as they require knowledge of the underlying contact network. Instead, a selection strategy that starts with a set of random nodes and then performs a random walk through a chain of neighbors reliably provides early warnings without requiring prior knowledge of the network. We find this method, called “random chain”, to be the most pragmatic for implementation in a real-world disease surveillance context.
60 APPLIED LIFE SCIENCES↗
A Particle-Based Conditional Sampling Scheme for the Simulation of Transport in Fractured Rock With Diffusion Into Stagnant Water and Rock Matrix
In situ experiments and field-scale characterization studies have pointed out that, in fractured crystalline media, groundwater flow is highly channelized. This implies that, at the scale of a single fracture, only part of the fracture surface area is in contact with flowing water, while the rest of in-plane water is essentially stagnant and can be accessed by solutes via molecular diffusion. Despite their importance for contaminant retention, to date, there are no numerical or analytical approaches that could be used to assess the implication of stagnant water zones on solute transport in realistic large-scale Discrete Fracture Network-based models. In this study, we present an efficient and flexible algorithm for the simulation of transport in fractured rock with diffusion into stagnant water and rock matrix. The algorithm is a generalization of a previously developed numerical framework for time domain particle tracking in sparsely fractured rock. The key of the generalization is that total time in fracture (τ f ) is first evaluated using a Monte Carlo sampling and then a second sampling is performed conditioned on τ f . The algorithm has been successfully validated against existing independent solutions and the implication of diffusion into stagnant water and secondary diffusion into the matrix has been assessed for a realistic modeling scenario. The results show that, due to diffusion into stagnant water, contaminants are more strongly retarded. This increased retention is more significant for sorbing species, as a larger number of sorption sites is accessible. A high sensitivity to the flowing channel/stagnant water zone geometry has also been observed.
54 ENVIRONMENTAL SCIENCES↗
Evaluate the System-Level Impact of Connected and Automated Vehicles Coupled with Shared Mobility: An Agent-based Simulation Approach
With the rapid growth of information and communication technologies, Connected and Automated Vehicles (CAVs) are deemed to be disruptive with the potential to significantly improve overall transportation system efficiency, however, may bring Vehicle Miles Traveled (VMT) increase or other issues. Further, shared mobility systems are another disruptive force that is reshaping our travel patterns. To quantify the combined impact of CAV and shared mobility on travel behavior, traffic performance and energy efficiency, we develop a mesoscopic simulation-based framework for mobility and energy efficiency evaluation considering the disruptive transportation technologies. Under this framework, we develop novel models for energy intensity and modal activity, and evaluated a variety of energy scenarios for different combinations of CAV applications, various levels of automation, roadway characteristics, and traffic conditions, while also varying different vehicle types and fuel/powertrain technologies. Based applying this modeling suite to a calibrated BEAM simulation network in Riverside California, it was found that cooperative automated driving in general will improve mobility, but automated vehicles, even when deployed in a shared autonomous fleet, will likely bring an increase of VMT (up to 36%) due to mode shifts and deadheading. Ride-hailing vehicles typically have better energy efficiency and a higher share of electric vehicles, which helps offset the negative impact from VMT increases when estimating the system-level energy consumption. In general, simulation results show a 6% increase in energy consumption for the scenarios with an increasing shift to ride-hailing modes.
42 ENGINEERING↗
Representing Carbon Dioxide Transport and Storage Network Investments within Power System Planning Models
Carbon dioxide (CO 2 ) capture and storage (CCS) is frequently identified as a potential component to achieving a decarbonized power system at least cost; however, power system models frequently lack detailed representation of CO 2 transportation, injection, and storage (CTS) infrastructure. In this paper, we present a novel approach to explicitly represent CO 2 storage potential and CTS infrastructure costs and constraints within a continental-scale power system capacity expansion model. In addition, we evaluate the sensitivity of the results to assumptions about the future costs and performance of CTS components and carbon capture technologies. We find that the quantity of CO 2 captured within the power sector is relatively insensitive to the range of CTS costs explored, suggesting that the cost of CO 2 capture retrofits is a more important driver of CCS implementation than the costs of transportation and storage. Finally, we demonstrate that storage and injection costs account for the predominant share of total costs associated with CTS investment and operation, suggesting that pipeline infrastructure costs have limited influence on the competitiveness of CCS.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Parameterization of generic positive sequence models to represent behavior of inverter based resources in low short circuit scenarios
The connection of large inverter based resources (IBR) in transmission systems is often located geographically and electrically far away from load centers. This, coupled with the displacement of synchronous machine plants, results in a reduction of the network short circuit strength at the point of connection. Under these conditions, state-of-the-art positive sequence simulation platforms and models can have difficulties maintaining numerical stability and/or providing an accurate representation of IBR plant dynamic behavior. As a result, computationally heavy time domain electromagnetic transient (EMT) simulations may be required to evaluate these systems. In this paper, a recently developed improved generic positive sequence model is parameterized to provide a representation of IBR behavior in low short circuit networks. Further, comparisons against generic and detailed EMT models demonstrate the suitability of the improved positive sequence model to study practical stability issues experienced with presently in-service plants. Such a model can provide some of the accuracy of an EMT representation with a much lower computational burden. The performance of the positive sequence model is validated against the behavior shown by both open white box and closed black box EMT domain models from around the world.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Navigating high-dimensional process-structure–property relations in nanocrystalline Pt-Au alloys with machine learning
For decades, materials scientists have relied on the process-structure–property paradigm to guide investigations into material behaviors. Traditional studies often examine a limited number of process-structure–property variables, striving to elucidate mechanisms governing material response. However, this approach is time consuming and can limit exploration, as well as the discovery of process-structure–property relations in novel materials. In this paper, we combined combinatorial sputter deposition and multi-modal high-throughput materials characterization with feedforward neural networks to establish high-dimensional process-structure–property relations in Pt-Au alloys, yielding nanocrystalline alloys with high hardness and low resistivity relevant to electrical contact switch applications. We mapped three indicators of process conditions (composition and two atomic deposition characteristics) onto four indicators of material structure (X-ray diffraction, film thickness, density, and surface roughness) and two indicators of material properties (hardness and resistivity), resulting in 784 unique combinations evaluated over a 13-dimensional space. The neural networks predicted Pt-Au alloys with 18–24 at.% Au, when deposited at specific conditions, to have a nanoindentation hardness up to 7.2 GPa. This high hardness value, comparable to some steels, represents a 3-fold improvement in hardness over “hard gold”, a commonly used electrical contact alloy, while maintaining requisite electrical conductivity. The neural network models provide an avenue to identify expected process windows capable of maximizing material performance.
Electrical contact materials↗
Iodine capture with mechanically robust heat-treated Ag-Al-Si-O xerogel sorbents
Silver-loaded heat-treated aluminosilicate xerogels (Ag-HTX) were evaluated as sorbents for iodine [I2(g)] capture. The heat treatment step was performed to help increase the mechanical integrity of the gels. The synthesized xerogels were characterized using powder X-ray diffraction, scanning electron microscopy, energy-dispersive X-ray spectroscopy, transmission electron microscopy, Brunauer-Emmett-Teller analysis, gravimetric iodine loading, nanoindentation, and dynamic mechanical analysis. The structural and chemical analyses of Ag-HTX showed uniform distribution of Ag throughout the gel network after Ag-exchange. After I2(g) capture, the AgI crystallites were observed in the sorbent verifying chemisorption as the iodine capture mechanism. Iodine loading of this xerogel was 0.43 g g-1 at 150°C over 1 d and 0.52 g g-1 at 22°C over 33 d. The specific surface area of Ag-HTX was 202 m2 g-1 and decreased to 87 m2 g-1 after iodine loading. The hardness of the heat-treated xerogel was 160 times higher than heat-treated aerogel of the same composition. The heat-treatment process increased the modulus value to 40.77 MPa from 6.99 MPa of as-made xerogel, demonstrating the need for this added step in the synthesis process. These results show that Ag-HTX is a promising sorbent for I2(g) capture with good loading capacity and mechanical stability.
xerogels, aerogels, iodine capture, hardness testi↗
Effect of image resolution on automated classification of chest X-rays
Deep learning (DL) models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays (CXRs). Recently available public CXR datasets include high resolution images, but state-of-the-art models are trained on reduced size images due to limitations on graphics processing unit memory and training time. As computing hardware continues to advance, it has become feasible to train deep convolutional neural networks on high-resolution images without sacrificing detail by downscaling. This study examines the effect of increased resolution on CXR classification performance. We used the publicly available MIMIC-CXR-JPG dataset, comprising 377,110 high resolution CXR images for this study. We applied image downscaling from native resolution to 2048 × 2048 pixels, 1024 × 1024 pixels, 512 × 512 pixels, and 256 × 256 pixels and then we used the DenseNet121 and EfficientNet-B4 DL models to evaluate clinical task performance using these four downscaled image resolutions. We find that while some clinical findings are more reliably labeled using high resolutions, many other findings are actually labeled better using downscaled inputs. We qualitatively verify that tasks requiring a large receptive field are better suited to downscaled low resolution input images, by inspecting effective receptive fields and class activation maps of trained models. Lastly, we show that stacking an ensemble across resolutions outperforms each individual learner at all input resolutions while providing interpretable scale weights, indicating that diverse information is extracted across resolutions.
47 OTHER INSTRUMENTATION↗
Development and transferability of neural-network models for plasma-surface interactions
Plasma-surface interactions are increasingly critical to modern technologies; yet, accurate molecular dynamics simulations remain limited by the capabilities of interatomic potentials. Deep Potentials (DPs) promise to revolutionize the field by providing a systematic method for producing accurate interatomic potentials. The primary challenge of DP development is selecting a dataset, which efficiently spans the set of atomic environments one expects to encounter in the subsequent molecular dynamics simulations. The computational cost of density functional theory calculations, which are the typical basis for DP development, makes it impossible to directly verify the quality of a given DP. To address this challenge, we explore the development of a deep-learned interatomic potential, “DeepREBO,” trained to reproduce the behavior of the REBO2 empirical potential, enabling direct validation of training methodology and transferability. Using an active learning framework, we begin with a minimal dataset and iteratively expand it to train a Deep Potential-Smooth Edition model that faithfully reproduces REBO2 results for 25 eV hydrogen bombardment of diamond (001), a particularly challenging case. We show that small, carefully curated datasets can outperform large, unguided ones, with effective models requiring fewer than 15 000 snapshots. Subsequent transferability tests demonstrate that while DeepREBO generalizes well to diamond (111) surfaces, performance degrades for amorphous carbon or higher-energy impacts, highlighting the need for use-case-specific training data. We also evaluate methods to improve short-range repulsion. This study outlines best practices for training robust deep potentials and underscores the importance of dataset design for predictive plasma simulations.
Ab-initio molecular dynamics↗
Photovoltaic Analysis and Response Support (PARS) Platform for Solar Situational Awareness and Resiliency Services
The project's primary objective is to develop a digital-twin based Photovoltaic (PV) Analysis and Response Support (PARS) platform, which aims to provide real-time situational awareness and optimal response plans. This platform is designed to enhance the performance of hybrid PV systems, making them competitive with or even superior to conventional generation resources. The PARS platform enabled the project team to develop and evaluate an extensive suite of grid support functionalities for the hybrid PV systems to enhance grid performance, across key areas including visibility, dispatchability, security, resilience, and reliability. Given the global push toward achieving 100% clean energy by 2035, there is a significant increase in the integration of inverter-based resources (IBRs) throughout the energy grid. Effectively managing the inherent variability and uncertainty associated with IBRs is crucial for ensuring cost-effectiveness, reliability, and security in both the main grid and islanded microgrids. Constrained to a limited array of IEEE test systems or standard feeder models, traditional IBR modeling struggles to assimilate new field data, accurately reflect system dynamics, and adapt to the evolving energy landscape. In our project, we embraced a Digital Twin (DT) strategy for crafting the PARS platform. A digital twin acts as a precise virtual counterpart of a physical system, built on historical data and continuously honed with real-time insights. This enables the high-fidelity DT to accurately mirror current system operations and forecast future scenarios. Consequently, the PARS platform becomes an ideal environment for testing and refining monitoring, control, power, and energy management algorithms designed to boost hybrid PV system performance. The defining feature of the PARS platform, distinguishing it from other advanced simulation tools, is its exceptional adaptability. This is achieved by employing actual network topologies and utilizing real-time field data for fine-tuning and calibration, ensuring a close emulation of real-world conditions. The project deliverables include: 1) High-fidelity IBR models and tools for real-time parameterization, utilizing real-time field measurements to refine IBR models for enhanced accuracy and performance; 2) Grid-forming and Grid-following capabilities to deliver resilience services, including blackstart, voltage and frequency support, cold-load pick-up, power reserves, and three-phase load balancing across grid-connected and microgrid settings; 3) Machine learning-based forecasting tools and methods for generating synthetic data and topologies, creating diverse and realistic simulation environments for evaluating varied operational scenarios; 4) Advanced microgrid power and energy management algorithms for optimizing the integration and operation of PV, storage, and demand response resources within both feeder and community scales. The power grid data sets are provided by four utility companies in North Carolina and the New York Power Administration. Acting as industry advisors, our industry partners communicated stakeholder needs and regulatory standards to the research teams, aiding technology transfer by incorporating the developed methodologies into their daily operations. This collaboration ensures that the PARS platform, functioning as a power system digital twin, enhances our understanding of IBR dynamic behaviors and enables the development and evaluation of IBR control functions that match or exceed the capabilities of conventional synchronous generators.
14 SOLAR ENERGY↗
PANTHER: A Programmable Architecture for Neural Network Training Harnessing Energy-Efficient ReRAM
The wide adoption of deep neural networks has been accompanied by ever-increasing energy and performance demands due to the expensive nature of training them. Additionally, numerous special-purpose architectures have been proposed to accelerate training: both digital and hybrid digital-analog using resistive RAM (ReRAM) crossbars. ReRAM-based accelerators have demonstrated the effectiveness of ReRAM crossbars at performing matrix-vector multiplication operations that are prevalent in training. However, they still suffer from inefficiency due to the use of serial reads and writes for performing the weight gradient and update step. A few works have demonstrated the possibility of performing outer products in crossbars, which can be used to realize the weight gradient and update step without the use of serial reads and writes. However, these works have been limited to low precision operations which are not sufficient for typical training workloads. Moreover, they have been confined to a limited set of training algorithms for fully-connected layers only. To address these limitations, we propose a bit-slicing technique for enhancing the precision of ReRAM-based outer products, which is substantially different from bit-slicing for matrix-vector multiplication only. We incorporate this technique into a crossbar architecture with three variants catered to different training algorithms. To evaluate our design on different types of layers in neural networks (fully-connected, convolutional, etc.) and training algorithms, we develop PANTHER, an ISA-programmable training accelerator with compiler support. Our design can also be integrated into other accelerators in the literature to enhance their efficiency. Our evaluation shows that PANTHER achieves up to 8.02×, 54.21×, and 103× energy reductions as well as 7.16×, 4.02×, and 16× execution time reductions compared to digital accelerators, ReRAM-based accelerators, and GPUs, respectively.
42 ENGINEERING↗