Results for “data fidelity” — Page 6

DOE OSTI2022

Detection of multi-reference character imbalances enables a transfer learning approach for virtual high throughput screening with coupled cluster accuracy at DFT cost

Appropriately identifying and treating molecules and materials with significant multi-reference (MR) character is crucial for achieving high data fidelity in virtual high-throughput screening (VHTS). Despite development of numerous MR diagnostics, the extent to which a single value of such a diagnostic indicates the MR effect on a chemical property prediction is not well established. We evaluate MR diagnostics for over 10 000 transition-metal complexes (TMCs) and compare to those for organic molecules. We observe that only some MR diagnostics are transferable from one chemical space to another. By studying the influence of MR character on chemical properties (i.e., MR effect) that involve multiple potential energy surfaces (i.e., adiabatic spin splitting, ΔE H–L , and ionization potential, IP), we show that differences in MR character are more important than the cumulative degree of MR character in predicting the magnitude of an MR effect. Motivated by this observation, we build transfer learning models to predict CCSD(T)-level adiabatic ΔE H–L and IP from lower levels of theory. By combining these models with uncertainty quantification and multi-level modeling, we introduce a multi-pronged strategy that accelerates data acquisition by at least a factor of three while achieving coupled cluster accuracy (i.e., to within 1 kcal mol –1 MAE) for robust VHTS.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

High Fidelity Simulations of Air-Cooled Reactor Cavity Cooling System

High Temperature Gas Reactor (HTGR) designs incorporate passive safety systems (e.g., the Reactor Cavity Cooling System [RCCS]) that utilize natural principles to manage heat dissipation from the reactor pressure vessel (RPV) during accidents or routine shutdowns. The industry community is experiencing a pressing need for advanced simulation tools that can accurately assess the performance of these types of systems. In the literature, a knowledge gap exists concerning high-fidelity data for the RCCS, and this gap is one of the areas of focus of the present study. This research focuses on a specific RCCS designed for General Atomics' Modular High-Temperature Gas Reactor (GA-MHTGR). Experimental studies on a scaled version (can be seen in Figure 1) of the air-cooled RCCS used in GA-MHTGR were conducted by the University of Wisconsin-Madison (UW-Madison). This work contributes to a broader initiative aimed at establishing a numerical benchmark based on the UW-Madison experiments.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

NASA NTRSDate not supplied

NASA Delay Tolerant Networks: Operational, Evolving, an Ready for Expansion

The future of humanity’s presence beyond Earth depends on the successful commercialization of space. For commercialization to succeed, companies need cost-efficient architectures to support their business models and minimize risks for human capital, design, development, and operations. An ongoing challenge to any space enterprise is the reality that terrestrial network technologies are insufficient to provide reliable communications between assets in space. Whether you need to ensure your valuable data is safely transmitted to the ground or reliably delivered between platforms in orbit, ensuring data integrity over intermittent communication links is a necessity. Current solutions to space communications rely heavily on manual recording, storing, and retrieval of data from spacecraft. The current standard in space communication protocols, Consultative Committee for Space Data Systems (CCSDS) Space Packet standard, is reliant on inflexible network architectures based around mission-critical infrastructure to ensure data delivery. However, by automating the recording, storing, retrieval, and verification of data with Delay Tolerant Networks (DTN), the operator is freed from the dependence on manual data management and expensive mission critical infrastructure. NASA has been developing delay tolerant systems since the late 1990’s. Multiple DTN implementations have been established during that time, each suited to different use cases. Most notably, the DTN deployment for the International Space Station (ISS) includes demonstration of two DTN technologies: Interplanetary Overlay Network (ION) and Delay Tolerant Network Marshall Enterprise (DTNME). Beyond ISS, there are even more NASA DTN deployments being considered. Now that DTN implementations are maturing, it is appropriate to reflect upon these decades of work, review the integration and performance of the existing ISS deployment, and explore the future possibilities for DTN deployment industry-wide. The ISS DTN deployment is a complex architecture consisting of different DTN implementations for the onboard and ground network environments. The ION DTN implementation is being used in the on-board network. The Huntsville Operations Support Center (HOSC) DTN implementation, DTNME, is used by the ground network supporting ISS and will soon be a second onboard gateway too. The two implementations work cooperatively to provide high fidelity data services to flight operations users and payload developers across the globe. Though the two implementations yield a quality service, limitations are evident. Data rate, data storage, and device management are constrained by the services themselves and the complex nature of the deployment. Evolution of operations concepts will improve system capabilities and stability, but significant improvement will require additional development to the implementations themselves and to the overall deployment architecture. Taking advantage of the ongoing development and operation of the ISS DTN service will be central to the success of the future evolutions of NASA DTN deployments while demonstrating the benefits of DTN’s low-cost reliable data communication protocols for the growing commercial space industry. A broad effort on DTN integration and support is necessary to promote expansion beyond existing applications. NASA is developing several useful DTN implementations across a number of different systems: ION, DTNME, High-Rate DTN (HDTN), Bundle Protocol Library (BPLib), and others. To prevent fragmentation, DTN implementation teams need to communicate, collaborate, and integrate with one another to build a solid operational foundation for new DTN deployments. The establishment of a group that can assist new DTN users with understanding the purpose of each DTN implementation, provide best practices, and serve as a general knowledge base is paramount. Potential use of DTN on Gateway and other future NASA missions further drives the need for streamlined communication between DTN implementation teams. A well-integrated and highly engaged NASA DTN working group should help provide system architects the best DTN solutions for future commercial space efforts. This paper will first review the history of DTN implementations, explore the shortcoming of current space networking solutions given available limits in technology, and therefore establish the need for Delay Tolerant Networking in space communications. Secondly, the authors will explore NASA’s array of DTN implementations and highlight their usefulness to space applications. Thirdly, this paper will establish general DTN implementation distinguishing factors. Fourthly, the authors will discuss attempts to create a generic DTN comparison matrix, and the authors will review potential future topics in DTN innovation and collaboration, highlighting several key future efforts. Finally, this paper will describe how the institution of a NASA DTN Working Group will benefit DTN adoption across the governmental and commercial space sector. The goal of this paper is to encourage enthusiasm for DTN, share strategies for improving DTN on both current and future applications, promote the collaboration of DTN implementation groups within the international space operations community, and open the conversations about DTN, priorities, complexities, and innovation to the wider spaceflight industry.

DTN↗

NASA NTRSDate not supplied

NASA Delay Tolerant Networks: Operational, Evolving, an Ready for Expansion

The future of humanity’s presence beyond Earth depends on the successful commercialization of space. For commercialization to succeed, companies need cost-efficient architectures to support their business models and minimize risks for human capital, design, development, and operations. An ongoing challenge to any space enterprise is the reality that terrestrial network technologies are insufficient to provide reliable communications between assets in space. Whether you need to ensure your valuable data is safely transmitted to the ground or reliably delivered between platforms in orbit, ensuring data integrity over intermittent communication links is a necessity. Current solutions to space communications rely heavily on manual recording, storing, and retrieval of data from spacecraft. The current standard in space communication protocols, Consultative Committee for Space Data Systems (CCSDS) Space Packet standard, is reliant on inflexible network architectures based around mission-critical infrastructure to ensure data delivery. However, by automating the recording, storing, retrieval, and verification of data with Delay Tolerant Networks (DTN), the operator is freed from the dependence on manual data management and expensive mission critical infrastructure. NASA has been developing delay tolerant systems since the late 1990’s. Multiple DTN implementations have been established during that time, each suited to different use cases. Most notably, the DTN deployment for the International Space Station (ISS) includes demonstration of two DTN technologies: Interplanetary Overlay Network (ION) and Delay Tolerant Network Marshall Enterprise (DTNME). Beyond ISS, there are even more NASA DTN deployments being considered. Now that DTN implementations are maturing, it is appropriate to reflect upon these decades of work, review the integration and performance of the existing ISS deployment, and explore the future possibilities for DTN deployment industry-wide. The ISS DTN deployment is a complex architecture consisting of different DTN implementations for the onboard and ground network environments. The ION DTN implementation is being used in the on-board network. The Huntsville Operations Support Center (HOSC) DTN implementation, DTNME, is used by the ground network supporting ISS and will soon be a second onboard gateway too. The two implementations work cooperatively to provide high fidelity data services to flight operations users and payload developers across the globe. Though the two implementations yield a quality service, limitations are evident. Data rate, data storage, and device management are constrained by the services themselves and the complex nature of the deployment. Evolution of operations concepts will improve system capabilities and stability, but significant improvement will require additional development to the implementations themselves and to the overall deployment architecture. Taking advantage of the ongoing development and operation of the ISS DTN service will be central to the success of the future evolutions of NASA DTN deployments while demonstrating the benefits of DTN’s low-cost reliable data communication protocols for the growing commercial space industry. A broad effort on DTN integration and support is necessary to promote expansion beyond existing applications. NASA is developing several useful DTN implementations across a number of different systems: ION, DTNME, High-Rate DTN (HDTN), Bundle Protocol Library (BPLib), and others. To prevent fragmentation, DTN implementation teams need to communicate, collaborate, and integrate with one another to build a solid operational foundation for new DTN deployments. The establishment of a group that can assist new DTN users with understanding the purpose of each DTN implementation, provide best practices, and serve as a general knowledge base is paramount. Potential use of DTN on Gateway and other future NASA missions further drives the need for streamlined communication between DTN implementation teams. A well-integrated and highly engaged NASA DTN working group should help provide system architects the best DTN solutions for future commercial space efforts. This paper will first review the history of DTN implementations, explore the shortcoming of current space networking solutions given available limits in technology, and therefore establish the need for Delay Tolerant Networking in space communications. Secondly, the authors will explore NASA’s array of DTN implementations and highlight their usefulness to space applications. Thirdly, this paper will establish general DTN implementation distinguishing factors. Fourthly, the authors will discuss attempts to create a generic DTN comparison matrix, and the authors will review potential future topics in DTN innovation and collaboration, highlighting several key future efforts. Finally, this paper will describe how the institution of a NASA DTN Working Group will benefit DTN adoption across the governmental and commercial space sector. The goal of this paper is to encourage enthusiasm for DTN, share strategies for improving DTN on both current and future applications, promote the collaboration of DTN implementation groups within the international space operations community, and open the conversations about DTN, priorities, complexities, and innovation to the wider spaceflight industry.

DTN↗

DOE OSTI2023

BeyondPlanck: VII. Bayesian estimation of gain and absolute calibration for cosmic microwave background experiments

We present a Bayesian calibration algorithm for cosmic microwave background (CMB) observations as implemented within the global end-to-end BEYONDPLANCK framework and applied to the Planck Low Frequency Instrument (LFI) data. Following the most recent Planck analysis, we decomposed the full time-dependent gain into a sum of three nearly orthogonal components: one absolute calibration term, common to all detectors, one time-independent term that can vary between detectors, and one time-dependent component that was allowed to vary between one-hour pointing periods. Each term was then sampled conditionally on all other parameters in the global signal model through Gibbs sampling. The absolute calibration is sampled using only the orbital dipole as a reference source, while the two relative gain components were sampled using the full sky signal, including the orbital and Solar CMB dipoles, CMB fluctuations, and foreground contributions. We discuss various aspects of the data that influence gain estimation, including the dipole-polarization quadrupole degeneracy and processing masks. Comparing our solution to previous pipelines, we find good agreement in general, with relative deviations of -0.67% (-0.84%) for 30 GHz, 0.12% (-0.04%) for 44 GHz and -0.03% (-0.64%) for 70 GHz, compared to Planck PR4 and Planck 2018, respectively. We note that the BEYONDPLANCK calibration was performed globally, which results in better inter-frequency consistency than previous estimates. Additionally, WMAP observations were used actively in the BEYONDPLANCK analysis, which both breaks internal degeneracies in the Planck data set and results in an overall better agreement with WMAP. Finally, we used a Wiener filtering approach to smoothing the gain estimates. We show that this method avoids artifacts in the correlated noise maps as a result of oversmoothing the gain solution, which is difficult to avoid with methods like boxcar smoothing, as Wiener filtering by construction maintains a balance between data fidelity and prior knowledge. Although our presentation and algorithm are currently oriented toward LFI processing, the general procedure is fully generalizable to other experiments, as long as the Solar dipole signal is available to be used for calibration.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2021

Machine Learning Assisted Safety Modeling and Analysis of Advanced Reactors

With the advances in computational power and numerical methods, analysts can now rely on first-principle simulations to predict ultra-fine details in a variety of applications. Advances in machine learning (ML) have produced algorithms that can now learn high-level abstractions via hierarchical models. This project aims to leverage advances in ML techniques and the available high-resolution simulation data to develop a novel modeling and simulation (M\&S) methodology for reactor safety analysis. While application-agnostic ML techniques are available, complex physics constraints need to be incorporated into ML techniques to build ML-based closures for computationally efficient predictive simulations. This project intends to develop a physics-guided data-driven multi-scale methodology for M\&S of advanced reactors. The project focuses on thermal fluid (T/F) phenomena, which play major roles in advanced reactor safety. Specifically, we propose a data-driven coarse-mesh turbulence model based on local flow features for the transient analysis of thermal mixing and stratification in a sodium-cooled fast reactor (SFR). The model has a coarse-mesh setup to ensure computational efficiency, while it is trained by fine-mesh computational fluid dynamics (CFD) data with Reynolds-averaged Navier-Stokes (RANS) turbulence model to ensure accuracy. Three different neural networks are developed and tested for loss-of-flow transients in the hot pool of SFR, i.e. the densely connected convolutional neural network (DCNN), long-short-term-memory network based on proper orthogonal decomposition (POD-LSTM), and the DCNN informed by LSTM (DCNN-LSTM). The performances of these three neural networks are evaluated based on baseline models. The DCNN-LSTM model has been chosen for further hyperparameter optimization. Furthermore, based on a simplified two-dimensional case, uncertainty quantification (UQ) of the developed ML-based closure are investigated with three methods, i.e. Monte Carlo dropout, deep ensemble, and Bayesian neural network. The developed ML-based turbulent viscosity closure relation based on deep ensemble is then integrated into the system analysis module SAM and serves as a term in the conservation equations. Such a SAM-ML based procedure guarantees that the obtained results are consistent with the physical constraints of the thermal-fluid system. The SAM-ML simulation on the same loss-of-flow transient showed comparable accuracy with the CFD simulation but with a much coarser mesh setup. Last but not least, the ML-based closure improvement with the support of higher-fidelity data from large eddy simulation (LES) is discussed. As a first step towards this direction, a baseline LES simulation is performed to obtain comparable data with RANS results. Based on the early results, future investigation on further improving the ML-based closure is discussed. We believe the developed approach that combines scientific machine learning with nuclear system analysis code can benefit the advanced reactor community as more accurate safety analyses will better characterize reactor safety margins and reduce licensing efforts.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2020

Data processing and quality verification for improved photovoltaic performance and reliability analytics

Not Available

14 SOLAR ENERGY↗

DOE OSTI2026

Deep learning–based digital twins for heat pumps

Heat pumps are effective cooling and heating appliances to save energy in buildings. However, traditional heat pump models are challenging to integrate with building demands in a co-simulation environment because of the nonlinear thermodynamics of refrigerants. Developing digital twin representatives for heat pumps capable of faster calculations with good accuracy is desirable. This study aimed to establish a generic deep learning–based digital twin for heat pumps with a large amount of high-fidelity data. Two refrigerants for two different heat pumps were considered: an air source heat pump with refrigerant R-410A, an air source heat pump with refrigerant CO 2 , a water source heat pump with refrigerant R-410A, and a water source heat pump with refrigerant CO 2 . Furthermore, results showed that the deep learning (long short-term memory) models effectively represented these four heat pumps as a digital twin: (a) accuracy for training and testing showed smaller than 0.02 for heating electricity and heating demands, and (b) the digital twins showed good consistency with original data for heating electricity and heating demands (root mean square errors of less than 0.12 W and 0.19 W, respectively). Therefore, deep learning–based heat pump models can be used in the co-simulation of building mechanical systems.

Air source heat pump↗

DOE OSTI2026

Real-time plasma monitoring framework for advanced plasma control and ML-research in DIII-D

Real-time and adaptive plasma control is crucial for robust tokamak operation, requiring sensitivity and tolerance measurements of the plasma state. This paper presents the implementation of an integrated real-time plasma monitoring framework on the DIII-D tokamak to support advanced control approaches, including machine-learning (ML) methods. The system is built on the SHIELD framework, a high-performance modular architecture that provides a unified pipeline for integrating diverse diagnostics. The framework leverages high-bandwidth digitizers, fast numerical processing, and deterministic, low-latency interconnects to stream high-fidelity data from diagnostics such as electron cyclotron emission (ECE), beam emission spectroscopy (BES), CO interferometers, and a visible tangential divertor camera (TangTV). The system’s validity is demonstrated through direct comparisons of real-time and offline data. Furthermore, we present two key applications of the developed plasma monitoring system with ML-based plasma control strategies, including real-time divertor detachment and active Alfvén Eigenmode control. As a result, this work presents a robust and scalable approach for integrating high-frequency, multidimensional diagnostics into advanced control algorithms for future fusion devices.

AI/ML↗

DOE OSTI2020

NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations

In the last 50 years there has been a tremendous progress in solving numerically the Navier-Stokes equations using finite differences, finite elements, spectral, and even meshless methods. Yet, in many real cases, we still cannot incorporate seamlessly (multi-fidelity) data into existing algorithms, and for industrial-complexity applications the mesh generation is time consuming and still an art. Moreover, solving ill-posed problems (e.g., lacking boundary conditions) or inverse problems is often prohibitively expensive and requires different formulations and new computer codes. Here, we employ physics-informed neural networks (PINNs), encoding the governing equations directly into the deep neural network via automatic differentiation, to overcome some of the aforementioned limitations for simulating incompressible laminar and turbulent flows. We develop the Navier-Stokes flow nets (NSFnets) by considering two different mathematical formulations of the Navier-Stokes equations: the velocity-pressure (VP) formulation and the vorticity-velocity (VV) formulation. Since this is a new approach, we first select some standard benchmark problems to assess the accuracy, convergence rate, computational cost and flexibility of NSFnets; analytical solutions and direct numerical simulation (DNS) databases provide proper initial and boundary conditions for the NSFnet simulations. The spatial and temporal coordinates are the inputs of the NSFnets, while the instantaneous velocity and pressure fields are the outputs for the VP-NSFnet, and the instantaneous velocity and vorticity fields are the outputs for the VV-NSFnet. This is unsupervised learning and, hence, no labeled data are required beyond boundary and initial conditions and the fluid properties. The residuals of the VP or VV governing equations, together with the initial and boundary conditions, are embedded into the loss function of the NSFnets. No data is provided for the pressure to the VP-NSFnet, which is a hidden state and is obtained via the incompressibility constraint without extra computational cost. Unlike the traditional numerical methods, NSFnets inherit the properties of neural networks (NNs), hence the total error is composed of the approximation, the optimization, and the generalization errors. Here, we empirically attempt to quantify these errors by varying the sampling (“residual”) points, the iterative solvers, and the size of the NN architecture. For the laminar flow solutions, we show that both the VP and the VV formulations are comparable in accuracy but their best performance corresponds to different NN architectures. The initial convergence rate is fast but the error eventually saturates to a plateau due to the dominance of the optimization error. For the turbulent channel flow, we show that NSFnets can sustain turbulence at , but due to expensive training we only consider part of the channel domain and enforce velocity boundary conditions on the subdomain boundaries provided by the DNS data base. We also perform a systematic study on the weights used in the loss function for balancing the data and physics components, and investigate a new way of computing the weights dynamically to accelerate training and enhance accuracy. In the last part, we demonstrate how NSFnets should be used in practice, namely for ill-posed problems with incomplete or noisy boundary conditions as well as for inverse problems. We obtain reasonably accurate solutions for such cases as well without the need to change the NSFnets and at the same computational cost as in the forward well-posed problems. As a result, we also present a simple example of transfer learning that will aid in accelerating the training of NSFnets for different parameter settings.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Deep learning of experimental electrochemistry for battery cathodes across diverse compositions

Artificial intelligence (AI) has emerged as a tool for discovering and optimizing novel battery materials. However, the adoption of AI in battery cathode representation and discovery is still limited due to the complexity of optimizing multiple performance properties and the scarcity of high-fidelity data. Here, we present a machine learning model (DRXNet) for battery informatics and demonstrate the application in the discovery and optimization of disordered rocksalt (DRX) cathode materials. We have compiled the electrochemistry data of DRX cathodes over the past 5 years, resulting in a dataset of more than 19,000 discharge voltage profiles on diverse chemistries spanning 14 different metal species. Learning from this extensive dataset, our DRXNet model can capture critical features in the cycling curves of DRX cathodes under various conditions. Our approach offers a data-driven solution to facilitate the rapid identification of novel cathode materials, accelerating the development of next-generation batteries for carbon neutralization.

25 ENERGY STORAGE↗

DOE OSTI2020

DDxNet: a deep learning model for automatic interpretation of electronic health records, electrocardiograms and electroencephalograms

Effective patient care mandates rapid, yet accurate, diagnosis. With the abundance of non-invasive diagnostic measurements and electronic health records (EHR), manual interpretation for differential diagnosis has become time-consuming and challenging. This has led to wide-spread adoption of AI-powered tools, in pursuit of improving accuracy and efficiency of this process. While the unique challenges presented by each modality and clinical task demand customized tools, the cumbersome process of making problem-specific choices has triggered the critical need for a generic solution to enable rapid development of models in practice. In this spirit, we develop DDxNet, a deep architecture for time-varying clinical data, which we demonstrate to be well-suited for diagnostic tasks involving different modalities (ECG/EEG/EHR), required level of characterization (abnormality detection/phenotyping) and data fidelity (single-lead ECG/22-channel EEG). Using multiple benchmark problems, we show that DDxNet produces high-fidelity predictive models, and sometimes even provides significant performance gains over problem-specific solutions.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2024

High-Fidelity Velocity and Concentration Measurements of Turbulent Buoyant Jets

Accurate models of turbulent buoyant flows are essential for the design of the cooling circuit of nuclear reactors and passive safety systems. However, available models fail to fully capture the physics of turbulent mixing when buoyancy becomes predominant with respect to momentum. Therefore, high-fidelity experiments of well-controlled fundamental flows are needed to develop and validate more accurate models. We analyze experiments of positive and negative turbulent buoyant jets, both in uniform and stratified environments, with the aim of understanding the thermal hydraulics of turbulent mixing with variable density and providing high-fidelity data for the development and validation of turbulence models. Non-intrusive, simultaneous particle image velocimetry and laser-induced fluorescence measurements were carried out to acquire instantaneous velocity and concentration fields on a vertical section parallel to the axis of a jet in the self-similar region. The refractive index matching method was applied to measure high-resolution buoyant jets with up to 8.6% density difference. These data are free of the typical errors that characterize optical measurements of buoyancy-driven flows (e.g. natural and mixed convection) where the refractive index of the fluid is inhomogeneous throughout the measurement domain. Turbulent statistics and entrainment of buoyant jets in uniform and stratified environments are presented. These data are compared with non-buoyant jets in a uniform environment, as a reference to investigate the effects of buoyancy and stratification on turbulent mixing. The results will be used for the assessment of current turbulence models and as a basis for the development of a new model that captures turbulent mixing.

laser induced fluorescence↗

DOE OSTI2025

Revealing the Structure and Dynamics of Self-Generated Electric and Magnetic Fields Near Plasma Stagnation in Laser-Driven Hohlraums

By coupling newly developed triparticle charged particle radiography with radiography reconstruction algorithms and novel reconstruction postprocessing techniques, the spatial structure and time evolution of self-generated electric and magnetic fields in laser-driven vacuum hohlraums have been quantitatively revealed. Through high-fidelity data from a series of experiments, it is shown that late in the hohlraum evolution (after the end of laser drive) these fields are strongly correlated in both space and time, providing evidence that their evolution is primarily dominated by advection with the plasma flow. At these late times, plasma flow velocities inferred from both gross radiography analysis and field reconstructions (and corroborated with Thomson scattering measurements) indicate that the plasma is approaching stagnation near the hohlraum axis. Finally, these experiments provide not only new physical insight into spontaneously generated hohlraum fields, but also provide important spatially and temporally resolved information for future benchmarking of numerical codes.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2026

Bayesian Physics Informed Spatio-Temporal Network for Streamflow Data Imputation

Reliable reconstruction of incomplete streamflow records is critical for improving hydrological forecasting, flood preparedness, and water resource management. However, large observational gaps and uncertainties in governing physical parameters limit the accuracy of traditional statistical and machinelearning imputation frameworks. To address these challenges, we develop a Bayesian Physics-Informed Spatio-Temporal Network (BPI-STNet) that jointly captures spatial and temporal dependencies while enforcing hydrologic consistency through embedded physical constraints. The framework integrates a GraphSAGE-LSTM architecture to model spatial connectivity across gauges and temporal flow dynamics, coupled with a Bayesian update mechanism to estimate uncertain parameters in a simplified water-balance framework. Unlike conventional physics-informed networks that rely on sampling-based posterior estimation, BPI-STNet derives an analytic solution to the inverse problem, allowing closed-form Bayesian updates of uncertain parameters Λ={α,β,k} using Gaussian priors and likelihoods. Applied to daily observations from the Susquehanna River Basin (1980-2022), BPI-STNet achieves substantial improvements over a purely data-driven RGNN baseline, which reduced RMSE by 23 % and MAE by 9 %, and achieving an average NSE values up to 0.96. The results demonstrate that coupling Bayesian inference with physics-informed learning yields physically consistent, uncertainty-aware reconstructions that preserve the temporal persistence and statistical distribution of observed flows. The proposed framework establishes a generalizable paradigm for data-sparse hydrologic systems where both data fidelity and physical interpretability are essential.

Krishnan Kutty Ambika, Anukesh [ORNL] (ORCID:00000↗

DOE OSTI2021

Scalable Plug-and-Play ADMM with Convergence Guarantees

Plug-and-play priors (PnP) is a broadly applicable methodology for solving inverse problems by exploiting statistical priors specified as denoisers. Recent work has reported the state-of-the-art performance of PnP algorithms using pre-trained deep neural nets as denoisers in a number of imaging applications. However, current PnP algorithms are impractical in large-scale settings due to their heavy computational and memory requirements. This work addresses this issue by proposing an incremental variant of the widely used PnP-ADMM algorithm, making it scalable to problems involving a large number measurements. Here, we theoretically analyze the convergence of the algorithm under a set of explicit assumptions, extending recent theoretical results in the area. Additionally, we show the effectiveness of our algorithm with nonsmooth data-fidelity terms and deep neural net priors, its fast convergence compared to existing PnP algorithms, and its scalability in terms of speed and memory.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

JANUS: Resilient and Adaptive Data Transmission for Enabling Timely and Efficient Cross-Facility Scientific Workflows

In modern science, the growing complexity of large-scale scientific projects has led to an increasing reliance on cross-facility scientific workflows, where resources and expertise from multiple institutions and geographic locations are leveraged to accelerate scientific discovery. These workflows often require transmitting huge amounts of scientific data through wide-area networks. Although high-speed networks like ESnet and transfer services such as Globus have improved data mobility, several challenges remain. The sheer volume of data can overwhelm network bandwidth, widely used transport protocols such as TCP suffer from inefficiencies due to retransmissions triggered by packet loss, and existing fault-tolerance mechanisms like erasure coding introduce substantial overhead. In this paper, we propose Janus, a resilient and adaptable data transmission approach designed for cross-facility scientific workflows. Unlike traditional TCP-based methods, Janus leverages UDP, integrates erasure coding for fault tolerance, and combines it with error-bounded lossy compression to reduce overhead. This novel design allows users to balance data transmission time and accuracy, optimizing transfer performance based on specific scientific requirements. Additionally, Janus dynamically adjusts erasure coding parameters in response to real-time network conditions, ensuring efficient data transfers even in fluctuating environments. We develop optimization models for determining ideal configurations and implement adaptive data transfer protocols to enhance reliability. Through extensive simulations and real-network experiments, we demonstrate that Janus significantly improves transfer efficiency while maintaining data fidelity.

Esaulov, Vladislav [Georgia State University, Atla↗

DOE OSTI2025

HydraGNN v4.0

The new version of HydraGNN v4.0 provides additional core capabilities, such as: Inclusion of multi-body atomistic cluster expansion MACE, polarizable atom interaction neural network PAINN, and equivariant principal neighborhood aggregation (PNAEq) among the message passing layers supported -Inclusion of graph transformers to directly model long-range interactions between nodes that are distant in the graph topology Integration of graph transformers with message passing layers by combining the graph embedding generated by the two mechanisms, which allows for an improved expressivity of the HydraGNN architecture Improved re-implementation of multi-task learning (MTL) to allow its use for stabilized training across imbalanced, multi-source, multi-fidelity data Introduction of multi-task parallelism, a newly proposed type of model parallelism specifically for MTL architectures, which allows to dispatch different output decoding heads to different GPU devices Integration of multi-task parallelism with pre-existing distributed data parallelism to enable a 2D parallelization for distributed training Improved portability of the distributed training across Intel GPUs, which has been testes on ALCF exascale supercomputer Aurora Inclusion of 2-level fine-grained energy profilers portable across NVIDIA, AMD, and Intel GPUs to monitor the power and energy consumption associated with different functions executed by the HydraGNN code during data pre-load and training Restructuring of previous examples and inclusion of new sets of examples to illustrate the download, preprocess, and training of HydraGNN models on new large-scale open-source datasets for atomistic materials modeling (e.g., Alexandria, Transition1x, OMat24, OMol25)

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗