Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Transformer model”

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 955 records · Page 53
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Robust Automatic EXAFS First-Shell Fits

Extended X-ray absorption fine structure (EXAFS) is a widely used technique for atomic structure determination. Fourier transformation connects EXAFS in k space and R space. However, determining the appropriate k-range for the transformation can be challenging, but critical for the first-shell fit. In this study, we present an automatic method to determine the k-range using the Larch package and a Python program. The first step is to estimate spectral noise across a series of k-ranges with a fixed minimum value and identify the optimal maximum value in the k-range (k max ). The k max is determined by an empirical noise threshold that marks the point where the noise level in the Fourier transformed spectrum changes dramatically. Using the obtained k max value, the first shell is modeled to determine the minimum k value (k min ) by optimizing the background function through alignment of the spectrum with theory. The optimal k min corresponds to the point of the minimum R-factor, which quantifies the difference between the experimental and fitted spectrum. Our method was tested on various typical datasets and yielded suitable k-ranges for Fourier transformation and accurate first-shell fits. This approach helps avoid unreliable, irreproducible data analysis, especially for noisy data from diluted samples, and enables robust automatic first-shell EXAFS fitting.

EXAFS analysis↗
DOE OSTI2021

Hierarchical quark and lepton masses with sequential U(1) gauge symmetry

Instead of right-handed neutrino singlets, the standard model is extended to include lepton triplets (Σ + , Σ 0 , Σ - ). Each quark and lepton family may now transform under an anomaly-free U(1) X gauge symmetry, known already for many years. A new sequential application is presented, using just the one Higgs doublet of the standard model, together with two U(1) X Higgs singlets. The resulting structure has hierarchical quark and lepton masses, as well as a viable seesaw neutrino mass matrix.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Evaluation of Nuclear Spent Fuel Disposal in Clay-Bearing Rock - Process Model Development and Experimental Studies

The DOE R&D program under the Spent Fuel Waste Science Technology (SFWST) campaign has made key progress in modeling and experimental approaches towards the characterization of chemical and physical phenomena that could impact the long-term safety assessment of heat-generating nuclear waste disposition in deep clay/shale/argillaceous rock. International collaboration activities such as heater tests and postmortem analysis of samples recovered from these have elucidated key information regarding changes in the engineered barrier system (EBS) material exposed to years of thermal loads. Chemical and structural analyses of sampled bentonite material from such tests has as well as experiments conducted on these are key to the characterization of thermal effects affecting bentonite clay barrier performance and the extent of sacrificial zones in the EBS during the thermal period. Thermal, hydrologic, and chemical data collected from heater tests and laboratory experiments has been used in the development, validation, and calibration of THMC simulators to model near-field coupled processes. This information leads to the development of simulation approaches (e.g., continuum vs. discrete) to tackle issues related to flow and transport at various scales of the host-rock and EBS design concept. Consideration of direct disposal of large capacity dual-purpose canisters (DPCs) as part of the back-end SNF waste disposition strategy has generated interest in improving our understanding of the effects of elevated temperatures on the EBS design. This is particularly important for backfilled repository concepts where temperature plays a key role in the EBS behavior and long-term performance. This report describes multiple R&D efforts on disposal in argillaceous geologic media through development and application of coupled THMC process models, experimental studies on clay/metal/cement barrier and host-rock (argillite) material interactions, molecular dynamic (MD) simulations of water transport during (swelling) clay dehydration, first-principles studies of metaschoepite (UO 2 corrosion product) stability, and advances in thermodynamic plus surface complexation database development. Drift-scale URL experiments provides key data for testing hydrological-chemical (HC) model involving strong couplings of fluid mixing and barrier material chemical interactions. The THM modeling focuses on heater test experiments in argillite rock and gas migration in bentonite as part of international collaboration activities at underground research laboratories (URLs). In addition, field testing at an URL involves in situ analysis of fault slip behavior and fault permeability. Pore-scale modeling of gas bubble migration is also being investigated within the gas migration modeling effort. Interaction experiments on bentonite samples from heater test under ambient and elevated temperatures permit the evaluation of ion exchange, phase stability, and mineral transformation changes that could impact clay swelling. Advances in the development, testing, and implementation of a spent nuclear fuel (SNF) degradation model coupled with canister corrosion focus on the effects of hydrogen gas generation and its integration with Geologic Disposal Safety Assessment (GDSA). GDSA integration activities includes evaluation of groundwater chemistries in shale formations.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

Preconceptual Designs of Coupled Power Delivery between a 4-Loop PWR and 100-500 MWe HTSE Plants

This study develops a preconceptual design for the integration between a large-scale high-temperature electrolysis facility and a NPP. Two hydrogen facility sizes are considered: 100 MWnom and 500 MWnom. Both steam supply designs use cold reheat steam extraction as a heat source. A reboiler inside the protected area of the power plant transfers steam heat to the demineralized water supply for the hydrogen plant. After the heat transfer, the extracted steam condenses and returns to the condenser while the process steam routes out of the protected area to the electrolyzers. Electrical power is tapped off from the high voltage side of the GSU transformer, where it is then transported via a 345-kV transmission line to the hydrogen facility. Circuit breakers and disconnects are located at both ends of the transmission line. Step down transformers and miscellaneous switchgear/buses are located at the end of the transmission line inside the HTEF boundary. Control capabilities for the steam interfacing equipment and electrical dispatch are accessible from the Main Control Room, and protective relays for the transmission line are located inside the Relay Room. Computer modeling was performed for the thermal and electrical designs. PEPSE analysis provided the steady-state parameters for thermal extraction from the turbine cycle. These parameters were used to inform transients and size equipment in combination with Applied Flow Technology (AFT) Arrow and AFT Fathom modeling for steam and water piping, respectively. Electrical transients were analyzed using PSCAD. An ETAP model was used to evaluate power flow and short circuit, which enabled the sizing of transformers and protective equipment. A cost estimate was developed for both integration designs when considering plant separation distances of 250 m and 500 m. From these estimates, the modifications for thermal and electrical interfacing of a first-of-a-kind nuclear-integrated hydrogen facility are anticipated to cost between $60–250/kWnom., where the subscript “nom” refers to the nominal size of the hydrogen plant. On a thermal power basis, the thermal power has a cost of approximately $8/MWth for a 500 MWnom high temperature electrolysis plant located 500 m distant from the NPP. That value decreases to approximately $7.5/MWth for a 250 m separation distance between the hydrogen plant and the NPP. This value is lower than previous estimates of the cost of heat extracted from NPPs primarily because in this work the steam is extracted from cold reheat instead of the main steam line, which reduces the cost of the dispatched steam by approximately $3.5/MWth. Nuclear steam extraction can provide a profit avenue for many plants and is not restricted to hydrogen production. Ammonia production, oil refining, and paper production, among other industrial processes all require thermal energy, which can be provided by NPPs. Future work should look further at the details of thermal extraction for a variety of use cases. This can include increased levels of extraction and multiple simultaneous users. Additionally, site-specific studies should be performed to develop industry experience and improve cost accuracy.

08 HYDROGEN↗
DOE OSTI2023

Hydropower representation in water and energy system models: a review of divergences and call for reconciliation

Abstract Reservoir-based hydropower systems represent key interactions between water and energy systems and are being transformed under policy initiatives driven by increasing water and energy demand, the desire to reduce environmental impacts, and interacting effects of climate change. Such policies are often guided by complex system models, whereby divergence in system representations can potentially translate to incompatible planning outcomes, thereby undermining any planning that may rely on them. We review different approaches and assumptions in hydropower representation in water and energy systems. While the models and issues are relevant globally, the review focuses on applications in California given its extensive development of energy and water models for policy planning, but discusses the extent to which these observations apply to other regions. Structurally, both water-driven and energy-driven management models are similar. However, in energy models, hydropower is often represented as a single-priority output. Water management models typically allocate water for competing priorities, which are generally uninformed by dynamic electricity load demand, and often result in a lower priority for hydropower. In water models, constraints are increasingly resolved for non-energy components (e.g. inflow hydrology and non-energy water demand); few analogues exist for energy models. These limitations may result in inadequate representations of each respective sector, and vastly different planning outcomes for the same facilities between the two different sectors. These divergent modeling approaches manifest themselves in California where poorly reconciled outcomes may affect decisions in hydropower licensing, electricity grid flexibility and decarbonization, and planning for environmental water. Fully integrated water-energy models are computationally intensive and specific to certain regions, but better representation of each domain in respective efforts would help reconcile divergences in planning and management efforts related to hydropower across energy and water systems.

Rheinheimer, David E. (ORCID:0000000315259069)↗
DOE OSTI2025

Applying deep learning methods to develop new models of molecular charge transfer, nonadiabatic dynamics, and nonlinear spectroscopy in the condensed phase

Photon- and field-induced charge transfer has central importance in the generation and storage of electricity, the novel properties of materials, photo-induced catalysis, and electro-optic activity (e.g., photovoltaic cells, fuel cells, and organic chromophores for use in optical fibers and light-emission diodes). These non-equilibrium electronic and chemical transformations are probed by ultrafast, nonlinear spectroscopies. Accurate simulations play a crucial role in our ability to understand, optimize, and control these transformations. This project applies modern deep learning and machine learning (ML) methods to dramatically improve models of electronic dynamics, electronic-nuclear dynamics, and spectroscopic measurements for improved simulations of chemistry in complex environments, far from equilibrium phenomena, and processes in extreme environments, such as materials exposed to strong or resonant fields. This project develops accurate neural net models that go beyond predictive capability to also provide new insight into the fundamental physics underlying electron and nuclear dynamics. To achieve its objectives, this project explores and develops customized versions of high-capacity deep learning algorithms/models. These techniques are developed with an emphasis on fundamental chemical insight, not just predictive accuracy, to assist the development of the next generation of quantum simulation methods.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

IMoFi - Intelligent Model Fidelity: Physics-Based Data-Driven Grid Modeling to Accelerate Accurate PV Integration (Final Report)

This report summarizes the work performed under a project funded by U.S. DOE Solar Energy Technologies Office (SETO) to use grid edge measurements to calibrate distribution system models for improved planning and grid integration of solar PV. Several physics-based data-driven algorithms are developed to identify inaccuracies in models and to bring increased visibility into distribution system planning. This includes phase identification, secondary system topology and parameter estimation, meter-to-transformer pairing, medium-voltage reconfiguration detection, determination of regulator and capacitor settings, PV system detection, PV parameter and setting estimation, PV dynamic models, and improved load modeling. Each of the algorithms is tested using simulation data and demonstrated on real feeders with our utility partners. The final algorithms demonstrate the potential for future planning and operations of the electric power grid to be more automated and data-driven, with more granularity, higher accuracy, and more comprehensive visibility into the system.

14 SOLAR ENERGY↗
DOE OSTI2022

IMoFi (Intelligent Model Fidelity): Physics-Based Data-Driven Grid Modeling to Accelerate Accurate PV Integration Updated Accomplishments

This report summarizes the work performed under a project funded by U.S. DOE Solar Energy Technologies Office (SETO), including some updates from the previous report SAND2022-0215, to use grid edge measurements to calibrate distribution system models for improved planning and grid integration of solar PV. Several physics-based data-driven algorithms are developed to identify inaccuracies in models and to bring increased visibility into distribution system planning. This includes phase identification, secondary system topology and parameter estimation, meter-to-transformer pairing, medium-voltage reconfiguration detection, determination of regulator and capacitor settings, PV system detection, PV parameter and setting estimation, PV dynamic models, and improved load modeling. Each of the algorithms is tested using simulation data and demonstrated on real feeders with our utility partners. The final algorithms demonstrate the potential for future planning and operations of the electric power grid to be more automated and data-driven, with more granularity, higher accuracy, and more comprehensive visibility into the system.

14 SOLAR ENERGY↗
DOE OSTI2024

A Hydropower facility as an Energy Water Signal Processor

In recent times, various efforts have been made to address the challenge of adequately representing hydropower systems in modeling frameworks, accounting for the lack of data to represent the multiple constraints in hydropower operation. This research is a pilot data-driven methodology for characterizing, classifying, and comparing the water-to-energy and energy-to-water signal transformations that hydropower facilities as signal processors accomplish. In this study, a Box Jenkins transfer function/noise model is used to identify the relationship between reservoir inflows and outflows. For examining the feasibility of this methodology, 5-minute fleet data for five storage and five run-of-river facilities was provided by the Tennessee Valley Authority (TVA) and transfer function models are developed. The influence of past inflow and outflow values on the current outflow decisions was investigated and summarized by examining the results of Box Jenkins methodology. Finally, dominance analysis was introduced to add value to the Box Jenkins model results and provide different stakeholders with a set of concepts to convey the functionality of hydropower.

13 HYDRO ENERGY↗
DOE OSTI2026

Least-cost Optimal Distribution Grid Expansion (LODGE): Utility Pilots

The Least-cost Optimal Distribution Grid Expansion (LODGE) model provides the optimal portfolio of distribution system upgrades—e.g., voltage regulators, feeder reconductoring, transformer upgrades and non-wires alternatives (NWA), such as strategic siting of storage and distributed generation—to interconnect distributed energy resources (DERs) and enable load growth. It can be used to assess grid infrastructure costs and explore policy and regulatory solutions for distribution planning and DER valuation.In 2025, Berkeley Lab conducted three pilot analyses to validate LODGE results with empirical utility data before the model’s first release in 2026. The pilots, done with utilities in Washington, Colorado, and New Mexico, provide examples that illustrate how the model works, what it can do, and the value of the analysis.

Heleno, Miguel↗
DOE OSTI2021

Data-driven Optimal Control Strategy for Virtual Synchronous Generator via Deep Reinforcement Learning Approach

This paper aims at developing a data-driven optimal control strategy for virtual synchronous generator (VSG) in the scenario where no expert knowledge or requirement for system model is available. Firstly, the optimal and adaptive control problem for VSG is transformed into a reinforcement learning task. Specifically, the control variables, i.e., virtual inertia and damping factor, are defined as the actions. Meanwhile, the active power output, angular frequency and its derivative are considered as the observations. Moreover, the reward mechanism is designed based on three preset characteristic functions to quantify the control targets: (1) maintaining the deviation of angular frequency within special limits; (2) preserving well-damped oscillations for both the angular frequency and active power output; (3) obtaining slow frequency drop in the transient process. Next, to maximize the cumulative rewards, a decentralized deep policy gradient algorithm, which features model-free and faster convergence, is developed and employed to find the optimal control policy. With this effort, a data-driven adaptive VSG controller can be obtained. By using the proposed controller, the inverter-based distributed generator can adaptively adjust its control variables based on current observations to fulfill the expected targets in model-free fashion. Finally, simulation results validate the feasibility and effectiveness of the proposed approach.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

FFT-based model for irradiated aggregate microstructures in concrete

The concrete biological shield of light water reactors is exposed to neutron and gamma irradiation throughout its lifetime, which results in the long-term degradation of the concrete’s mechanical properties. Under neutron irradiation, the concrete’s aggregates are subjected to radiation-induced volumetric expansion (RIVE), which strongly depends on the mineral content of the aggregate and exhibits the largest expansion in silicate-bearing minerals. In this work, the authors used the fast Fourier transform-based code Microstructure-Oriented Scientific Analysis of Irradiated Concrete (MOSAIC) in 2D to model the expansion of five different aggregates provided by the Japan Concrete Aging Management Program (JCAMP). Comparable rock specimens were irradiated at the JEEP-II test reactor. The model uses realistic aggregate microstructure reconstruction based on high-resolution characterization images. The model accounts for anisotropic RIVE, thermal expansion, and the associated initiation and propagation of damage. The RIVE models are calibrated based on expansion data in the literature. The authors assume that damage occurs exclusively at interfaces between the particles that compose an aggregate and that these interfaces also exhibit swelling. Using a micromechanical model, the evolution of Young’s modulus with RIVE is calculated for each aggregate and compared with Russian irradiation data. The modeled linear expansion agrees well with the experimentally measured expansion. Furthermore, the model also predicts that anisotropic RIVE and thermal expansion result in an earlier onset of damage with neutron fluence than in the isotropic case.

36 MATERIALS SCIENCE↗
DOE OSTI2023

UVCGAN: UNet Vision Transformer cycle-consistent GAN for unpaired image-to-image translation

Unpaired image-to-image translation has broad applications in art, design, and scientific simulations. One early breakthrough was CycleGAN that emphasizes one-to-one mappings between two unpaired image domains via generative-adversarial networks (GAN) coupled with the cycle-consistency constraint, while more recent works promote one-to-many mapping to boost diversity of the translated images. Motivated by scientific simulation and one-to-one needs, this work revisits the classic CycleGAN framework and boosts its performance to outperform more contemporary models without relaxing the cycle-consistency constraint. To achieve this, we equip the generator with a Vision Transformer (ViT) and employ necessary training and regularization techniques. Compared to previous best-performing models, our model performs better and retains a strong correlation between the original and translated image. An accompanying ablation study shows that both the gradient penalty and self-supervised pre-training are crucial to the improvement. To promote reproducibility and open science, the source code, hyperparameter configurations, and pre-trained model are available at https: //github.com/LS4GAN/uvcgan.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Variability and Diversity Load Model Tool [SWR-20-03]

The motivation for the development of this tool and the underlying algorithms and methods was to enable the development of high-temporal resolution, realistic time-series data for quasi-static time-series (QSTS) analysis of distribution systems. Often, aggregated load profile data for a distribution circuit is available (e.g. feeder loading data collected via SCADA at the utility substation) and, while this data is typically accurate it masks the considerable variability of the 100’s or 1000’s of individual loads connected on the circuit. This tool was developed to model both the increased variability expected for these individual loads (e.g. the load of a single distribution transformer connected to 8-12 houses) and the expected diversity between loads on the circuit. It is important to note that the difference in variability and diversity, in the context of this tool, is that variability modeling only adds representative variability due to disaggregated load characteristics (e.g. the presence in the load profile of loads turning off and on like an air conditioner/oven) while the average energy profile remains the same as the user supplied power profile. Diversity modeling generates multiple individual load profiles which, in aggregate, sum to the user supplied power profile. Diversity is effectively variability in the energy usage over longer periods of time than seen in the variability model. Put another way, variability modeling supplies the expected variability due to the operation of various end-use loads and diversity modeling supplies the usage differences due to human behavior, schedules, etc. This load modeling tool was developed for use in generating data for distribution systems. Modeling is summarized by two major functions: 1) taking low resolution load profiles and adding intra-seconds variability onto the profiles, and 2) taking a user supplied load profile and distribution factors and adding both diversity and variability to the user supplied profile.

Zhu, Xiangqi↗
DOE OSTI2025

Intercomparison of Deep Learning Model Architectures for Atmospheric River Prediction

With a rapid surge in the application of machine learning (ML) for a diverse range of tasks in climate science, the present study addresses a challenge for climate scientists when selecting the optimal ML or deep learning (DL) architecture for a given application. In particular, a DL intercomparison study was performed with a focus on forecasting the position of atmospheric rivers (ARs) on short-range time scales (up to 5-day lead times). AR predictions from multiple DL architectures, including various types of convolutional autoencoders and a vision transformer (ViT), were compared against ECMWF ERA5 reanalysis and hindcasts from a global climate model. DL models with similar trainable parameters were trained on ERA5 reanalysis data and AR positions derived from a thresholding algorithm to ensure a fair comparison among the DL models. Each model’s performance and accuracy in forecasting AR location and key input fields within a 5-day window were assessed using metrics of root-mean-square error, anomaly correlation, and mean intersection over union. The ViT architecture outperformed other autoencoder models in most of the metrics. Incorporating additional meteorological fields only yielded slight improvements in forecasting certain fields at longer lead times. The results also suggest that a smaller number of input time steps or smaller number of autoregressive steps can achieve better prediction skills, while also improving the overall computational efficiency. This research offers valuable insights into the strengths and weaknesses of different DL techniques for AR forecasting, hopefully guiding the development of improved models for forecasting this phenomenon.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Variational Adiabatic Gauge Transformation on Real Quantum Hardware for Effective Low-Energy Hamiltonians and Accurate Diagonalization

Effective low-energy theories represent powerful theoretical tools to reduce the complexity in modeling interacting quantum many-particle systems. However, common theoretical methods rely on perturbation theory, which limits their applicability to weak interactions. Here we introduce the Variational Adiabatic Gauge Transformation (VAGT), a nonperturbative hybrid quantum algorithm that can use nowadays quantum computers to learn the variational parameters of the unitary circuit that brings the Hamiltonian to either its block-diagonal or full-diagonal form. If a Hamiltonian can be diagonalized via a shallow quantum circuit, then VAGT can learn the optimal parameters using a polynomial number of runs. The accuracy of VAGT is tested through numerical simulations, as well as simulations on Rigetti and IonQ quantum computers.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Distributed Cross-Channel Hierarchical Aggregation for Foundation Models

Vision-based scientific foundation models hold significant promise for advancing scientific discovery and innovation. This potential stems from their ability to aggregate images from diverse sources—such as varying physical groundings or data acquisition systems—and to learn spatio-temporal correlations using transformer architectures. However, tokenizing and aggregating images can be compute-intensive, a challenge not fully addressed by current distributed methods. In this work, we introduce the Distributed Cross-Channel Hierarchical Aggregation (D-CHAG) approach designed for datasets with a large number of channels across image modalities. Our method is compatible with any model-parallel strategy and any type of vision transformer architecture, significantly improving computational efficiency. We evaluated D-CHAG on hyperspectral imaging and weather forecasting tasks. When integrated with tensor parallelism and model sharding, our approach achieved up to a 75% reduction in memory usage and more than doubled sustained throughput on up to 1,024 AMD GPUs on the Frontier Supercomputer.

Tsaris, Aristeidis (aris) [ORNL] (ORCID:0000000277↗