Search NASA⌕ Search

SEARCH · Search NASA

Results for “BASELINE REGULARIZATION”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Design, dynamic modeling and wave basin verification of a Hybrid Wave–Current Energy Converter

Massive and high-density Marine and Hydrokinetic (MHK) energy is contained in the ocean, including waves, tidal streams, and ocean currents. Traditional MHK energy converters harvest energy from only a single MHK energy source, which does not fully exploit the energy potential that co-exists in multiple forms of MHK energy in the ocean. This paper presents the design and dynamics of a Hybrid Wave–Current Energy Converter (HWCEC) that can simultaneously convert both wave and current energy to electricity with a single Power Take-off (PTO) through the engagement and disengagement of the three one-way clutches. The critical design parameters are analyzed through modeling and simulation, including the transmission ratios, the electrical impedance, and the turbine-heave plate distance. Water basin tests in a wave–current tank were conducted, which shows the prototyped HWCEC can improve the electric power output by 38%–71% for regular waves and 79% for irregular waves, while providing a 70% reduction of the Peak to Average Ratio (PAR) of power compared to the baseline Wave Energy Converter (WEC). Here we further analyze the occurring percentages of the instant power of both WEC and HWCEC in irregular wave tests, shedding new insight on how HWCEC reduces the PAR.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Surrogate model evaluation and building energy benchmarking for commercial buildings

Building energy consumption benchmarking involves challenges associated with various energy patterns for different building types; heating, ventilating, and air-conditioning (HVAC) system types; and climates. Given significant variation in energy use patterns, accurate prediction of long-term energy use using surrogate models remains challenging. Multiple linear regression (MLR) is commonly used for building energy benchmarking because of its simple structure; however, it lacks accuracy compared to other black-box models. Although many studies have compared surrogate models and offer guidance on model selection based on metrics, they do not provide detailed analysis on improving the surrogate model accuracy. In this paper, we implement a surrogate model using polynomial ridge regression (i.e., MLR with interaction terms combined with ridge regularization) for small office and retail strip mall buildings across six HVAC system types and all climate zones, for electricity and natural gas in baseline and proposed scenarios. A simulation workflow is developed using OpenStudio TM /EnergyPlus TM to generate simulation data using measures over a wide range of efficiency inputs. Enhancements based on statistical insights are used for improving the model accuracy using filters, input transformations, and change points. Surrogate models achieved average coefficient of variation of the root mean squared error (CVRMSE) values of 2.17, 1.06, 2.05, and 3.26 for proposed electricity, proposed natural gas, baseline electricity, and baseline natural gas, respectively, with enhancements reducing CVRMSE by an average of 14.9% across all combinations. We provide model interpretation via Shapley additive explanations to determine which input variables most influence energy consumption and provide supportive arguments for enhancements.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2024

Dark Matter Reconstruction in LBAI Experiments with Imperfect Data

Long-baseline atom interferometer (LBAI) experiments offer unprecedented sensitivity to ultralight scalar dark matter (DM) [1], however reconstruction of a putative DM signal with traditional frequency-domain analysis requires ``perfect data (i.e., regularly-sampled with no missing samples). In a real LBAI experiment, there will undoubtedly be imperfections in the data leading to downtime. This downtime can arise from operational considerations (e.g., maintenance), the operational environment (motion of people and animals [2] or elevators), and robustness of the experimental apparatus (e.g., bad atom launches). In this work, we investigate the impact of various downtime models on the overall DM sensitivity of such an experiment. We compare the sensitivity for each downtime model as determined by a ``compound FFT analysis to a baseline no-downtime case. We also show how much sensitivity can be regained by moving to a Lomb-Scargle frequency analysis, as in [2]. Furthermore, we demonstrate reconstruction of the DM wave s phase as well as its frequency. [1] D. Antypas, et al, ``New Horizons: Scalar and Vector Ultralight Dark Matter (2022). arXiv:2203.14915 [2] J. Carlton and C. McCabe, ``From RATs to riches: mitigating anthropogenic and synanthropic noise in atom interferometer searches for ultra-light dark matter (2023). arXiv:2308.101731

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

RADA

The code for regularized adversarial domain adaption is provided here. The RADA methodology is developed for correcting climate model biases using an adversarial learning strategy. The code here covers the data processing, the algorithm, the baseline models, and the evaluation.

Anderson, GemmaJ.↗
DOE OSTI2024

Evaluating the Effects of Detector Modeling Uncertainties on Sterile Neutrino Oscillation Analysis with the ICARUS Detector

he ICARUS T600 LArTPC detector was refurbished after an initial run at the underground LNGS labs and is currently taking data within its experimental hall at Fermilab after full commissioning. Regular data taking began in May 2021 with neutrinos from the Booster Neutrino Beam (BNB) and the Neutrinos at the Main Injector (NuMI) off-axis beam. As the far detector of the Short-Baseline Neutrino (SBN) Program, the ICARUS detector's capability in searching for both muon neutrino disappearance and electron neutrino appearance will allow for unprecedented sensitivity to light sterile neutrinos with eV-scale mass. The ultimate sensitivity of the detector to sterile neutrino oscillations depends on the understanding of the detector response and the uncertainties remaining after calibrating the detector model. This poster will review how detector-model-related uncertainties are quantified, evaluated and their impact on expected sterile neutrino oscillation sensitivities using the existing ICARUS data. The work pursued here is broadly applicable to other ICARUS analysis pathways such as neutrino-argon cross-sections and Beyond the Standard Model physics searches. It will finally discuss improvement pathways to reduce the systematic uncertainties to the level needed for the joint sterile neutrino oscillation analysis with the Short Baseline Near Detector (SBND) within the SBN Program.

43 PARTICLE ACCELERATORS↗
DOE OSTI2021

Power System Event Identification Based on Deep Neural Network With Information Loading

Online power system event identification and classification are crucial to enhancing the reliability of transmission systems. In this study, we develop a deep neural network (DNN) based approach to identify and classify power system events by leveraging real-world measurements from hundreds of phasor measurement units (PMUs) and labels from thousands of events. Two innovative designs are embedded into the baseline model built on convolutional neural networks (CNNs) to improve the event classification accuracy. First, we propose a graph signal processing based PMU sorting algorithm to improve the learning efficiency of CNNs. Second, we deploy information loading based regularization to strike the right balance between memorization and generalization for the DNN. Numerical results based on real-world dataset from the Eastern Interconnection of the U.S power transmission grid show that the combination of PMU based sorting and the information loading based regularization techniques help the proposed DNN approach achieve highly accurate event identification and classification results.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Differentiable lagrangian shock hydrodynamics with application to stable shock acceleration of density interfaces

We develop a gradient based optimization approach for the equations of compressible, Lagrangian hydrodynamics and demonstrate how it can be employed to automatically uncover strategies to control hydrodynamic instabilities arising from shock acceleration of density interfaces. Strategies for controlling the Richtmyer-Meshkov instability (RMI) are of great benefit for inertial confinement fusion (ICF) where shock interactions with many small imperfections in the density interface lead to instabilities which rapidly grow over time. These instabilities lead to mixing which, in the case of laser driven ICF, quenches the runaway fusion process ruining the potential for positive energy return. Here, we demonstrate that control of these instabilities can be achieved by optimization of initial conditions with ( > 100) parameters. Optimizing over a large parameter space like this is not possible with gradient-free optimization strategies. This requires computation of the gradient of the outputs of a numerical solution to the equations of Lagrangian hydrodynamics with respect to the inputs. We show that the efficient computation of these gradients is made possible via a judicious application of (i) adjoint methods, the exact formal representation of sensitivities involving partial differential equations, and (ii) automatic differentiation (AD), the algorithmic calculation of derivatives of functions. Careful regularization of multiple operators including artificial viscosity and timestep control is required. We perform design optimization of > 100 parameter energy field driving the Richtmyer Meshkov instability showing significant suppression while simultaneously enhancing the acceleration of the interface relative to a nominal baseline case.

Hydrophysics↗
DOE OSTI2021

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↗
DOE OSTI2021

The De-Icing Comparison Experiment (D-ICE): a study of broadband radiometric measurements under icing conditions in the Arctic

Abstract. Surface-based measurements of broadband shortwave (solar) and longwave (infrared) radiative fluxes using thermopile radiometers are made regularly around the globe for scientific and operational environmental monitoring. The occurrence of ice on sensor windows in cold environments – whether snow, rime, or frost – is a common problem that is difficult to prevent as well as difficult to correct in post-processing. The Baseline Surface Radiation Network (BSRN) community recognizes radiometer icing as a major outstanding measurement uncertainty. Towards constraining this uncertainty, the De-Icing Comparison Experiment (D-ICE) was carried out at the NOAA Atmospheric Baseline Observatory in Utqiaġvik (formerly Barrow), Alaska, from August 2017 to July 2018. The purpose of D-ICE was to evaluate existing ventilation and heating technologies developed to mitigate radiometer icing. D-ICE consisted of 20 pyranometers and 5 pyrgeometers operating in various ventilator housings alongside operational systems that are part of NOAA's Barrow BSRN station and the US Department of Energy Atmospheric Radiation Measurement (ARM) program North Slope of Alaska and Oliktok Point observatories. To detect icing, radiometers were monitored continuously using cameras, with a total of more than 1 million images of radiometer domes archived. Ventilator and ventilator–heater performance overall was skillful with the average of the systems mitigating ice formation 77 % (many >90 %) of the time during which icing conditions were present. Ventilators without heating elements were also effective and capable of providing heat through roughly equal contributions of waste energy from the ventilator fan and adiabatic heating downstream of the fan. This provided ∼0.6 ∘C of warming, enough to subsaturate the air up to a relative humidity (with respect to ice) of ∼105 %. Because the mitigation technologies performed well, a near complete record of verified ice-free radiometric fluxes was assembled for the duration of the campaign. This well-characterized data set is suitable for model evaluation, in particular for the Year of Polar Prediction (YOPP) first Special Observing Period (SOP1). We used the data set to calculate short- and long-term biases in iced sensors, finding that biases can be up to +60 W m−2 (longwave) and −211 to +188 W m−2 (shortwave). However, because of the frequency of icing, mitigation of ice by ventilators, cloud conditions, and the timing of icing relative to available sunlight, the biases in the monthly means were generally less than the aggregate uncertainty attributed to other conventional sources in both the shortwave and longwave.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Open Call LDRD: Physically Informed Autoencoders for Galactic Redshift Regression

Physical constraints have been suggested to make neural network models more generalizable, act scientifically plausible, and be more data-efficient over unconstrained baselines. In this report, we present preliminary work on evaluating the effects of adding soft physical constraints to computer vision neural networks trained to estimate the conditional density of redshift on input galaxy images for the Sloan Digital Sky Survey. We introduce physically motivated soft constraint terms that are not implemented with differential or integral operators. We frame this work as a simple ablation study where the effect of including soft physical constraints is compared to an unconstrained baseline. We compare networks using standard point estimate metrics for photometric redshift estimation, as well as metrics to evaluate how faithful our conditional density estimate represents the probability over the ensemble of our test dataset. We find no evidence that the implemented soft physical constraints are more effective regularizers than augmentation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Automated Waterbox Inspection for Nuclear Power Plants Using Computer Vision - Based Change Detection

Nuclear power plant waterboxes require regular inspection for leaks, missing components, and structural damage during maintenance outages. Traditional manual inspection is time-consuming and poses safety risks from confined space entry. We developed an automated computer vision system for drone-based waterbox inspection in partnership with Florida Light and Power. Our approach uses feature detection and matching to identify critical changes between baseline and current inspection images, automatically flagging additions (leaks/debris), removals (missing plugs), and translations (displaced components) while compensating for drone movement and environmental variations. We systematically evaluated six feature matching methods, from classical approaches (SIFT+BF) to state-of-the-art neural networks (SuperPoint+SuperGlue), using both standard benchmarks (HPatches) and waterbox-specific validation with real-world augmentations. SuperPoint+SuperGlue achieved superior performance with 7.82 pixels RMSE and 100% success rate—2.8x better accuracy than our baseline. While the pre-trained model has commercial licensing restrictions for nuclear deployment, our findings validate this architecture for custom training. We implemented a real-time GUI demonstrating the SIFT+BF approach for immediate deployment, processing drone feeds at 30 FPS with color-coded change visualization. Future work includes training a custom SuperPoint+SuperGlue model on waterbox data and integrating Vision-Language Models for automated reporting and maintenance guidance.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗
DOE OSTI2025

Neural network-based classification and regression of magnetohydrodynamic modes in tokamaks

We present a machine learning-based magnetohydrodynamic (MHD) classifier and regressor that utilizes real or complex-valued 3D magnetic sensor array data to determine neoclassical tearing mode (NTM) onset times in tokamaks with millisecond accuracy. The input dataset consists of poloidal profiles of complex Fourier amplitudes with an n = 1 toroidal mode number from 144 human-labeled ITER Baseline Scenario discharges in the DIII-D tokamak, spanning both tearing-dominated and sawtooth-dominated regimes. Since m, n = 2,1 NTMs frequently emerge alongside sawteeth at the same frequency in this scenario, the focus is on isolating the m = 1 and m = 2 components of the n = 1 MHD mode near the tearing onset. To improve model regularization and prediction stability, singular value decomposition was applied to balance the sawtooth and tearing datasets. The enriched datasets facilitated training neural networks that learn the key distinguishing features of sawtooth and tearing modes in the poloidal profiles of their magnetic amplitude and phase. When the modes occur independently, the networks achieve perfect classification due to the modes’ distinct characteristics and low measurement noise. In the more experimentally relevant case where both modes coexist, the networks maintain exceptional performance across key metrics. Tests on synthetic data with known ground truth demonstrate the superior accuracy of the neural network trained on complex-valued input compared to models using real amplitude, phase, or pseudo-complex data, achieving both a mean time delay and standard deviation below 1 ms. Notably, standard linear regression methods fitting the dominant singular modes to the data closely match the neural network’s performance. Applying these methods across a broad range of H-mode scenarios will enable future studies to systematically identify dominant NTM triggers as scenario-specific variables, paving the way for more effective tearing mode avoidance strategies in future fusion reactor designs.

machine learning↗
DOE OSTI2021

Basin scale distributions of dissolved manganese, nickel, zinc and cadmium in the Mediterranean Sea

Samples for dissolved trace metal concentrations were collected during GEOTRACES expedition GA04-N in summer/spring in the Mediterranean Sea, starting in the Atlantic Ocean and sampling both deep basins of the Mediterranean Sea. Outflow of Mediterranean Outflow Water leads to elevated concentrations of Mn, Ni and Zn in the Atlantic Ocean, but a concentration minimum in the Atlantic distribution of Cd. Nevertheless, when comparing the in- and outflow, the Mediterranean is a net source of Cd to the Atlantic Ocean. Surface concentrations of Mn, Ni, Zn and Cd are elevated in the Mediterranean relative to the Atlantic Ocean where Ni and Cd gradually increased along the eastward surface water flow path, Zn reached a homogenous concentration in the order of 1 to 1.5 nM, and Mn displayed a patchy surface distribution. The observed differences are attributable to the different dynamics of their biogeochemical cycling, notably the partitioning between the dissolved and particulate phases due to biological uptake, scavenging and possibly organic complexation. The elevated surface concentrations of Mn, Ni, Zn and Cd in the Mediterranean are derived from atmospheric deposition, where most likely Zn and Cd are mainly sourced from anthropogenic origin, Mn mostly from lithogenic origin and Ni from both anthropogenic and lithogenic origin. Dissolved Zn and Cd, as well as phosphate and nitrate, display striking inter-basin fractionations with elevated concentrations in the deep water of the western basin compared to the deep eastern basin, without a coinciding increase in the apparent oxygen utilization. Given that physical circulation or contribution from biogenic particulate metals cannot explain the elevated dissolved concentrations, an external non-biological source is required. This source is most likely a vertical flux of metal laden particles dissolving through the water column of the western Mediterranean where these particles, most likely from anthropogenic origin, are derived from either atmospheric deposition or particulate material deposited on the continental shelves that makes its way into the deep basin. Furthermore, to confirm or detect trends in dissolved metal concentrations in the deep basin, regular basin wide assessments of the trace metal distributions in the Mediterranean are needed. The distributions of Mn, Ni, Zn and Cd in the Sea of Marmara illustrate that all of these metals can be affected by anthropogenic surface sources and highlight the different susceptibilities of the dissolved metal distributions to supply from remineralization, and to removal through scavenging. This study provides a first baseline to assess future changes and underlines that the Mediterranean marine environment is susceptible to anthropogenic disturbances, with varying effects for different metals due to differing source strengths and biogeochemical cycles.

58 GEOSCIENCES↗
DOE OSTI2022

Effectiveness of supervised exercise, home-based exercise, or walk advice strategies on walking performance and muscle endurance in patients with intermittent claudication (SUNFIT trial): a randomized clinical trial

Abstract Aims Supervised exercise is a guideline-recommended treatment in intermittent claudication (IC). Hospital-based supervised exercise programmes (SEPs) are underutilized, while home-based structured exercise programmes (HSEPs) have attracted interest. The results from HSEP in IC are inconsistent and may confer no benefit over walk advice (WA) and be less effective than SEP. The aim of the study was to compare the effectiveness of best medical treatment, including Nordic pole WA alone, or WA + SEP or WA + HSEP for patients with IC. Methods and results This three-armed, multicentre randomized clinical trial enrolled patients with IC; all patients received best medical treatment including walking poles and the advice of regular Nordic pole walking (WA). For HSEP and SEP, additional exercise programmes were provided. The primarily investigated hypothesis was a non-inferiority analysis of SEP vs. HSEP regarding the 6-min walk test (6MWT) maximum distance, with a pre-defined non-inferiority margin of 50 m. Supporting outcomes included muscle endurance tests and the walking impairment questionnaire. Outcomes were assessed at baseline, 3, 6, and 12 months by a blinded evaluator. Altogether 166 patients (mean age 72 years; 59% males) were randomized. In HSEP and SEP, 24 and 26% patients, respectively, were fully exercise adherent. All three groups improved pain-free walking distance over time, but there were no significant intergroup differences. The intergroup 6MWT difference between SEP and HSEP from 0 to 12 months was –11.6 m, 95% confidence interval: –36.4 to 13.0 m (i.e. within the pre-specified non-inferiority margin). Conclusion The HSEP was non-inferior to SEP in patients with IC. There were no significant differences observed between the three groups at 1 year. Registration ClinicialTrials.gov: NCT02341716.

Sandberg, Anna (ORCID:0000000258432784)↗
DOE OSTI2025

Day-Ahead Forecasting with Federated LSTM to Plan Energy Sharing in a Community Microgrid

Energy balancing in microgrids is a key enabler of resilience. Community microgrids located close to each other have the added benefit of networking and sharing surplus energy, if available. Such complex decision-making runs on optimization that requires reliable short-term (up to very-short-term) forecasts of energy generation and consumption for scheduling or trading. Each microgrid may also opt to not expose their sensitive data such as consumption patterns of individual businesses or residences. This paper investigates a federated approach to dayahead forecasting that trains naive long short-term memory (LSTM) at each business in a microgrid and aggregates weights at the microgrid controller using proximal regularization. This approach ensures that the controller has access only to energy surplus/deficit and not the actual generation or consumption values, avoiding unwanted exposure of sensitive data. A community microgrid in Adjuntas, Puerto Rico with 3 businesses is selected as a case study with a laboratory-scale computing setup. A central LSTM forecaster, where sensitive data from businesses are aggregated at the controller, is implemented as a baseline for qualifying the results. This work serves as a proof-of-concept for scaling the approach to networked and nested microgrids with more complex control options.

Sundararajan, Aditya [ORNL] (ORCID:000000033577854↗
DOE OSTI2020

Performance Assessments of Demand Flexibility from Grid-Interactive Efficient Buildings: Issues and Considerations

This SEE Action Network report explains basic concepts and fundamental considerations for assessing the actual demand flexibility performance of buildings participating in demand flexibility programs and responding to time-varying retail rates. Demand flexibility is the capability of distributed energy resources (DERs) to adjust a building’s load profile across different timescales. Assessments determine the timing, location, quantity, and quality of grid services provided. The results can be used for financial settlements and to improve performance of demand flexibility, support its consideration in resource potential studies and electricity system planning, and contribute to cost-effectiveness evaluations. While practitioners and regulators regularly find opportunities to improve performance assessments of demand-related services, to a large degree current best practices are sufficient for basic service offerings as demand flexibility is implemented today. However, advances in assessment practices will be required in a future with grid-interactive efficient buildings (GEBs) that provide continuous demand flexibility by integrating multiple DERs and flexibility modes (load shed, load shift, modulate, and generate). Example practices include using new baseline constructs—including whether baselines are even required for certain building flexibility modes or configurations, deploying more advanced metering and analytics, further developing cybersecurity standards, improving communication standards for increased interoperability, and establishing performance metrics and assessment procedures for load modulation as it is more fully defined and implemented. This report provides information on prioritizing and designing performance assessment elements for demand flexibility programs and time-varying retail rates, assessment protocols, and research and development needs. Additional research sponsored by DOE’s Building Technologies Office offers more detailed technical information on assessing performance of demand flexibility, including definition of performance metrics.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

Solvent Hold Tank Sample Results for MCU-20-32-33-34 (Quarterly Report, September 2020)

In late FY13, MCU implemented the Next Generation Solvent (NGS) flow sheet. Facility personnel added a non-radioactive, NGS “cocktail” containing the new Extractant (MaxCalix) and a new Suppressor (TiDG) to the SHT heel to implement the NGS flow sheet. The resulting “blend” solvent (“NGS blend solvent”) is essentially NGS with residual amounts of calix[4]arene-bis(tert-octylbenzo-crown-6) (BOBCalixC6) and trioctylamine (TOA). For process monitoring, SHT samples are sent to Savannah River National Laboratory (SRNL) to examine solvent composition changes over time. With the exception of Isopar™ L which is regularly added to the SHT due to its high vapor pressure, this report shows the cumulative chemical composition data, including impurities like mercury, of the September 2020 (MCU-20-32-33-34) solvent SHT sample. A summary report of the September SHT solvent sample was issued earlier. This report examines the cumulative results from these and several past monthly reports. These samples are intended to verify that the solvent is within the specified composition range. A baseline “scratch” solvent - a scratch solvent is a preparation of all 6 solvent components (this includes the old extractant BoBCalix and the old suppressor TrioctylAmine that remained when solvent was converted to Next Generation Solvent [NGS]) of the composition that approximates the blend of cocktail and heel solvent - was prepared in the lab (September 2018) and used for comparison and evaluation. The results from the analyses are presented in this document. This report is the last characterization analysis of the SHT solvent. MCU entered a lay-up state and about 128 gallons of the SHT solvent were pumped into 55 gallons drums for future disposal.

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

Physics-Driven Convolutional Autoencoder Approach for CFD Data Compressions: Preprint

With the growing size and complexity of turbulent flow models, data compression approaches are of the utmost importance to analyze, visualize, or restart the simulations. Recently, in-situ autoencoder-based compression approaches have been proposed and shown to be effective at producing reduced representations of turbulent flow data. However, these approaches focus solely on training the model using point-wise sample reconstruction losses that do not take advantage of the physical properties of turbulent flows. In this paper, we show that training autoencoders with additional physics-informed regularizations, e.g., enforcing incompressibility and preserving enstrophy, improves the compression model in three ways: (i) the compressed data better conform to known physics for homogeneous isotropic turbulence without negatively impacting point-wise reconstruction quality, (ii) inspection of the gradients of the trained model uncovers changes to the learned compression mapping that can facilitate the use of explainability techniques, and (iii) as a performance byproduct, training losses are shown to converge up to 12x faster than the baseline model.

auto-encoders↗