Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “model benchmarking”

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 1,693 records · Page 94

Observational benchmarks inform representation of soil organic carbon dynamics in land surface models

Abstract. Representing soil organic carbon (SOC) dynamics in Earth system models (ESMs) is a key source of uncertainty in predicting carbon–climate feedbacks. Machine learning models can help identify dominant environmental controllers and establish their functional relationships with SOC stocks. The resulting knowledge can be integrated into ESMs to reduce uncertainty and improve predictions of SOC dynamics over space and time. In this study, we used a large number of SOC field observations (n=54 000), geospatial datasets of environmental factors (n=46), and two machine learning approaches (namely random forest, RF, and generalized additive modeling, GAM) to (1) identify dominant environmental controllers of global and biome-specific SOC stocks, (2) derive functional relationships between environmental controllers and SOC stocks, and (3) compare the identified environmental controllers and predictive relationships with those in models used in Phase 6 of the Coupled Model Intercomparison Project (CMIP6). Our results showed that the diurnal temperature, drought index, cation exchange capacity, and precipitation were important observed environmental predictors of global SOC stocks. While the RF model identified 14 environmental factors that describe climatic, vegetation, and edaphic conditions as important predictors of global SOC stocks (R2=0.61, RMSE = 0.46 kg m−2), current ESMs oversimplify the relationships between environmental factors and SOC, with precipitation, temperature, and net primary productivity explaining > 96 % of the variability in ESM-modeled SOC stocks. Further, our study revealed notable disparities among the functional relationships between environmental factors and SOC stocks simulated by ESMs compared with observed relationships. To improve SOC representations in ESMs, it is imperative to incorporate additional environmental controls, such as the cation exchange capacity, and refine the functional relationships to align more closely with observations.

54 ENVIRONMENTAL SCIENCES↗

Benchmarking and Exploring Parameter Space of the 2-Phase Bubble Tracking Model for Liquid Mercury Target Simulation

High intensity proton pulses strike the Spallation Neutron Source (SNS)’s mercury target to provide bright neutron beams. These strikes deposit extensive energy into the mercury and its steel vessel. Prediction of the resultant loading on the target is difficult when helium gas is intentionally injected into the mercury to reduce the loading and to mitigate the pitting damage on the vessel. A 2-phase material model that incorporates the Rayleigh-Plesset (R-P) model is expected to address this complex multi-physics dynamics problem by including the bubble dynamics in the liquid mercury. We present a study comparing the measured target strains in the SNS target station with the simulation results of the solid mechanics simulation framework. We investigate a wide range of various physical model parameters, including the number of bubble families, bubble size distribution, viscosity, surface tension, etc. to understand their impact on simulation accuracy. Our initial findings reveal that using 8-10 bubble families in the model renders a simulation strain envelope that covers the experimental ones. Further optimization studies are planned to predict the strain response more accurately.

Lin, Lianshan↗

Uncertainty Quantification with MCMC method and BFBT Data

The reliability of predictions of the system codes is closely related to the validation of their physical models. For example, the accuracy of void fraction prediction in a Boiling Water Reactor (BWR) using TRACE code depends on the uncertainties of closure relations developed for the two-phase flow models. This work focus on quantifying the uncertainties of two physical models, interfacial drag model and subcooled boiling model, using Markov Chain Monte Carlo (MCMC) method and BFBT experimental benchmark data. We applied the MCMC and MLE algorithms to BFBT benchmark data and estimated the uncertainty distribution of two model parameters: subcooled boiling heat transfer coefficient and interfacial drag (bubbly/slug rod bundle) coefficient. These algorithms tend to give various estimation results. Despite the differences in these algorithms, we find that the estimation results are consistent. After obtaining the uncertainty estimates, we validated the uncertainty estimates by applying the estimated distributions of model parameters to TRACE prediction. The new predictions succeed in reducing the void fraction prediction error, but it is not possible to eliminate all prediction error. We are only adjusting two model parameters, and there exists uncertainties in all model parameters. Application of the algorithms to other physical models and experimental data will be helpful for future work. (authors)

42 ENGINEERING↗

Application of Phase-Field Techniques to Hydraulically- and Deformation-Induced Fracture.

Phase-field techniques provide an alternative approach to fracture problems which mitigate some of the computational expense associated with tracking the crack interface and the coalescence of individual fractures. The technique is extended to apply to hydraulically driven fracture such as would occur during fracking or CO 2 sequestration. Additionally, the technique is applied to a stainless steel specimen used in the Sandia Fracture Challenge. It was found that the phase-field model performs very well, at least qualitatively, in both deformation-induced fracture and hydraulically-induced fracture, though spurious hourglassing modes were observed during coupled hydralically-induced fracture. Future work would include performing additional quantitative benchmark tests and updating the model as needed.

42 ENGINEERING↗

Whisper Use of Nuclear Data Covariances [Slides]

Whisper is statistical analysis code using sensitivity/uncertainty-based methods to determine baseline upper subcritical limit (USL) for nuclear criticality safety. Features of Whisper 1.1 include: GLLS method implemented to compute adjusted covariance based on current benchmark suite (1,100+ ICSBEP models), BLO “low-fidelity” covariance data used (44 energy groups), adjusted covariance is pre-computed and saved, and adjusted cross sections are NOT computed. Potential future efforts include: an extension to include angular distributions in benchmark selection and in GLLS adjustment, a move toward more modern covariance data (ENDF/B-VIII.0) and different group structure, and compute and store adjusted cross sections (trivial).

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Transient Modelling of HTGR Thermal Load Follow in Modelica

The ability of nuclear power plants to load follow electrical power demand is particularly important as we move to higher variable renewable (VRE) energy production on low carbon energy grids. High temperature gas reactors (HTGRs) are capable of load following by adjusting thermal output, but their high temperatures and thermal inertia significantly impact the power transients they are able to follow. To aid investigation into new integrated energy systems with advanced reactors, it is necessary to have an open-source transient model of a full HTGR primary and secondary. This paper introduces two new open-source models for a HTGR and corresponding balance of plant (BOP) written for the HYBRID library developed at Idaho National Laboratory. A transient thermal load follow test case was taken by the work from Brits et al. to benchmark our open-source model against transient operation results from other proprietary codes that exist. The control scheme for the two new models in combination is discussed and a comparison of the response in transient load follow between the two codes performed.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Modeling Urban Scenarios & Experiments: Fort Indiantown Gap Data Collections Summary and Analysis

This report summarizes experimental radiation detector, contextual sensor, weather, and global positioning system (GPS) data collected to inform and validate a comprehensive, operational radiation transport modeling framework to evaluate radiation detector system and algorithm performance. This framework will be used to study the influence of systematic effects (such as geometry, background activity, background variability, environmental shielding, etc.) on detector responses and algorithm performance using synthetic time series data. This work consists of performing data collection campaigns at a canonical, controlled environment for complete radiological characterization to help construct and benchmark a high-fidelity model with quantified system geometries, detector response functions, and source terms for background and threat objects. This data also provides an archival, benchmark dataset that can be used by the radiation detection community. The data reported here spans four data collection campaigns conducted between May 2015 and September 2016.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Online thermal profile prediction for large format additive manufacturing: A hybrid CNN-LSTM based approach

Large format additive manufacturing (LFAM) is an advanced 3D printing technique that efficiently fabricates large-scale components through a layer-by-layer extrusion and deposition process. Accurate surface layer temperature monitoring is essential to prevent manufacturing failures and ensure final product quality. Traditional physics-based offline approaches for simulating thermal behavior are often inefficient and complex, posing challenges on real-time, in-situ monitoring. Here, to address this, we propose a data-driven hybrid CNN-LSTM model to predict sequential thermal images of arbitrary length using real-time infrared thermal imaging. In this approach, a Convolutional Neural Networks (CNN) is trained offline to capture spatial features, reduce dimensional complexity, and enhance time efficiency, while a stacked Long Short-Term Memory (LSTM) is applied online to capture temporal information for improved prediction of future thermal behavior in subsequent printing layers. Model performance is evaluated using MSE, SSIM, and PSNR metrics and is benchmarked against stacked LSTM and convolutional LSTM models, demonstrating superior accuracy and applicability. Additionally, to mitigate noise from moving extruders and gantry backgrounds in thermal images, a fine-tuned semantic segmentation model is implemented offline to extract printing geometry, enabling precise temperature tracking along the tool path for further thermal analysis. The frameworks developed in this study significantly advance temperature monitoring, thermal analysis, and in-situ manufacturing control for LFAM, bridging the gap between theoretical modeling and practical application.

Geometry extraction↗

Enhanced Co-Expression Extrapolation (COXEN) Gene Selection Method for Building Anti-Cancer Drug Response Prediction Models

The co-expression extrapolation (COXEN) method has been successfully used in multiple studies to select genes for predicting the response of tumor cells to a specific drug treatment. Here, we enhance the COXEN method to select genes that are predictive of the efficacies of multiple drugs for building general drug response prediction models that are not specific to a particular drug. The enhanced COXEN method first ranks the genes according to their prediction power for each individual drug and then takes a union of top predictive genes of all the drugs, among which the algorithm further selects genes whose co-expression patterns are well preserved between cancer cases for building prediction models. We apply the proposed method on benchmark in vitro drug screening datasets and compare the performance of prediction models built based on the genes selected by the enhanced COXEN method to that of models built on genes selected by the original COXEN method and randomly picked genes. Models built with the enhanced COXEN method always present a statistically significantly improved prediction performance (adjusted p-value ≤ 0.05). Our results demonstrate the enhanced COXEN method can dramatically increase the power of gene expression data for predicting drug response.

60 APPLIED LIFE SCIENCES↗

2D numerical simulation of unsteady flows for large scale floods prediction in real time

The challenge of finding a compromise between computational time and level of accuracy and robustness has traditionally expanded the use simplified models rather than full two-dimensional (2D) models for flood simulation. This work presents a GPU accelerated 2D shallow water model for the simulation of flood events in real time. In particular, an explicit first-order finite volume scheme is detailed to control the numerical instabilities that are likely to appear when used in complex topography. The model is first validated with the benchmark test case of the Toce River (Italy) and numerical fixes are demonstrated to be necessary. The model is next applied to reproduce real events in a reach of the Ebro River (Spain) in order to compare simulation results with field data. The second case deals with a large domain (744 km 2 ) and long flood duration (up to 20 days) allowing an analysis of the performance and speed-up achieved by different GPU devices. We report the high values of fit between observed and simulated results as well as the computational times achieved are encouraging to propose the use of the model as forecasting system.

54 ENVIRONMENTAL SCIENCES↗

Machine learning and ligand binding predictions: A review of data, methods, and obstacles

We report that computational predictions of ligand binding is a difficult problem, with more accurate methods being extremely computationally expensive. The use of machine learning for drug binding predictions could possibly leverage the use of biomedical big data in exchange for time-intensive simulations. This paper reviews current trends in the use of machine learning for drug binding predictions, data sources to develop machine learning algorithms, and potential problems that may lead to overfitting and ungeneralizable models. A few popular datasets that can be used to develop virtual high-throughput screening models are characterized using spatial statistics to quantify potential biases. We can see from evaluating some common benchmarks that good performance correlates with models with high-predicted bias scores and models with low bias scores do not have much predictive power. A better understanding of the limits of available data sources and how to fix them will lead to more generalizable models that will lead to novel drug discovery.

59 BASIC BIOLOGICAL SCIENCES↗

Dynamic System Simulation of Fissile Solution Systems

Dynamic System Simulation (DSS) utilizes state-variables described by differential or difference equations to model system evolution. DSS techniques have been applied to develop a family of models to examine the time-dependent operational behavior of fissile solution systems. The base, or generic model, predicts the behavior of an aqueous homogeneous reactor (AHR), configured with a single cooling loop, operating at atmospheric pressure. A model specific for SUPO (Super Power), a uranium solution fueled AHR that operated at Los Alamos National Laboratory from 1951 to 1974, tailored from the generic model demonstrated close correlation with experimental data for steady-state operation1. SUPO is considered the benchmark for steady-state AHR operation. Subsequently specific models were tailored from the generic model for a variety of historical reactors including KEWB (Kinetics Experiments Water Boiler) “A-2” and “B-5” cores and Silene, which is considered the benchmark for AHR pulse operations. These models demonstrated DSS techniques could reliably be extended to different core geometries (SUPO and KEWB “A-2” were spherical; KEWB “B-5” and Silene cylindrical)2. This family of models also shows close correlation with experimental data in all modes of fissile solution system operation including pulse, free evolution, and steady-state. System response due to rate and amplitude of reactivity insertion closely matches experimental data. This report describes additional extensions of this family of models to include a wider design space for fissile solution systems. Pressurized cores, cooling schemes involving multiple cooling loops of various geometries, and accelerator-driven sub-critical system concepts are modeled.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

A comparative study of discrete fracture network and equivalent continuum models for simulating flow and transport in the far field of a hypothetical nuclear waste repository in crystalline host rock

One of the major challenges of simulating flow and transport in the far field of a geologic repository in crystalline host rock is related to reproducing the properties of the fracture network over the large volume of rock with sparse fracture characterization data. Various approaches have been developed to simulate flow and transport through the fractured rock. The approaches can be broadly divided into Discrete Fracture Network (DFN) and Equivalent Continuum Model (ECM). The DFN explicitly represents individual fractures, while the ECM uses fracture properties to determine equivalent continuum parameters. In this paper, we compare DFN and ECM in terms of upscaled observed transport properties through generic fracture networks. The major effort was directed on making the DFN and ECM approaches similar in their conceptual representations. This allows for separating differences related to the interpretation of the test conditions and parameters from the differences between the DFN and ECM approaches. The two models are compared using a benchmark test problem that is constructed to represent the far field (1 × 1 × 1 km 3 ) of a hypothetical repository in fractured crystalline rock. The test problem setting uses generic fracture properties that can be expected in crystalline rocks. The models are compared in terms of the: 1) effective permeability of the domain, and 2) nonreactive solute breakthrough curves through the domain. The principal differences between the models are mesh size, network connectivity, matrix diffusion and anisotropy. We demonstrate how these differences affect the flow and transport. Finally, we identify the factors that should be taken in consideration when selecting an approach most suitable for the site-specific conditions.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Feasibility analyses for HEU to LEU fuel conversion of the LAUE Langivin Institute (ILL) High Flux Reactor (RHF).

The High Flux Reactor (RHF) of the Laue Langevin Institute (ILL) based in Grenoble, France is a research reactor designed primarily for neutron beam experiments for fundamental science. It delivers one of the most intense neutron fluxes worldwide, with an unperturbed thermal neutron flux of 1.5 x 10{sup 15} n/cm{sup 2}/s in its reflector. The reactor has been conceived to operate at a nuclear power of 57 MW but currently operates at 52 MW. The reactor currently uses a Highly Enriched Uranium (HEU) fuel. In the framework of its non-proliferation policies, the international community presently aims to minimize the amount of nuclear material available that could be used for nuclear weapons. In this geopolitical context, most worldwide research and test reactors have already started a program of conversion to the use of Low Enriched Uranium (LEU) fuel. A new type of LEU fuel based on a mixture of uranium and molybdenum (UMo) is expected to allow the conversion of compact high performance reactors like the RHF. This report presents the results of reactor design, performance and steady state safety analyses for conversion of the RHF from the use of HEU fuel to the use of UMo LEU fuel. The objective of this work was to show that is feasible, under a set of manufacturing assumptions, to design a new RHF fuel element that could safely replace the HEU element currently used. The new proposed design has been developed to maximize performance, minimize changes and preserve strong safety margins. Neutronics and thermal-hydraulics models of the RHF have been developed and qualified by benchmark against experiments and/or against other codes and models. The models developed were then used to evaluate the RHF performance if LEU UMo were to replace the current HEU fuel 'meat' without any geometric change to the fuel plates. Results of these direct replacement analyses have shown a significant degradation of the RHF performance, in terms of both neutron flux and cycle length. Consequently, ANL and ILL have collaborated to investigate alternative designs. A promising candidate design has been selected and studied, increasing the total amount of fuel without changing the external plate dimensions by relocating the burnable poison. In this way, changes required in the fuel element are reasonably small. With this new design, neutronics analyses have shown that performance could be maintained at a high level: 2 day decrease of cycle length (to 47.5 days at 58.3 MW) and 1-2% decrease of brightness in the cold and hot sources in comparison to the current typical operation. In addition, studies have shown that the thermal-hydraulic and shutdown margins for the proposed LEU design would satisfy technical specifications.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Uncertainty Quantification Techniques of SCALE/TSUNAMI

The Standardized Computer Analysis for Licensing Evaluation (SCALE) code system developed at Oak Ridge National Laboratory (ORNL) includes Tools for Sensitivity and Uncertainty Analysis Methodology Implementation (TSUNAMI). The TSUNAMI code suite can quantify the predicted change in system responses, such as k eff , reactivity differences, or ratios of fluxes or reaction rates, due to changes in the energy-dependent, nuclide-reaction-specific cross-section data. Where uncertainties in the neutron cross-section data are available, the sensitivity of the system to the cross-section data can be applied to propagate the uncertainties in the cross-section data to an uncertainty in the system response. Uncertainty quantification is useful for identifying potential sources of computational biases and highlighting parameters important to code validation. Traditional validation techniques often examine one or more average physical parameters to characterize a system and identify applicable benchmark experiments. However, with TSUNAMI correlation coefficients are developed by propagating the uncertainties in neutron cross-section data to uncertainties in the computed responses for experiments and safety applications through sensitivity coefficients. The bias in the experiments, as a function of their correlation coefficient with the intended application, is extrapolated to predict the bias and bias uncertainty in the application through trending analysis or generalized linear least squares techniques, often referred to as 'data adjustment.' Even with advanced tools to identify benchmark experiments, analysts occasionally find that the application models include some feature or material for which adequately similar benchmark experiments do not exist to support validation. For example, a criticality safety analyst may want to take credit for the presence of fission products in spent nuclear fuel. In such cases, analysts sometimes rely on 'expert judgment' to select an additional administrative margin to account for gap in the validation data or to conclude that the impact on the calculated bias and bias uncertainty is negligible. As a result of advances in computer programs and the evolution of cross-section covariance data, analysts can use the sensitivity and uncertainty analysis tools in the TSUNAMI codes to estimate the potential impact on the application-specific bias and bias uncertainty resulting from nuclides not represented in available benchmark experiments. This paper presents the application of methods described in a companion paper.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Flexible, integrated modeling of tokamak stability, transport, equilibrium, and pedestal physics

The STEP (Stability, Transport, Equilibrium, and Pedestal) integrated-modeling tool has been developed in OMFIT to predict stable, tokamak equilibria self-consistently with core-transport and pedestal calculations. STEP couples theory-based codes to integrate a variety of physics, including magnetohydrodynamic stability, transport, equilibrium, pedestal formation, and current-drive, heating, and fueling. The input/output of each code is interfaced with a centralized ITER-Integrated Modelling & Analysis Suite data structure, allowing codes to be run in any order and enabling open-loop, feedback, and optimization workflows. This paradigm simplifies the integration of new codes, making STEP highly extensible. STEP has been verified against a published benchmark of six different integrated models. Core-pedestal calculations with STEP have been successfully validated against individual DIII-D H-mode discharges and across more than 500 discharges of the H98,y2 database, with a mean error in confinement time from experiment less than 19%. STEP has also reproduced results in less conventional DIII-D scenarios, including negative-central-shear and negative-triangularity plasmas. Predictive STEP modeling has been used to assess performance in several tokamak reactors. Simulations of a high-field, large-aspect-ratio reactor show significantly lower fusion power than predicted by a zero-dimensional study, demonstrating the limitations of scaling-law extrapolations. STEP predictions have found promising scenarios for an EXhaust and Confinement Integration Tokamak Experiment, including a high-pressure, 80%-bootstrap-fraction plasma. ITER modeling with STEP has shown that pellet fueling enhances fusion gain in both the baseline and advanced-inductive scenarios. Finally, STEP predictions for the SPARC baseline scenario are in good agreement with published results from the physics basis.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

The G+M eclipsing binary V530 Orionis: a stringent test of magnetic stellar evolution models for low-mass stars

We report extensive photometric and spectroscopic observations of the 6.1 day period, G+M-type detached double-lined eclipsing binary V530 Ori, an important new benchmark system for testing stellar evolution models for low-mass stars. We determine accurate masses and radii for the components with errors of 0.7% and 1.3%, as follows: M {sub A} = 1.0038 ± 0.0066 M {sub ☉}, M {sub B} = 0.5955 ± 0.0022 M {sub ☉}, R {sub A} = 0.980 ± 0.013 R {sub ☉}, and R {sub B} = 0.5873 ± 0.0067 R {sub ☉}. The effective temperatures are 5890 ± 100 K (G1 V) and 3880 ± 120 K (M1 V), respectively. A detailed chemical analysis probing more than 20 elements in the primary spectrum shows the system to have a slightly subsolar abundance, with [Fe/H] = –0.12 ± 0.08. A comparison with theory reveals that standard models underpredict the radius and overpredict the temperature of the secondary, as has been found previously for other M dwarfs. On the other hand, models from the Dartmouth series incorporating magnetic fields are able to match the observations of the secondary star at the same age as the primary (∼3 Gyr) with a surface field strength of 2.1 ± 0.4 kG when using a rotational dynamo prescription, or 1.3 ± 0.4 kG with a turbulent dynamo approach, not far from our empirical estimate for this star of 0.83 ± 0.65 kG. The observations are most consistent with magnetic fields playing only a small role in changing the global properties of the primary. The V530 Ori system thus provides an important demonstration that recent advances in modeling appear to be on the right track to explain the long-standing problem of radius inflation and temperature suppression in low-mass stars.

79 ASTRONOMY AND ASTROPHYSICS↗