SEARCH · Engineering Papers
Results for “Mathematical Computing”
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.
A stable 1D multigroup high-order low-order method
The high-order low-order (HOLO) method is a recently developed moment-based acceleration scheme for solving time-dependent thermal radiative transfer problems, and has been shown to exhibit orders of magnitude speedups over traditional time-stepping schemes. However, a linear stability analysis by Haut et al. (2015 Haut, T. S., Lowrie, R. B., Park, H., Rauenzahn, R. M., Wollaber, A. B. (2015). A linear stability analysis of the multigroup High-Order Low-Order (HOLO) method. In Proceedings of the Joint International Conference on Mathematics and Computation (M&C), Supercomputing in Nuclear Applications (SNA) and the Monte Carlo (MC) Method; Nashville, TN, April 19–23, 2015. American Nuclear Society.) revealed that the current formulation of the multigroup HOLO method was unstable in certain parameter regions. Since then, we have replaced the intensity-weighted opacity in the first angular moment equation of the low-order (LO) system with the Rosseland opacity. Furthermore, this results in a modified HOLO method (HOLO-R) that is significantly more stable.
Scientific Computing at Los Alamos National Laboratory
Abstract Not Provided
Collaboratory on Mathematics for Mesoscopic Modeling of Materials (CM4)(Final Technical Report)
Our work on the CM4 grant focused on new mathematical and computational approaches for stochastic methods for mesoscopic systems. Our work included continuum SPDEs, molecular coarse-grained models, and hybrid particle-continuum approaches. This work was motivated by applications in soft materials, fluidic devices, and interfacial phenomena. Throughout the project we collaborated with several staff researchers, including from Pacific Northwest National Laboratories (PNNL) and Sandia National Laboratory (SNL) as well as other members of the CM4 project. Below we summarize briefly a few specific accomplishments from this work over the funding period. A full list of published papers and software releases supported by the grant are listed at the end.
Data-Driven Model Validation Across Dimensions
Data-driven model validation across dimensions in mathematical and computational biology assumptions are often made (e.g., symmetry) to reduce the problem from three spatial dimensions (3D) to two (2D). However, some experimental datasets, such as cell counts obtained via flow cytometry, represent the entire 3D biological object. For purpose of model calibration and validation, it is sometimes necessary to compare these biological datasets with model outputs. In this work, we propose a methodology for scaling 2D model outputs to compare with 3D experimental datasets, and we discuss the application of this methodology to two examples: agent-based models of granuloma formation and skeletal muscle tissue. The accuracy of the method is evaluated in artificially generated scenarios.
Reduced-order modeling for efficient cross section library development in high-temperature gas reactor pebble-bed depletion analysis
Accurate modeling of running-in and equilibrium conditions in pebble-bed reactors (PBRs) requires precise microscopic multigroup neutron cross sections. In Griffin, deterministic neutronics calculations rely on multivariate interpolation over large cross section libraries, resulting in significant memory usage and performance bottlenecks. This work, together with a companion paper on Griffin integration, explores reduced-order models (ROMs) to replace interpolation with lightweight surrogates. Several ROM techniques are benchmarked, with deep neural networks (DNNs) demonstrating superior memory efficiency, scalability, and predictive accuracy. A total of 295 DNNs were trained to build a comprehensive isotope library, integrated into Griffin through a custom LibTorch interface for depletion analysis. Initial results demonstrate that DNN-based ROMs drastically reduce memory demands while preserving accuracy, enabling finer tabulations and additional state variables without overhead. In conclusion, the framework also supports online cross section generation and real-time DNN updates through transfer learning, improving fidelity by capturing self-shielding and evolving nuclide compositions during burnup.
Estimation of cerebral blood flow velocity during breath-hold challenge using artificial neural networks
The effect of untreated Obstructive Sleep Apnoea (OSA) on cerebral haemodynamics and CA impairment is an active field of research interest. A breath-hold challenge is usually used in clinical and research settings to simulate cardiovascular and cerebrovascular changes that mimic OSA events. This work utilises temporal arterial oxygen saturation (SpO 2 ) and photoplethysmography (PPG) signals to estimate the temporal cerebral blood flow velocity (CBFv) waveform. Measurements of CBFv, SpO 2 , and PPG, were acquired concurrently from volunteers performing two different protocols of breath-hold challenge in the supine position. Past values of the SpO 2 and PPG signals were used to estimate the current values of CBFv using different permutations and topologies of supervised learning with shallow artificial neural networks (ANNs). Measurements from one protocol were used to train the ANNs and find the optimum topologies, which in turn were tested using the other protocol. Data collected from 10 normotensive, healthy subjects (four females, age 28.5 ± 6.1 years, Body Mass Index (BMI) 24.0 ± 4.7 kg/m 2 ) were used in this study. The results show that different subjects have different optimum topologies for ANNs, thus indicating the effects of inter-subject variability on ANNs. Successfully reconstructed blind waveforms for the same subject group in the second protocol showed a reasonable accuracy of 60–80% estimation compared to the measured waveforms.
Developing a digital twin framework for remotely monitoring nuclear reactor facilities
A digital twin must seek to represent all applicable functional components of the system of interest. Different expertise is required for understanding the physical system being modeled than the skills needed for transforming those models into a functional digital twin through physics modeling, machine learning analysis, and visualization. The diversity of knowledge requires a multi-disciplinary team to ensure all system details are captured. Team members also need a method to verify that the data they generate within their domain can be effectively communicated to professionals in other fields. To address this challenge, this work provides an approach for developing a digital twin framework to remotely monitoring nuclear facilities. Through this, general knowledge of the framework is presented along with two examples to solidify the process. The AGN-201 digital twin and microreactor digital twins provide varying levels of complexity in a potential nuclear facility, where common threads are identified and lessons learned are provided. The goal of this research is to aid future researchers by providing a formula for a successful digital twin and in turn reducing the development time of nuclear system digital twins, specifically for remote monitoring.
Modeling heat pipe startup and noncondensable gases in Sockeye
For this work, a one-dimensional gas mixture flow model was developed and implemented in the heat pipe code Sockeye to model the effects of noncondensable gases. Additionally, a startup model based on the dusty gas model was implemented to model the transition from rarefied gas dynamics to continuum flow, which occurs during the frozen startup of high-temperature heat pipes. Multiple startup and noncondensable gas models were tested against experimental data for sodium heat pipes, showing excellent agreement. Additionally, the newly developed gas mixture model for modeling noncondensable gas is further tested with a theoretical case study with arbitrary heating configurations. Finally, several recommendations and conclusions are made from the studies in this work to guide future heat pipe modeling efforts.
Assessment of RELAP5-3D Code with Molten Salt Heat Transfer Experiments
The accurate thermal-hydraulic assessment of thermal storage systems is crucial for addressing the integrity and performance of the storage system design. This is particularly important for thermal storage systems using molten salts as a heat transport and storage medium, as the chemical corrosion and erosion by the high-temperature molten salts can cause serious damage to the structural components of the storage system. To understand these effects for TerraPower’s Natrium® Demonstration Reactor design, the system thermal-hydraulic analysis code, RELAP5-3D, is utilized to analyze the thermal storage system. This study evaluates two heat transfer correlations implemented in the RELAP5-3D code—the Dittus-Boelter and Gnielinski correlations—using selected benchmark cases from two molten salt heat transfer experiments conducted at Xi’an Jiao Tong University and the German Aerospace Center. The Nusselt numbers calculated using the Gnielinski correlation agree with the experimental data within ±5% of the relative deviations at 300°C and 400°C. With the increasing Reynolds numbers, the Dittus-Boelter correlation underestimates the Nusselt numbers by up to 22% compared to the Gnielinski correlation. Comparison of the RELAP5-3D calculation results with experimental data at a high temperature of 550°C revealed that the thermo-chemical characteristics of solar salt at high temperatures surpassing the decomposition temperatures of nitrate salts reduce the accuracy of heat transfer model of the RELAP5-3D code. Specifically, it has been observed that the temperatures of the wall surface and liquid near the wall can surpass the decomposition temperatures, even though the bulk temperature of molten salts remains significantly lower than these decomposition temperatures. In summary, this study recommends the use of the Gnielinski correlation for the heat transfer analysis of molten salt, rather than the Dittus-Boelter correlation.
Modeling Flow in a Microreactor Core: From High-Fidelity CFD to Subchannel Analysis
The accurate prediction of pressure drop in tightly packed, low–Reynolds number (Re) bare rod bundles is essential for the thermal-hydraulic design of the Microreactor Applications Research Validation and EvaLuation (MARVEL) reactor and other microreactor concepts. However, existing friction factor correlations, particularly the upgraded Cheng-Todreas (UCTD) correlation, have limited validation for the small pitch-to-diameter ratios (P/Ds) and transitional flow conditions characteristic of these systems. In this work, we perform high-fidelity large-eddy simulations (LESs) of both an infinite bare rod bundle and a finite bare 37-pin scalloped rod bundle across the range 1000≤ 𝑅𝑒 ≤5500. The simulations reveal strong gap vortex–driven transitional behavior and indicate that the UCTD may underpredict the friction factor by up to 28% at 𝑃/𝐷 = 1.05. Using the LES-calculated pressure drops, we formulate a new friction factor correlation that follows the Cheng-Todreas functional form but is calibrated for low-Re and tightly packed geometries representative of a MARVEL-like reactor. The correlation is implemented in the MOOSE (Multiphysics Object-Oriented Simulation Environment) subchannel module and compared against both the LES and UCTD predictions. Across all subchannel types, the proposed model reduces the streamwise velocity differences from as high as 44% (UCTD) to below 9%, and decreases the pressure gradient differences from 13% to 25% (UCTD) to 0.7% to 7% relative to the LES results. These results suggest that the new correlation has the potential to improve the pressure drop and flow field predictions for such geometries, highlighting the importance of high-fidelity simulations in supporting microreactor thermal-hydraulic model development and motivating future pressure drop experiments for compact rod bundles to further validate these findings.
Accelerated Irradiation Testing and Post-Irradiation Characterization: U.S.-Based Capabilities for Advanced Nuclear Systems and Radioisotope Production
Irradiation experiments and post-irradiation examinations, together referred to as irradiation testing (IRT), are prerequisites for nuclear fuel and material qualification for the deployment of new and advanced reactors, as well as radioisotope production, thereby ensuring regulatory compliance. Qualified research and test reactors (RTRs) and testing facilities are essential to enable IRT to verify performance and safety under prototypical reactor conditions. In the past, qualification of new fuels or structural materials required about 20 years. Synergist strategies, advanced tools, and qualified methods are needed to greatly reduce this timeframe of IRT and radioisotope production. This study, termed accelerated-IRT, focuses on identifying gaps and leveraging U.S.-based RTRs and material testing capabilities, leveraging the preliminary evaluation and qualification of selected RTRs to provide a generic as well as specific-case solution paths forward, ensuring adherence to stringent regulatory standards. Furthermore, IRT and radioisotope production utilizing qualified RTRs necessarily includes modeling and simulation to support the design (i.e. neutronics, thermal, and structural aspects) and manufacturing of irradiation test specimens, vehicles, capsules, apparatuses, and flow loops. In addition, IRT can be improved by applying advanced manufacturing techniques and in-pile sensors and instrumentation, as discussed in this study.
Detection of Diversion in a Realistic Heat Pipe Microreactor Using Supervised Machine Learning
Microreactors (MRs) pose new challenges for international safeguards. Here, their small size and mass reproducibility make them ideal for deployment in greater numbers and in remote locations, making the job of safeguards inspectors more challenging. Machine learning (ML) is currently being applied to many fields to augment human performance and increase automation; in particular, ML could be used to provide insight for international inspectors to help detect the diversion of nuclear fuel from MR cores. Four ML model types (k-nearest neighbors, decision tree, random forest, and histogram-based gradient boosted ensemble) were trained on integrated flux and critical control drum angle data generated with Serpent 2 for a realistic heat pipe MR design, achieving nearly 100% binary classification accuracy of nominal and diversion core configurations by the end of 1 full power year for three of the four model types. Regression model variants were also trained, using the same input data, for predicting the number of fuel pins diverted. Root-mean-square errors below 5% of the total number of fuel pins were achieved by the 1 full power year mark for all models.
A Low Mach Number Model for Moist Atmospheric Flows
A low Mach number model for moist atmospheric flows is introduced that accurately incorporates reversible moist processes in flows whose features of interest occur on advective rather than acoustic time scales. Total water is used as a prognostic variable, so that water vapor and liquid water are diagnostically recovered as needed from an exact Clausius–Clapeyron formula for moist thermodynamics. Low Mach number models can be computationally more efficient than a fully compressible model, but the low Mach number formulation introduces additional mathematical and computational complexity because of the divergence constraint imposed on the velocity field. Here in this paper, latent heat release is accounted for in the source term of the constraint by estimating the rate of phase change based on the time variation of saturated water vapor subject to the thermodynamic equilibrium constraint. Finally, the authors numerically assess the validity of the low Mach number approximation for moist atmospheric flows by contrasting the low Mach number solution to reference solutions computed with a fully compressible formulation for a variety of test problems.
Integrating Cyber-Informed Engineering into Process Automation
As organizations increasingly automate their core missions and essential functions to address business risks and enhance efficiency, process automation becomes pivotal. This shift, involving minimal or no manual intervention, significantly impacts an organization's cyber-risk landscape. While automation drives efficiencies, it also introduces new cyber risks if not properly managed. Cyber-Informed Engineering (CIE) provides a proactive framework for managing these digital risks, enhancing cyber-resilience in process automation. This document supports organizations in applying CIE principles to mitigate the cyber risks associated with automation. The outlined approach can be independently implemented to improve any organization’s cyber-resilience, ensuring that the advantages of automation do not result in unaddressed or unmanaged digital risks. It serves as a starting point, offering considerations for integrating CIE principles and practices into organizational processes. CIE is presented as an iterative process, fostering continuous improvement and reinforcing the engineering and operational cultures to manage digital risks effectively. The document is structured as follows: Section 1 provides background on CIE and process automation, and their integration. Section 2 explores the twelve CIE principles in the context of process automation, highlighting key questions, engineering considerations, and implications for digital risk management. Section 3 synthesizes the findings and offers recommendations to advance resilience by design.