Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “PHYSICS”

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 1,189 records · Page 66
All sourcesNASA NTRSDOE OSTI
DOE OSTI2014

Economic and Physical Modeling of Land Use in GCAM 3.0 and an Application to Agricultural Productivity, Land, and Terrestrial Carbon

We explore the impact of changes in agricultural productivity on global land use and terrestrial carbon using the new agriculture and land use modeling approach developed for Global Change Assessment Model (GCAM) version 3.0. This approach models economic land use decisions with regional, physical, and technological specificity while maintaining economic and physical integration with the rest of the GCAM model. Physical land characteristics and quantities are tracked explicitly, and crop production practices are modeled discretely to facilitate coupling with physical models. Economic land allocation is modeled with non-linear functions in a market equilibrium rather than through a constrained optimization. In this paper, we explore three scenarios of future agriculture productivity in all regions of the globe over this century, ranging from a high growth to a zero growth level. The higher productivity growth scenario leads to lower crop prices, increased production of crops in developing nations, preservation of global forested lands and lower terrestrial carbon emissions. The scenario with no productivity improvement results in higher crop prices, an expansion of crop production in the developed world, loss of forested lands globally, and higher terrestrial carbon emissions.

Wise, Marshall A.↗
DOE OSTI2017

Testing autonomous cyber-physical systems using fuzzing features from convolutional neural networks: work-in-progress

Autonomous cyber-physical systems rely on modern machine learning methods such as deep neural networks to control their interactions with the physical world. Testing of such intelligent cyber-physical systems is a challenge due to the huge state space associated with high-resolution visual sensory inputs. We demonstrate how fuzzing the input using patterns obtained from the convolutional filters of an unrelated convolutional neural network can be used to test computer vision algorithms implemented in intelligent cyber-physical systems. Our method discovers interesting counterexamples to a pedestrian detection algorithm implemented in the popular OpenCV library. Our approach also unearths counterexamples to the correct behavior of an autonomous car similar to NVIDIA's end-to-end self-driving deep neural net running on the Udacity open-source simulator.

Raj, Sunny↗
DOE OSTI2014

Assessing the Effects of Data Compression in Simulations Using Physically Motivated Metrics

This paper examines whether lossy compression can be used effectively in physics simulations as a possible strategy to combat the expected data-movement bottleneck in future high performance computing architectures. We show that, for the codes and simulations we tested, compression levels of 3–5X can be applied without causing significant changes to important physical quantities. Rather than applying signal processing error metrics, we utilize physics-based metrics appropriate for each code to assess the impact of compression. We evaluate three different simulation codes: a Lagrangian shock-hydrodynamics code, an Eulerian higher-order hydrodynamics turbulence modeling code, and an Eulerian coupled laser-plasma interaction code. We compress relevant quantities after each time-step to approximate the effects of tightly coupled compression and study the compression rates to estimate memory and disk-bandwidth reduction. We find that the error characteristics of compression algorithms must be carefully considered in the context of the underlying physics being modeled.

Laney, Daniel↗
DOE OSTI2024

Should We Conserve Entropy or Energy when Computing CAPE with Mixed-Phase Precipitation Physics?

Abstract The rapidly increasing resolution of global atmospheric reanalysis and climate model datasets necessitates finding methods for computing convective available potential energy (CAPE) both efficiently and accurately. To this end, this article compares two common methods for computing CAPE which conserve either energy or entropy. Inaccuracies in these computations arise from both physical and numerical errors. For instance, computing CAPE with entropy conserved results in physical errors from nonequilibrium phase transitions but minimizes numerical errors because solutions are analytic at each height. In contrast, computing CAPE with energy conserved avoids these physical errors, but accumulates numerical errors that are grid-resolution-dependent because the numerical integration of a differential equation is required. Analysis of CAPE computed with large databases of soundings from the tropical Amazon and midlatitude storm environments shows that physical errors from the entropy method are typically 1%–3% as large as CAPE, which is comparable to the numerical errors from conserving energy with grid spacing of 25 and 250 m using explicit first-order and second-order integration schemes, respectively. Errors in entropy-based CAPE calculations are also insensitive to vertical grid spacing, in contrast to energy-based calculations whose error strongly scales with the grid spacing. It is shown that entropy-based methods are advantageous when intercomparing datasets with differing vertical resolution because they produce accurate and reasonably fast results that are insensitive to grid resolution, whereas a second-order energy-based method is advantageous when analyzing data with a consistent vertical resolution because of its superior computational efficiency. Significance Statement Convective available potential energy (CAPE) is a measure of instability in the atmosphere that helps forecasters and researchers understand when and where thunderstorms will form. The purpose of this article is to identify the most efficient and accurate methods for computing CAPE. Two methods are considered here, one that relates to the entropy (a measure of thermodynamic disorder) of an air parcel and one that relates to the energy of an air parcel. Results indicate that the entropy method is most accurate and insensitive to the resolution of the data used for the calculation (which can vary considerably), whereas the energy method uses the least computation time.

Peters, John M.↗
DOE OSTI2020

A holistic cyber-physical security protocol for authenticating the provenance and integrity of structural health monitoring imagery data

Modern infrastructure systems, such as bridges, dams, power generation stations, and buildings increasingly have an intrinsic cyber-physical nature to them. Infrastructure now commonly, includes actuators, network connections, sensors, control systems, and computational resources. It is of increasing concern that modern infrastructure is vulnerable to cyber-attacks that can damage both the cyber and physical nature of the infrastructure. To date, the physical and cyber health of infrastructure has been considered separately. However, the increasing concerns associated with the cyber-physical security of infrastructure coupled with the emergence of 5G networks made using components that are not universally considered trustworthy, and the emergence of techniques for creating deepfakes and adversarial examples suggests the time has come to begin considering cyber health and structural health with a more holistic approach. In this work, a protocol is developed for ensuring the imagery data captured by a structural health monitoring system can be unambiguously attributed to legitimate sensors associated with the structural health monitoring system. A computer vision approach based on the idea of mutual information is then presented to detect damage in an image. This work presents the protocol for authenticating the provenance of imager data and demonstrates that this protocol does not have overly adverse effects when used with the mutual information-based technique for detecting damage in the resulting imagery data.

Jung, HweeKwon↗
DOE OSTI2023

Elucidating regulatory processes of intense physical activity by multi-omics analysis

Abstract Background Physiological and biochemical processes across tissues of the body are regulated in response to the high demands of intense physical activity in several occupations, such as firefighting, law enforcement, military, and sports. A better understanding of such processes can ultimately help improve human performance and prevent illnesses in the work environment. Methods To study regulatory processes in intense physical activity simulating real-life conditions, we performed a multi-omics analysis of three biofluids (blood plasma, urine, and saliva) collected from 11 wildland firefighters before and after a 45 min, intense exercise regimen. Omics profiles post- versus pre-exercise were compared by Student’s t -test followed by pathway analysis and comparison between the different omics modalities. Results Our multi-omics analysis identified and quantified 3835 proteins, 730 lipids and 182 metabolites combining the 3 different types of samples. The blood plasma analysis revealed signatures of tissue damage and acute repair response accompanied by enhanced carbon metabolism to meet energy demands. The urine analysis showed a strong, concomitant regulation of 6 out of 8 identified proteins from the renin-angiotensin system supporting increased excretion of catabolites, reabsorption of nutrients and maintenance of fluid balance. In saliva, we observed a decrease in 3 pro-inflammatory cytokines and an increase in 8 antimicrobial peptides. A systematic literature review identified 6 papers that support an altered susceptibility to respiratory infection. Conclusion This study shows simultaneous regulatory signatures in biofluids indicative of homeostatic maintenance during intense physical activity with possible effects on increased infection susceptibility, suggesting that caution against respiratory diseases could benefit workers on highly physical demanding jobs.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Physics-assisted generative adversarial network for X-ray tomography

X-ray tomography is capable of imaging the interior of objects in three dimensions non-invasively, with applications in biomedical imaging, materials science, electronic inspection, and other fields. The reconstruction process can be an ill-conditioned inverse problem, requiring regularization to obtain satisfactory results. Recently, deep learning has been adopted for tomographic reconstruction. Unlike iterative algorithms which require a distribution that is known a priori , deep reconstruction networks can learn a prior distribution through sampling the training distributions. In this work, we develop a Physics-assisted Generative Adversarial Network (PGAN), a two-step algorithm for tomographic reconstruction. In contrast to previous efforts, our PGAN utilizes maximum-likelihood estimates derived from the measurements to regularize the reconstruction with both known physics and the learned prior. Compared with methods with less physics assisting in training, PGAN can reduce the photon requirement with limited projection angles to achieve a given error rate. The advantages of using a physics-assisted learned prior in X-ray tomography may further enable low-photon nanoscale imaging.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

The Accelerator and Beam Physics of the Muon g-2 Experiment at Fermilab

The physics case of the Muon g-2 Experiment at Fermilab is outstanding and has recently attracted significant attention from its first official results. Although its measurements involve high energy physics methods, such as counting positron production rates with the use of calorimeters and beam diagnostics with tracking detectors, this experiment is strongly bound to accelerator and beam physics. This paper reviews the principles of the experiment and the details necessary to provide a solid ground for the beam-dynamics uncertainties and the corrections of the systematic effects influencing the output of the experiment: a single numerical value, which may unveil new physics.

43 PARTICLE ACCELERATORS↗
DOE OSTI2012

Pre-Shot Simulations of Near-Field and Far-Field Ground Motion for the Source Physics Experiment (SPE) Explosions at the Climax Stock, Nevada National Security Site: SPE3

The Source Physics Experiment (SPE) is designed to improve ground motion prediction capabilities for underground nuclear testing through the development of new physics-based models together with validation via new chemical explosion data. The SPE shots consist of different size and depth chemical explosions in a granite borehole at the Climax Stock on the Nevada National Security Site (NNSS). The third shot in the series (SPE3) will be the same size (1000 kg TNT equivalent) and depth (45 m canister centroid) as the previous shot (SPE2). The main purpose of the SPE3 is to investigate the effects of the damage zone caused by the SPE2 chemical explosion on the SPE3 explosion, keeping all other parameters (e.g. size, depth, explosive type) the same. This SPE3 pre-shot report addresses progress made in simulation of the expected wavefield in the both the near- and the far-field. In this report we consider the far-field to be where seismic wave motion can be characterized by elastic or visco-elastic approximations, in contrast to the near-field where other, often non-linear effects, play a major role. The current simulations build on previous work from the SPE1 and SPE2 shots. The simulations are intended to highlight differences between SPE2 and SPE3 using a variety to approximations intended to mimic the expected effect of the damage zone. The simplest physics of damage zone effects suggest changes in the source region material properties (density, velocity), and in the yield strength (coupling) of the rock, should both impact the far-field signal. We can illustrate this in a simplistic way for P-waves in Figure A using the empirical Mueller-Muphy (1971) far-field explosion model (MM). First if the SPE2 explosion-induced damage reduces velocities and densities without affecting the yield strength or coupling, then by energy conservation the amplitudes for SPE 3 would increase as shown in the top of Figure A. Similarly if the explosion induced damage increased density (say by compacting pore space) the amplitude would decrease. This impedance effect, where the product of the density, wave-velocity, and ground-velocity squared is conserved, is well known. The second and harder effect to quantify, are changes in the energy coupling or yield strength. In the simple MM model this is represented by the coupling factor between cavity radius and yield. If we assume the granite was slightly weakened by SPE2 such that the cavity radius increases slightly then the overall amplitudes of SPE3 decrease relative to SPE2 as shown in the bottom of Figure A. In reality both effects, may exist due to the damage zone, and furthermore we know they are interrelated in complex ways. For S-waves we expect even more complex effects due to motion on joints and faults. For this reason we are developing physics-based near field simulation models coupled to far-field propagation codes to predict the overall effects of changes in the source emplacement region on observed signals. Near-field simulations have been performed using the Eulerian code (GEODYN). An ensemble model, calibrated to match records from past large-scale nuclear events is used. Both 2D axisymmetric (no faults) and fully three-dimensional GEODYN calculations (with /without faults) were conducted in which the damage zones sustained by SPE2 are retained as the initial condition for the SPE3 calculations. The results indicate that the waveform peak amplitudes are slightly decayed since the ensemble model includes no softening mechanism. The pulse widths are reduced, indicating the energy is dissipated more by the damage zone as expected. The dominant motions calculated in all the scenarios are spherically symmetric and non-spherical components are small as seen for the SPE2 calculations. The arrival time of the SPE3 waveform at the boreholes is shorter than SPE2 due to compaction during the previous explosion. As shown in the SPE 1 and 2 observations shear waves are generated, indicating the potential contributions from the induced motion on distributed joints. The effects of joints are evaluated using the Lagrangian hydrocode (GEODYN-L), which was benchmarked against the GEODYN calculations for the ensemble model. A new granite model calibrated for samples from the SPE site is used. The new model includes effects of poroelasticity and water saturation and may be scaled allowing application to both small and large scales. Parametric tests were performed to evaluate the effect of joints within the granite and the results demonstrate that in situ joints introduce anisotropy and significant tangential motions.

58 GEOSCIENCES↗
DOE OSTI2013

Empirical and physics based mathematical models of uranium hydride decomposition kinetics with quantified uncertainties.

Metal particle beds have recently become a major technique for hydrogen storage. In order to extract hydrogen from such beds, it is crucial to understand the decomposition kinetics of the metal hydride. We are interested in obtaining a a better understanding of the uranium hydride (UH3) decomposition kinetics. We first developed an empirical model by fitting data compiled from different experimental studies in the literature and quantified the uncertainty resulting from the scattered data. We found that the decomposition time range predicted by the obtained kinetics was in a good agreement with published experimental results. Secondly, we developed a physics based mathematical model to simulate the rate of hydrogen diffusion in a hydride particle during the decomposition. We used this model to simulate the decomposition of the particles for temperatures ranging from 300K to 1000K while propagating parametric uncertainty and evaluated the kinetics from the results. We compared the kinetics parameters derived from the empirical and physics based models and found that the uncertainty in the kinetics predicted by the physics based model covers the scattered experimental data. Finally, we used the physics-based kinetics parameters to simulate the effects of boundary resistances and powder morphological changes during decomposition in a continuum level model. We found that the species change within the bed occurring during the decomposition accelerates the hydrogen flow by increasing the bed permeability, while the pressure buildup and the thermal barrier forming at the wall significantly impede the hydrogen extraction.

08 HYDROGEN↗
DOE OSTI2013

Integrated Multiple Effects Software for Nuclear Physics Applications

The technical progress made during this project has been very substantial, and in most ways has exceeded the expectations of what was intended. In particular, the ability to treat CAD-based complex geometry, using the cut-cell finite-difference methodology, has been evolved to a very advanced state. Figure 1 illustrates the progress made, comparing what the modeling was capable of at the end of the Phase I project, to what it is capable of at the end of the Phase II project. This project had a strong component of collaborative research with activities at the TJNAF (JLab), including scientific assessment of specific nuclear physics hardware, and this collaboration was a main focus during the much of the Phase II project. During the final year of this project, and subsequent months since its completion, the Department of Energy has strongly shifted emphasis for SBIR projects towards commercialization. During this same time frame the software used for the modeling, Vorpal, underwent a similarly dramatic shift from a single product with any-and-all-phyiscs-in-one, to a suite of products, under the name Vsim, with each product individually targeted at one of the areas of: Electromagnetics, Microwave Devices, Plasma Devices, and Accelerator Physics. The focus area of this project, thermal analysis, was a new but important type of multi-physics application for the software involved, and the manner in which it has been implemented is of sufficient generality, that it continues life as a fifth target area, Thermal Transport. Thus, while a somewhat more subtle commercial deployment of this project’s developments was the original intention (e.g., more-physics-in-the-one), the result is in fact more in line with the newer emphasis, e.g., a direct and obvious one-to-one commercial product deployment of project capabilities.

Smithe, David↗
DOE OSTI2014

Connecting the physical and psychosocial space to Sandia's mission

Sandia Labs has corporate, lab-wide efforts to enhance the research environment as well as improve physical space. However, these two efforts are usually done in isolation. The integration of physical space design with the nurturing of what we call psychosocial space can foster more efficient and effective creativity, innovation, collaboration, and performance. This paper presents a brief literature review on how academia and industry are studying the integration of physical and psychosocial space and focuses on the efforts that we, the authors, have made to improve the research environment in the Cyber Engineering Research Lab (CERL), home to Group 1460. Interviews with subject matter experts from Silicon Valley and the University of New Mexico plus changes to actual spaces in CERL provided us with six lessons learned when integrating physical and psychosocial space. We describe these six key takeaways in hopes that Sandia will see this area as an evolving research capability that Sandia can both contribute to and benefit from.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

Development of Magnesium Diboride Superconducting Coils for Nuclear Physics Application

The primary objective of this project is to develop an affordable magnet technology at 10-25K and then take it to a level that it can be considered mature enough to be used in a real Nuclear Physics accelerator system. We believe that because of the lower cost of magnesium diboride (MgB2) as compared to the other HTS, it offers the best such chance. Once successfully demonstrated, this exciting and innovative technology could potentially revolutionize the field the same way the conventional superconducting NbTi magnets did several decades ago. MgB2 magnets could reduce the cost of ownership (capital +operation) while enhancing performance. Another innovation is the overall cooling scheme where the exhaust gas of an existing superconducting magnet cools the inner tube which conduction cools the MgB2 coils. In several instances during the upgrade of existing facilities or designing new facilities in the Nuclear Physics or High Energy Physics communities, it is found that magnets made neither with copper coils nor with conventional superconducting NbTi coils offer an acceptable solution. Whereas copper coils limit the performance of the machine and consume large amount of power, NbTi require 4 K cryogenic facilities. This is particularly applicable in those Nuclear Physics machines where the magnetic field requirements are rather modest. One such case has been found in the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Laboratory (BNL) where an e-lens system is under construction to increase RHIC luminosity. In this case, magnesium diboride (MgB2) operating at 10-25 K seems to offer an exciting and practical solution. The main objective in Phase II is to demonstrate the reliability and robustness of the MgB2 coil technology, so that it is ready for use in RHIC and other user facilities. We will do this by fabricating several MgB2 coils and by conducting an exhaustive series of lifetime cycle tests. The tests will involve about a thousand ramp-up and ramp-down cycles, about 20 over current or thermal-excursions hitting the quench limit and about 10 thermal cycles. Phase II will begin with a design modeling effort of the MgB2 coils for the e-lens system with enough detail so that the magnetic, thermal and stress requirements can be understood for the fabrication and testing of representative MgB2 coils. After the technology is proven in Phase II for reliability, we will propose to use this in Phase III to upgrade RHIC e-lens system for improved luminosity performance and expect that more applications will follow in other accelerator and medical facilities.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2017

Development of Multi-physics (Multiphase CFD + MCNP) simulation for generic solution vessel power calculation

The current study aims to predict the steady state power of a generic solution vessel and to develop a corresponding heat transfer coefficient correlation for a Moly99 production facility by conducting a fully coupled multi-physics simulation. A prediction of steady state power for the current application is inherently interconnected between thermal hydraulic characteristics (i.e. Multiphase computational fluid dynamics solved by ANSYS-Fluent 17.2) and the corresponding neutronic behavior (i.e. particle transport solved by MCNP6.2) in the solution vessel. Thus, the development of a coupling methodology is vital to understand the system behavior at a variety of system design and postulated operating scenarios. In this study, we report on the k-effective (keff) calculation for the baseline solution vessel configuration with a selected solution concentration using MCNP K-code modeling. The associated correlation of thermal properties (e.g. density, viscosity, thermal conductivity, specific heat) at the selected solution concentration are developed based on existing experimental measurements in the open literature. The numerical coupling methodology between multiphase CFD and MCNP is successfully demonstrated, and the detailed coupling procedure is documented. In addition, improved coupling methods capturing realistic physics in the solution vessel thermal-neutronic dynamics are proposed and tested further (i.e. dynamic height adjustment, mull-cell approach). As a key outcome of the current study, a multi-physics coupling methodology between MCFD and MCNP is demonstrated and tested for four different operating conditions. Those different operating conditions are determined based on the neutron source strength at a fixed geometry condition. The steady state powers for the generic solution vessel at various operating conditions are reported, and a generalized correlation of the heat transfer coefficient for the current application is discussed. The assessment of multi-physics methodology and preliminary results from various coupled calculations (power prediction and heat transfer coefficient) can be further utilized for the system code validation and generic solution vessel design improvement.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2012

International Linear Collider Physics and detectors: 2011 Status Report

The studies of physics and detectors for the International Linear Collider are an important parallel element to the effort for the ILC Technical Design Report. The studies comprise the physics opportunities, detector requirements, and detector development to achieve the challenging high performance demanded by the physics, as well as integration of detectors into the accelerator. The current phase of this effort began with a call for Letters of Intent (LOIs) in 2007 and will lead to the submission of Detailed Baseline Design (DBD) report together with the ILC Technical Design Report at the end of 2012. Here we summarise the current status of this process, review what it has accomplished and identify the work that still needs to be completed. This report, titled International Linear Collider Physics and Detectors: 2011 Status Report, does just this.

43 PARTICLE ACCELERATORS↗
DOE OSTI2018

Proceedings of the 1998 International Computational Accelerator Physics Conference

The 1998 International Computational Accelerator Physics Conference, ICAP’98, was held September 14-18, 1998, at the Monterey Convention Center in Monterey, California. The meeting continued the tradition of four earlier Computational Accelerator Physics conferences, held in La Jolla, California (1988), Los Alamos, New Mexico (1990), Pleasanton, California (1993), and Williamsburg, Virginia (1996). ICAP’98 was organized by the Stanford Linear Accelerator Center (SLAC), Los Alamos National Laboratory (LANL), and the National Energy Research Scientific Computing Center (NERSC), and received support from the U.S. Department of Energy. The aim of this conference series is to discuss recent advances in Computational Accelerator Physics and their applications to solving important problems related to present and future accelerators. The ICAP’98 program included plenary sessions on modeling challenges in next-generation accelerators, state-of-the art methods in computational accelerator physics, and a view of what high-end computing would be like in the 21st century. There were also focused sessions on accelerator design and control, modeling high intensity beam phenomena, electromagnetic modeling, advanced algorithms, software environments and libraries, visualization and graphic interfaces, modeling advanced accelerator concepts, modeling the beam-beam interaction, computing wakefields, and combined particle/field simulations. In addition to these oral sessions there was a poster session and a vendor display. Prior to the start of the technical program, a half-day tutorial on High Performance Computing was hosted by NERSC. The ICAP series recognizes that computing and, in particular, advanced computing, is playing an increasingly important role in the successful design and development of major accelerator facilities, and in research and development leading to advances in accelerator technology. We expect that this new emphasis on high-end computing will continue in future ICAP conferences.

Ko, Kwok↗
DOE OSTI2018

MULTI-PRED: A Software Module for Predictive Modeling of Coupled Multi-Physics Systems: User's Manual

This User’s Manual describes the code module MULTI-PRED, written in FORTRAN which implements the methodology for “predictive modeling of coupled multi-physics systems (PM-CMPS)” formulated by Cacuci (2014). This methodology fully takes into account the coupling terms between the systems but requires only the computational resources that would be needed to perform predictive modeling on each system separately. The PM-CMPS methodology uses the maximum entropy principle to construct an optimal approximation of the unknown a priori distribution based on a priori known mean values and uncertainties characterizing the experimental and computational parameters and results of interest responses, called for the multi-physics models under consideration. This “maximum entropy” a priori distribution is combined, using Bayes’ theorem, with the “likelihood” provided by the multi-physics simulation models to obtain a formal posterior distribution. Subsequently, the posterior distribution thus obtained is evaluated using the saddle-point method to obtain analytical expressions for the optimally predicted values for the multi-physics models parameters and responses along with corresponding reduced uncertainties. Noteworthy, the predictive modeling methodology for the coupled systems is constructed such that the systems can be considered sequentially rather than simultaneously, while preserving exactly the same results as if the systems were treated simultaneously. Consequently, very large coupled systems, which could perhaps exceed available computational resources if treated simultaneously, can be treated with the PM-CMPS methodology presented in this work sequentially and without any loss of generality or information, requiring just the resources that would be needed if the systems were treated sequentially. Three illustrative demonstration problems are also provided. The first problem presents the application of the PM-CMPS methodology to a simple particle diffusion problem which admits a closed-form analytical solution which facilitates a rapid understanding of this methodology and its predicted results. The second demonstration problem presents the application of the PM-CMPS methodology to the problem of inverse prediction, from detector responses in the presence of counting uncertainties, of the thickness of a homogeneous slab of material containing uniformly distributed gamma-emitting sources, for optically thin and thick slabs. This problem highlights the essential role played by the relative uncertainties (or, conversely, accuracies) of measured and computed responses. The third demonstration problem presents the application of the PM-CMPS methodology to the F-area cooling towers at the Savannah River National Lab. This problem demonstrates that the PM-CMPS methodology reduces the predicted response uncertainties not only at locations where measurements are available, but also at locations where measurements are not available.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Physics-Informed Machine Learning with Application to Solid Oxide Fuel Cell System Modeling and Optimization

Traditional data-driven machine learning (ML) techniques were combined with the physics-based SOFC-MP model toward improving SOFC system-level performance prediction. Four different physics-informed ML methods were proposed to demonstrate better prediction accuracy compared with traditional ML methods. First, it was preferred that the existing Kriging-based surrogate model not be entirely abandoned, so errors between Kriging regression prediction and true solution are learned using ML. The goal was to find the pattern of the error distribution as a function of model input so discrepancies between the Kriging prediction and true solution could be reduced by adding the additional ML-trained error term to the existing prediction. Second, deep neural networks (DNN) coupled with the mass balance model (MBM) have been developed to significantly decrease the reduced order model (ROM) prediction error with fewer training data compared to the traditional Kriging-based ROM and traditional DNN regression approaches. Third, DNN regression coupled with a neural networks (NN) classifier and MBM was developed. It provides superior performance on classifying the physically operational conditions for natural gas fuel cell (NGFC) than traditional classification approaches. Finally, two kinds of DNN transferring were studied and tested. One was transferring DNN for low-fidelity model data to a high-fidelity model. The second one was transferring DNN from state-of-the-art NGFC to advanced NGFC. These physic-informed ML frameworks can serve as a solid foundation for estimating U.S. Department of Energy-targeted fuel cell performance and exhibit great potential to be employed to other engineering applications.

30 DIRECT ENERGY CONVERSION↗