A data-based private learning framework for enhanced security against replay attacks in cyber-physical systems
Not provided.
SEARCH · Engineering Papers
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.
Not provided.
Turbulent mixed convection between two vertical, infinite parallel plates at different temperatures is studied using various two-equation turbulence models. The numerical simulations are performed at a turbulent Reynolds number of Re τ = 150 and a Grashof number of Gr = 9.6 × 10 5 . Comparisons are made against the highly trusted DNS results. Consistent with the DNS approach, the current simulations are performed using constant properties and the Boussinesq approximation to predict the influence of buoyancy. Previous studies have provided assessments of two-equation turbulence models for various scenarios, but often rely on a qualitative “eye” test in order to determine the most appropriate model to predict a given flow. This study aims to provide a new form of quantitative assessment that accounts for both the physics captured by the turbulence model as well as the magnitude of the system response quantities (SRQ) using a modified symmetric mean absolute percent error (SMAPE) method. This method is designed to be approachable to researchers at any level and can be applied to system response quantities from multiple research fields. Uncertainty quantification is also performed to determine the discretization error for each turbulence model. Recommendations are made as to which turbulence models best capture the physics – hydrodynamically and thermally – using both local and global validation metrics. Lastly, a sensitivity analysis is performed on the damping functions used in the most accurate models. This underpins the potential of model developments and adjustments most worth pursuing for buoyant flows. Finally, this framework provides a more physics-based comparative analysis of the selected turbulence models.
This report documents the methods and results of a modeling study on a series of data generated by 1-foot test columns for the Water Detritiation project. In order to compliment modeling work already in progress on this project, a purely empirical curve-fitting approach is used in this study to determine chromatographic performance indicators such as retention time, curve width, and the number of chromatographic stages. The data used in this study comes from four sets of chromatographic test separations performed in March through May of 2017 using 1-foot columns. All are using the same flow rates and temperature (room temperature, nominally 25 C°). The first set of data (referred to in this report as the Column 1 data) is a series of twelve breakthrough curve measurements. Prior to each of these twelve runs, the column was regenerated by flowing hydrogen gas through the column, for increasing amounts of time, ranging from 2 hours to 25 hours. The second set of data (referred to as the Column 2 data) is comprised of five measurements of breakthrough curves measured in similar fashion as the Column 1 data, across a range of regeneration times from 2 to 21 hours. The third set of data (referred to as the Purge data) is of three breakthrough curves, measured after a regeneration time of 8 hours, which takes place after a nitrogen purge of 15, 30, or 45 minutes. The fourth data set is a series of five breakthrough curve measurements using a 2 nd batch of stationary phase, taken after regeneration times ranging from 2 to 20 hours. The overall modeling approach is as follows. First, the data is pre-processed for formatting, artifact removal, and is normalized. The data is then fit by integrated versions of four common chromatographic models. The best model is chosen based on the predictive performance of the model, and chromatographic performance values are calculated based on the best model’s fit parameters. Each of these processes are detailed below.
Engineering defective structures in an attempt to modify properties is an established technique in materials chemistry, yet, no models exist which can predict the structure of perovskite compounds containing extrinsic point defects such as vacancies. An empirically derived predictive model, based solely on chemical composition and published ionic radii has been developed. Effective vacancy sizes were derived both empirically from an existing model for pseudocubic lattice-constants, as well as experimentally, from average bond lengths calculated from neutron diffraction data. Compounds of lanthanum-doped barium titanate and strontium-doped magnesium titanate were synthesized with vacancies engineered on the A and B sites. Effective vacancy sizes were then used in empirical models to predict changes in lattice constants. Experimentally refined bond lengths used in the derivation of an effective vacancy size seemed to overestimate the effect of the point defects. Conversely, using calculated vacancy sizes, derived from a previously reported predictive model, showed significant improvements in the prediction of the pseudocubic perovskite lattice. - Graphical abstract: Atomistic model of Sr{sub 0.3}Nd{sub 0.7}Mg{sub 0.35}Ti{sub 0.65}O{sub 3} and Rietveld refinement of neutron diffraction data. - Highlights: • Defective perovskites were synthesized using the organic steric entrapment method. • Oxygen tilt systems were solved through X-ray, electron, and neutron diffraction. • An effective vacancy size for the cations was calculated from experimental bond lengths. • Discrepancies between Shannon radii and experimental measurements are explored. • An empirical model for predicting a{sub pc}, with an absolute error of 0.20%, was developed.
Analytical and empirical solutions to engineering problems are usually preferred because of their convenience in applications. However, they are not always accessible in complex problems. A new class of solutions, based on machine learning (ML) models such as regression trees and neural networks (NNs), are proposed and their feasibility and value are demonstrated through the analysis of fracture toughness measurements. It is found that both solutions based on regression trees and NNs can provide accurate results for the specific problem, but NN-based solutions outperform regression-tree-based solutions in terms of their simplicity. Here this example demonstrates that ML solutions are a major improvement over analytical and empirical solutions in terms of both reliable functionality and rapid deployment. When analytical solutions are not available, the use of ML solutions can overcome the limitations of empirical solutions and substantially change the way that engineering problems are solved.
Urea-selective catalytic reduction (SCR) catalysts are regarded as the leading NOx aftertreatment technology to meet the 2010 NOx emission standards for on-highway vehicles running on heavy-duty diesel engines. However, issues such as low NOx conversion at low temperature conditions still exist due to various factors, including incomplete urea thermolysis, inhibition of SCR reactions by hydrocarbons and H2O. We have observed a noticeable reduction in the standard SCR reaction efficiency at low temperature with increasing water content. We observed a similar effect when hydrocarbons are present in the stream. This effect is absent under fast SCR conditions where NO ~ NO2 in the feed gas. As a first step in understanding the effects of such inhibition on SCR reaction steps, kinetic models that predict the inhibition behavior of H2O and hydrocarbons on NO oxidation are presented in the paper. A one-dimensional SCR model was developed based on conservation of species equations and was coded as a C-language S-function and implemented in Matlab/Simulink environment. NO oxidation and NO2 dissociation kinetics were defined as a function of the respective adsorbate’s storage in the Fe-zeolite SCR catalyst. The corresponding kinetic models were then validated on temperature ramp tests that showed good match with the test data. Such inhibition models will improve the accuracy of model based control design for integrated DPF-SCR aftertreatment systems.
At Washington State University (WSU), a modern cyber-physical system testbed has been implemented based on an industry grade distribution management system (DMS) that is integrated with remote terminal units (RTUs), smart meters, and a solar photovoltaic (PV). In addition, the real model from the Avista Utilities distribution system in Pullman, WA, is modeled in DMS. The proposed testbed environment allows students and instructors to utilize these facilities for innovations in learning and teaching. For power engineering education, this testbed helps students understand the interaction between a cyber system and a physical distribution system through industrial level visualization. The testbed provides a distribution system monitoring and control environment for students. Compared with a simulation based approach, the testbed brings the students' learning environment a step closer to the real world. The educational modules allow students to learn the concepts of a cyber-physical system and an electricity market through an integrated testbed. Furthermore, the testbed provides a platform in the study mode for students to practice working on a real distribution system model. Here, this paper describes the new educational modules based on the testbed environment. Three modules are described together with the underlying educational principles and associated projects.
We present an engineering-oriented summary of statistical methods for estimation of the reliability (or equivalently, failure probability) of one-shot devices such as explosive detonators and other weapon components. Estimates may be given as single points or intervals, based on pass/fail tests, margin analysis, computational models, expert judgment, or a combination of these. We focus on highly reliable devices for which few or no failures are expected to occur in testing.
As systems become more complex, systems engineers rely on experts to inform decisions. There are few experts and limited data in many complex new technologies. This challenges systems engineers as they strive to plan activities such as qualification in an environment where technical constraints are coupled with the traditional cost, risk, and schedule constraints. Bayesian network (BN) models provide a framework to aid systems engineers in planning qualification efforts with complex constraints by harnessing expert knowledge and incorporating technical factors. By quantifying causal factors, a BN model can provide data about the risk of implementing a decision supplemented with information on driving factors. This allows a systems engineer to make informed decisions and examine “what-if” scenarios. This paper discusses a novel process developed to define a BN model structure based primarily on expert knowledge supplemented with extremely limited data (25 data sets or less). The model was developed to aid qualification decisions—specifically to predict the suitability of six degrees of freedom (6DOF) vibration testing for qualification. The process defined the model structure with expert knowledge in an unbiased manner. Finally, validation during the process execution and of the model provided evidence the process may be an effective tool in harnessing expert knowledge for a BN model.
Accelerating the Design-Build-Test-Learn (DBTL) cycle in synthetic biology is critical to achieving rapid and facile bioengineering of organisms for the production of, e.g., biofuels and other chemicals. The Learn phase involves using data obtained from the Test phase to inform the next Design phase. As part of the Learn phase, mathematical models of metabolic fluxes give a mechanistic level of comprehension to cellular metabolism, isolating the principle drivers of metabolic behavior from the peripheral ones, and directing future experimental designs and engineering methodologies. Furthermore, the measurement of intracellular metabolic fluxes is specifically noteworthy as providing a rapid and easy-to-understand picture of how carbon and energy flow throughout the cell. Here, we present a detailed guide to performing metabolic flux analysis in the Learn phase of the DBTL cycle, where we show how one can take the isotope labeling data from a 13 C labeling experiment and immediately turn it into a determination of cellular fluxes that points in the direction of genetic engineering strategies that will advance the metabolic engineering process.For our modeling purposes we use the Joint BioEnergy Institute (JBEI) Quantitative Metabolic Modeling (jQMM) library, which provides an open-source, python-based framework for modeling internal metabolic fluxes and making actionable predictions on how to modify cellular metabolism for specific bioengineering goals. It presents a complete toolbox for performing different types of flux analysis such as Flux Balance Analysis, 13 C Metabolic Flux Analysis, and it introduces the capability to use 13 C labeling experimental data to constrain comprehensive genome-scale models through a technique called two-scale 13 C Metabolic Flux Analysis (2S- 13 C MFA) [1]. In addition to several other capabilities, the jQMM is also able to predict the effects of knockouts using the MoMA and ROOM methodologies. The use of the jQMM library is illustrated through a step-by-step demonstration, which is also contained in a digital Jupyter Notebook format that enhances reproducibility and provides the capability to be adopted to the user's specific needs. As an open-source software project, users can modify and extend the code base and make improvements at will, providing a base for future modeling efforts.
The application of asphalt materials in pavement engineering has been increasingly widespread and sophisticated over the past several decades. Variations in the properties of asphalt binder during mixing, transportation, and paving can affect the performance of asphalt pavement. However, the asphalt material is a non-homogeneous and complex organic substance, consisting of various molecules with widely various molecular weights, elemental compositions, and structures. This complexity leads to difficulties for researchers to clearly and immediately understand the properties of asphalt materials and their variations. The multi-scale research approach combines macroscopic experimental data and microscopic simulation results from a practical engineering perspective. It helps to improve the understanding of asphalt materials. The molecular dynamics (MD) simulation proposes a corresponding molecular model of asphalt material based on experimental data, and the simulation algorithm is able to derive properties similar to those of real asphalt. Here, this paper provides a comprehensive review of the current studies on MD simulation of asphalt materials, including modeling, properties, and multi-scale analysis. As a key part of the computational simulation, this paper discusses the typical asphalt binder and asphalt-aggregate interface models constructed by different groups, and also presents their differences from real samples and their feasibility based on fundamental properties. After the introduction of molecular models, the extensive work made by researchers based on molecular models is categorically reviewed and discussed. The strengths and weaknesses of MD simulation methods in the study of asphalt materials are also summarized in order to provide the reader with a more comprehensive understanding of the relevant contents and to guide subsequent research.
In this work, the deformation behavior of irradiated tungsten at elevated service temperatures is examined using mechanism-based crystal plasticity finite element framework by considering multiple mechanisms, including thermal softening and irradiation hardening. The previously developed mechanism-based crystal plasticity model is first utilized to assess the yield stress of unirradiated and irradiated tungsten upon low-temperature irradiation. Additionally, the engineering stress-strain response under uniaxial compression is then predicted by considering the effect of irradiation-induced defect clusters on the motion of deformation-induced dislocations, and the predicted results agree well with the experimental measurements. More importantly, the formation and evolution of plastic strain localization zones in the form of shear bands are qualitatively captured, and the maximal principal strain distribution is quantitatively linked to the dose of irradiation damage. Finally, the deformation-induced shear band as a result of plastic instability in the compressed tungsten specimen is discussed.
We report a theoretical equation of state (EOS) table for boron across a wide range of temperatures (5.1 10 4 {5.2 10 8 K) and densities (0.25{49 g/cm 3 ), and experimental shock Hugoniot data at unprecedented high pressures (5608 118 GPa). The calculations are performed with first-principles methods combining path integral Monte Carlo (PIMC) at high temperatures and density functional theory molecular dynamics (DFT-MD) methods at lower temperatures. PIMC and DFT-MD cross-validate each other by providing coherent EOS (difference <1.5 Hartree/boron in energy and <5% in pressure) at 5.1 10 5 K. The Hugoniot measurement is conducted at the National Ignition Facility using a planar shock platform. The pressure-density relation found in our shock experiment is on top of the shock Hugoniot profile predicted with our first-principles EOS and a semi-empirical EOS table (LEOS 50). We investigate the self diffusivity and the effect of thermal and pressure-driven ionization on the EOS and shock compression behavior in high pressure and temperature conditions. We also study the sensitivity of a polar direct-drive exploding pusher platform to pressure variations based on applying pressure multipliers to LEOS 50 and by utilizing a new EOS model based on our ab initio simulations via 1D radiation-hydrodynamic calculations. The results are valuable for future theoretical and experimental studies and engineering design in high energy density research.
Purpose: The clinical test cases presented in AAPM TG-119 are used to evaluate the accuracy of treatment planning and delivery through spot scanning proton beams. Methods: An IBA spot scanning delivery system has been commissioned to be used with the RayStation treatment planning system. Various test cases provided in TG-119 were used for planning and delivery verification. The CT dataset and structures as provided by TG-119 were imported into a mock patient. The plans were optimized using the multi field optimization (MFO) to achieve the desired goals. The planner was given the flexibility to achieve the given dose-volume goals by creating appropriate objectives and constraints. Beams were delivered to a phantom and measurements were performed at multiple depths using the MatrixxPT detector array. The analyses were performed on beam by beam basis and quantified using the gamma index. A tolerance of 3%/3 mm in 2D was used for gamma index analysis along with dose threshold of 10%. Results: The clinical goals for targets and critical structures were met or improved for all cases except the C-Shape target with difficult constraints. The minimum gamma index using the 3%/3mm as a criterion is 93.3% for one of the planes measured for C-Shape target. Using 2%/2mm as a criterion, the minimum gamma index drops to 70%. Only Prostate target has all the planes above >90% pass using the 2%/2mm criterion. Conclusion: The overall accuracy of the treatment planning and delivery is deemed clinically acceptable. The test cases with highly modulated beams can have steep gradients in the dose profiles that can reduce the gamma index pass rate. Gamma analysis based on 3D data may be needed for routine use of 2%/2mm criterion. In addition, improvements in modelling of spot profiles in dose engine may be required for further improving the gamma index pass rate.
This software includes the Jupyter notebooks, model pretraining and evaluation code for the EXPERT 2.0 Human-AI Reasoning Engine V0.1. It supports pre-training a Human-AI model for reasoning over multi-layer network representations. It includes prompt-based evaluation framework in a Jupyter notebook for AI reasoning and Jupyter widgets with AI-based techniques for evidence generation and uncertainty quantification to support human-AI reasoning.
SAND2025-07450O Impala is a software tool that performs Bayesian model calibration for emulator-based models. The program uses R code to perform the calculations. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.
At nuclear power plants (NPPs), anomaly detection and identification (i.e., determining the causes of anomalies) are important tasks for ensuring the safe and efficient operation of NPPs. These tasks are currently labor-intensive and costly, and are made more difficult by the size and complexity of NPP systems. An alternative approach to conducting these tasks is to automate them, such as via the reconstruction-based contribution method, which is a well-researched unsupervised machine learning method that uses a data-driven model of anomaly-free behavior to detect events and then identify each variable’s contributions to those events. The present effort developed a novel contribution approach that utilized a leave-one-variable-out (LOVO) model, with which each variable is predicted using all the other variables. The novelty lay in transforming this model into a reconstruction model and modifying the identification algorithm to work with the new reconstruction model. To evaluate this method in a controlled environment, a synthetic dataset based on spring-mass-damper (SMD) systems (commonly found in mechanical engineering references) was used, with known anomalies introduced into the system. The proposed method successfully detected the anomalies and afforded insights into their causes, thus enabling the appropriate identifications to be made.