Search NASA⌕ Search

SEARCH · Search NASA

Results for “layered decision model”

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 19 records

A Layered Decision Model for Cost-Effective System Security

System security involves decisions in at least three areas: identification of well-defined security policies, selection of cost-effective defence strategies, and implementation of real-time defence tactics. Although choices made in each of these areas affect the others, existing decision models typically handle these three decision areas in isolation. There is no comprehensive tool that can integrate them to provide a single efficient model for safeguarding a network. In addition, there is no clear way to determine which particular combinations of defence decisions result in cost-effective solutions. To address these problems, this paper introduces a Layered Decision Model (LDM) for use in deciding how to address defence decisions based on their cost-effectiveness. To validate the LDM and illustrate how it is used, we used simulation to test model rationality and applied the LDM to the design of system security for an e-commercial business case.

99 GENERAL AND MISCELLANEOUS↗

A machine learning approach for efficient multi-dimensional integration

Many physics problems involve integration in multi-dimensional space whose analytic solution is not available. The integrals can be evaluated using numerical integration methods, but it requires a large computational cost in some cases, so an efficient algorithm plays an important role in solving the physics problems. We propose a novel numerical multi-dimensional integration algorithm using machine learning (ML). After training a ML regression model to mimic a target integrand, the regression model is used to evaluate an approximation of the integral. Then, the difference between the approximation and the true answer is calculated to correct the bias in the approximation of the integral induced by ML prediction errors. Because of the bias correction, the final estimate of the integral is unbiased and has a statistically correct error estimation. Three ML models of multi-layer perceptron, gradient boosting decision tree, and Gaussian process regression algorithms are investigated. The performance of the proposed algorithm is demonstrated on six different families of integrands that typically appear in physics problems at various dimensions and integrand difficulties. The results show that, for the same total number of integrand evaluations, the new algorithm provides integral estimates with more than an order of magnitude smaller uncertainties than those of the VEGAS algorithm in most of the test cases.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Modeling protected species distributions and habitats to inform siting and management of pioneering ocean industries: A case study for Gulf of Mexico aquaculture

Marine Spatial Planning (MSP) provides a process that uses spatial data and models to evaluate environmental, social, economic, cultural, and management trade-offs when siting (i.e., strategically locating) ocean industries. Aquaculture is the fastest-growing food sector in the world. The United States (U.S.) has substantial opportunity for offshore aquaculture development given the size of its exclusive economic zone, habitat diversity, and variety of candidate species for cultivation. However, promising aquaculture areas overlap many protected species habitats. Aquaculture siting surveys, construction, operations, and decommissioning can alter protected species habitat and behavior. Additionally, aquaculture-associated vessel activity, underwater noise, and physical interactions between protected species and farms can increase the risk of injury and mortality. In 2020, the U.S. Gulf of Mexico was identified as one of the first regions to be evaluated for offshore aquaculture opportunities as directed by a Presidential Executive Order. We developed a transparent and repeatable method to identify aquaculture opportunity areas (AOAs) with the least conflict with protected species. First, we developed a generalized scoring approach for protected species that captures their vulnerability to adverse effects from anthropogenic activities using conservation status and demographic information. Next, we applied this approach to data layers for eight species listed under the Endangered Species Act, including five species of sea turtles, Rice’s whale, smalltooth sawfish, and giant manta ray. Next, we evaluated four methods for mathematically combining scores (i.e., Arithmetic mean, Geometric mean, Product, Lowest Scoring layer) to generate a combined protected species data layer. The Product approach provided the most logical ordering of, and the greatest contrast in, site suitability scores. Finally, we integrated the combined protected species data layer into a multi-criteria decision-making modeling framework for MSP. This process identified AOAs with reduced potential for protected species conflict. These modeling methods are transferable to other regions, to other sensitive or protected species, and for spatial planning for other ocean-uses.

54 ENVIRONMENTAL SCIENCES↗

Machine learning to predict final fire size at the time of ignition

Fires in boreal forests of Alaska are changing, threatening human health and ecosystems. Given expected increases in fire activity with climate warming, insight into the controls on fire size from the time of ignition is necessary. Such insight may be increasingly useful for fire management, especially in cases where many ignitions occur in a short time period. Here we investigated the controls and predictability of final fire size at the time of ignition. Using decision trees, we show that ignitions can be classified as leading to small, medium or large fires with 50.4 ± 5.2% accuracy. This was accomplished using two variables: vapour pressure deficit and the fraction of spruce cover near the ignition point. The model predicted that 40% of ignitions would lead to large fires, and those ultimately accounted for 75% of the total burned area. Other machine learning classification algorithms, including random forests and multi-layer perceptrons, were tested but did not outperform the simpler decision tree model. Applying the model to areas with intensive human management resulted in overprediction of large fires, as expected. This type of simple classification system could offer insight into optimal resource allocation, helping to maintain a historical fire regime and protect Alaskan ecosystems.

54 ENVIRONMENTAL SCIENCES↗

Hierarchical Transactive Control of Flexible Building Loads Under Distribution LMP

With grid modernization efforts, future distribution networks, which consist of various distributed generators and flexible loads, will be more flexible and active. All new network components of distributed energy resources (DERs) drive and enable the transition towards a market-based distribution net-work that seeks the optimal allocation of all DERs. To address challenges associated with DERs, one promising solution is to utilize demand-side flexibility of building loads facilitated by demand response (DR) programs and provide ancillary grid services through distribution-level markets. Under this new paradigm, this paper proposes an efficient DR management strategy incorporating emerging price signals of distribution markets, i.e., distribution locational marginal price (DLMP), based on a hierarchical transactive control approach. The proposed approach establishes a two-layer decision-making framework; the upper layer formulates a bilevel model to obtain an optimal demand response (ODR) under DLMP, and the lower layer employs model-free control to dispatch the (aggregated) ODR to individual end users. Numerical case studies using a modified IEEE 33 test network are performed to verify the effectiveness of the proposed approach; load shifting and peak shaving for the distribution system operator and payments’ reduction for end users while maintaining their comfort.

Park, Byungkwon↗

Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach

Wind speed prediction with spatio–temporal correlation is among the most challenging tasks in wind speed prediction. In this paper, the problem of predicting wind speed for multiple sites simultaneously is investigated by using spatio–temporal correlation. This paper proposes a model for wind speed prediction with spatio–temporal correlation, i.e., the predictive deep convolutional neural network (PDCNN). The model is a unified framework, integrating convolutional neural networks (CNNs) and a multi-layer perceptron (MLP). Firstly, the spatial features are extracted by CNNs located at the bottom of the model. Then, the temporal dependencies among these extracted spatial features are captured by the MLP. In this way, the spatial and temporal correlations are captured by PDCNN intrinsically. Finally, PDCNN generates the predicted wind speed by using the learnt spatio–temporal correlations. In addition, three error indices are defined to evaluate the prediction accuracy of the model on the wind turbine array. Experiment results on real-world data show that PDCNN can capture the spatio–temporal correlation effectively, and it outperforms the conventional machine learning models, including multi-layer perceptron, support vector regressor, decision tree, etc.

17 WIND ENERGY↗

Decayheatml

This code is designed to predict and analyze the decay heat generated in molten salt reactors (MSRs) using a hybrid approach that combines machine learning and segmented polynomial fitting. The accurate prediction of decay heat is essential for reactor safety and the optimization of spent fuel storage. The code operates through several key components: 1) Data Architecture: It incorporates a modular data architecture that handles various MSR-specific operational parameters such as power density, humidity content, and air ingress. These parameters are sampled using Sobol sequences to ensure comprehensive coverage of operational uncertainties. 2) Machine Learning Framework: The code employs a diverse set of machine learning models, including polynomial regression, decision trees, random forests, gradient boosting, support vector regression, k-nearest neighbors, multi-layer perceptrons, and symbolic regression. These models are trained to predict decay heat over a wide temporal range, from immediate shutdown up to 10,000 years. 3) Region-Optimized Training: The temporal domain is divided into multiple regions, each modeled separately to capture distinct decay heat characteristics across different time scales. This approach significantly improves the accuracy and interpretability of predictions. 4) Segmented Polynomial Interpretation (SPI): The SPI method translates machine learning predictions into piecewise polynomial equations. These equations are physically interpretable and can be directly integrated into existing engineering workflows and safety analyses. 5) Front-End Interfaces: The code includes both a Jupyter notebook interface for research development and a Streamlit web application for operational deployment. These interfaces allow users to interactively explore decay heat predictions, adjust operational parameters, and visualize results in real-time. 6) Applications: The framework supports various applications, including safety system validation and spent fuel container optimization. It enables real-time evaluation of worst-case decay heat scenarios, informing the design of passive safety systems and optimizing container designs for long-term storage. Overall, this code provides a robust, accurate, and user-friendly tool for predicting decay heat in MSRs, enhancing reactor safety, and optimizing spent fuel management.

Retamales, Mauricio Eduardo Tano [Idaho National L↗

Uncertainty quantification in machine learning for engineering design and health prognostics: A tutorial

On top of machine learning (ML) models, uncertainty quantification (UQ) functions as an essential layer of safety assurance that could lead to more principled decision making by enabling sound risk assessment and management. The safety and reliability improvement of ML models empowered by UQ has the potential to significantly facilitate the broad adoption of ML solutions in high-stakes decision settings, such as healthcare, manufacturing, and aviation, to name a few. In this tutorial, we aim to provide a holistic lens on emerging UQ methods for ML models with a particular focus on neural networks and the applications of these UQ methods in tackling engineering design as well as prognostics and health management problems. Towards this goal, we start with a comprehensive classification of uncertainty types, sources, and causes pertaining to UQ of ML models. Next, we provide a tutorial-style description of several state-of-the-art UQ methods: Gaussian process regression, Bayesian neural network, neural network ensemble, and deterministic UQ methods focusing on spectral-normalized neural Gaussian process. Established upon the mathematical formulations, we subsequently examine the soundness of these UQ methods quantitatively and qualitatively (by a toy regression example) to examine their strengths and shortcomings from different dimensions. Then, we review quantitative metrics commonly used to assess the quality of predictive uncertainty in classification and regression problems. Afterward, we discuss the increasingly important role of UQ of ML models in solving challenging problems in engineering design and health prognostics. In conclusion, two case studies with source codes available on GitHub are used to demonstrate these UQ methods and compare their performance in the life prediction of lithium-ion batteries at the early stage (case study 1) and the remaining useful life prediction of turbofan engines (case study 2).

97 MATHEMATICS AND COMPUTING↗

Deep convolutional autoencoders as generic feature extractors in seismological applications

The idea of using a deep autoencoder to encode seismic waveform features and then use them in different seismological applications is appealing. In this paper, we designed tests to evaluate this idea of using autoencoders as feature extractors for different seismological applications, such as event discrimination (i.e., earthquake vs. noise waveforms, earthquake vs. explosion waveforms), and phase picking. These tests involve training an autoencoder, either undercomplete or overcomplete, on a large amount of earthquake waveforms, and then using the trained encoder as a feature extractor with subsequent application layers (either a fully connected layer, or a convolutional layer plus a fully connected layer) to make the decision. By comparing the performance of these newly designed models against the baseline models trained from scratch, we conclude that the autoencoder feature extractor approach may only outperform the baseline under certain conditions, such as when the target problems require features that are similar to the autoencoder encoded features, when a relatively small amount of training data is available, and when certain model structures and training strategies are utilized. The model structure that works best in all these tests is an overcomplete autoencoder with a convolutional layer and a fully connected layer to make the estimation.

58 GEOSCIENCES↗

Estimating the influence of field inventory sampling intensity on forest landscape model performance for determining high-severity wildfire risk

Abstract Historically, fire has been essential in Southwestern US forests. However, a century of fire-exclusion and changing climate created forests which are more susceptible to uncharacteristically severe wildfires. Forest managers use a combination of thinning and prescribed burning to reduce forest density to help mitigate the risk of high-severity fires. These treatments are laborious and expensive, therefore optimizing their impact is crucial. Landscape simulation models can be useful in identifying high risk areas and assessing treatment effects, but uncertainties in these models can limit their utility in decision making. In this study we examined underlying uncertainties in the initial vegetation layer by leveraging a previous study from the Santa Fe fireshed and using new inventory plots from 111 stands to interpolate the initial forest conditions. We found that more inventory plots resulted in a different geographic distribution and wider range of the modelled biomass. This changed the location of areas with high probability of high-severity fires, shifting the optimal location for management. The increased range of biomass variability from using a larger number of plots to interpolate the initial vegetation layer also influenced ecosystem carbon dynamics, resulting in simulated forest conditions that had higher rates of carbon uptake. We conclude that the initial forest layer significantly affects fire and carbon dynamics and is dependent on both number of plots, and sufficient representation of the range of forest types and biomass density.

Science & Technology - Other Topics↗

Renewable Energy Data Explorer User Guide

This publication provides a user guide for the Renewable Energy Data Explorer and technical potential tool within the Explorer. The Renewable Energy Data Explorer is a dynamic, web-based geospatial analysis tool that facilitates renewable energy decision-making, investment, and deployment. It brings together renewable energy resource data and other modeled or measured geographic information system (GIS) layers, including land use, weather, environmental, population density, administrative, and grid data.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Repulsive Casimir forces and the role of surface modes

The Casimir repulsion between a metal and a dielectric suspended in a liquid has been thoroughly studied in recent experiments. In the present paper we consider surface modes in three layered systems modeled by dielectric functions guaranteeing repulsion. It is shown that surface modes play a decisive role in this phenomenon at short separations. For a toy plasma model we find the contribution of the surface modes at all distances.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Data-driven optimization of mixed-integer bi-level multi-follower integrated planning and scheduling problems under demand uncertainty

The coordination of interconnected elements across the different layers of the supply chain is essential for all industrial processes and the key to optimal decision-making. Yet, the modeling and optimization of such interdependent systems are still burdensome. Here we address the simultaneous modeling and optimization of medium-term planning and short-term scheduling problems under demand uncertainty using mixed-integer bi-level multi-follower programming and data-driven optimization. Bi-level multi-follower programs model the natural hierarchy between different layers of supply chain management holistically, while scenario analysis and data-driven optimization allow us to retrieve the guaranteed feasible solutions of the integrated formulation under various demand considerations. We address the data-driven optimization of this challenging class of problems using the DOMINO framework, which was initially developed to solve single-leader single-follower bi-level optimization problems to guaranteed feasibility. This framework is extended to solve single-leader multi-follower stochastic formulations and its performance is characterized by well-known single and multi-product process scheduling case studies. Through our data-driven algorithmic approach, we present guaranteed feasible solutions to linear and nonlinear mixed-integer bi-level formulations of simultaneous planning and scheduling problems and further characterize the effects of the scheduling level complexity on the solution performance, which spans over several hundred continuous and binary variables, and thousands of constraints.

42 ENGINEERING↗

Recrystallization and modification of the stainless-steel surface relief under photonic heat load in powerful plasma discharges

Targets made of ITER-grade 316L(N)-IG stainless steel and Russian-grade 12Cr18Ni10Ti stainless steel with a close composition were exposed at the QSPA-T plasma gun to plasma photonic radiation pulses simulating conditions of disruption mitigation in ITER. After a large number of pulses, modification of the stainless-steel surface was observed, such as the formation of a wavy structure, irregular roughness, and cracks on the target surface. X-ray and optic microscopic analyses of targets revealed changes in the orientation and dimensions of crystallites (grains) over a depth of up to 20 μm for 316L(N)-IG stainless steel after 200 pulses and up to 40 μm for 12Cr18Ni10Ti stainless steel after 50 pulses, which is significantly larger than the depth of the layer melted in one pulse (∼10 μm). In a series of 200 tests of ITER-grade 316L(N)-IG ITER stainless steel, a linear increase in the height of irregularity (roughness) with increasing number of pulses at a rate of up to ∼1 μm per pulse was observed. No alteration in the chemical composition of the stainless-steel surface in the series of tests was revealed. A model is developed that describes the formation of wavy irregularities on the melted metal surface with allowance for the nonlinear stage of instability of the melted layer with a vapor/plasma flow above it. A decisive factor in this case is the viscous flow of the melted metal from the troughs to tops of the wavy structure. The model predicts saturation of the growth of the wavy structure when its amplitude becomes comparable with its wavelength. Approaches to describing the observed stochastic relief and roughness of the stainless-steel surface formed in the series of tests are considered. The recurrence of the melting-solidification process in which mechanisms of the hill growth compete with the spreading of the material from the hills can result in the formation of a stochastic relief.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

NUCLEAR WASTE VITRIFICATION EFFICIENCY COLD CAP REACTIONS

The cost and schedule of nuclear waste treatment and immobilization are greatly affected by the rate of glass production. Various factors influence the performance of a waste-glass melter. One of the most significant, and also one of the least understood, is the process of batch melting. Studies are being conducted to gain fundamental understanding of the batch reactions, particularly those that influence the rate of melting, and models are being developed to link batch makeup and melter operation to the melting rate. Batch melting takes place within the cold cap, i.e., a batch layer floating on the surface of molten glass. The conversion of batch to glass consists of various chemical reactions, phase transitions, and diffusion-controlled processes. These include water evaporation (slurry feed contains as high as 60% water), gas evolution, the melting of salts, the formation of borate melt, reactions of borate melt with molten salts and with amorphous oxides (Fe{sub 2}O{sub 3} and Al{sub 2}O{sub 3}), the formation of intermediate crystalline phases, the formation of a continuous glass-forming melt, the growth and collapse of primary foam, and the dissolution of residual solids. To this list we also need to add the formation of secondary foam that originates from molten glass but accumulates on the bottom of the cold cap. This study presents relevant data obtained for a high-level-waste melter feed and introduces a one-dimensional (1D) mathematical model of the cold cap as a step toward an advanced three-dimensional (3D) version for a complete model of the waste glass melter. The 1D model describes the batch-to-glass conversion within the cold cap as it progresses in a vertical direction. With constitutive equations and key parameters based on measured data, and simplified boundary conditions on the cold-cap interfaces with the glass melt and the plenum space of the melter, the model provides sensitivity analysis of the response of the cold cap to the batch makeup and melter conditions. The model demonstrates that batch foaming has a decisive influence on the rate of melting. Understanding the dynamics of the foam layer at the bottom of the cold cap and the heat transfer through it appears crucial for a reliable prediction of the rate of melting as a function of the melter-feed makeup and melter operation parameters. Although the study is focused on a batch for waste vitrification, the authors expect that the outcome will also be relevant for commercial glass melting.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Improved Meteorological Input for Atmospheric Release Decision support Systems and an Integrated LES Modeling System for Atmospheric Dispersion of Toxic Agents: Homeland Security Applications

When hazardous material is accidently or intentionally released into the atmosphere, emergency response organizations look to decision support systems (DSSs) to translate contaminant information provided by atmospheric models into effective decisions to protect the public and emergency responders and to mitigate subsequent consequences. The Department of Homeland Security (DHS)-led Interagency Modeling and Atmospheric Assessment Center (IMAAC) is one of the primary DSSs utilized by emergency management organizations. IMAAC is responsible for providing 'a single piont for the coordination and dissemination of Federal dispersion modeling and hazard prediction products that represent the Federal position' during actual or potential incidents under the National Response Plan. The Department of Energy's (DOE) National Atmospheric Release Advisory Center (NARAC), locatec at the Lawrence Livermore National Laboratory (LLNL), serves as the primary operations center of the IMAAC. A key component of atmospheric release decision support systems is meteorological information - models and data of winds, turbulence, and other atmospheric boundary-layer parameters. The accuracy of contaminant predictions is strongly dependent on the quality of this information. Therefore, the effectiveness of DSSs can be enhanced by improving the meteorological options available to drive atmospheric transport and fate models. The overall goal of this project was to develop and evaluate new meteorological modeling capabilities for DSSs based on the use of NASA Earth-science data sets in order to enhance the atmospheric-hazard information provided to emergency managers and responders. The final report describes the LLNL contributions to this multi-institutional effort. LLNL developed an approach to utilize NCAR meteorological predictions using NASA MODIS data for the New York City (NYC) region and demonstrated the potential impact of the use of different data sources and data parameterizations on IMAAC/NARAC fate and transport predictions. A case study involving coastal sea breeze circulation patterns in the NYC region was used to investigate the sensitivity of atmospheric dispersion results on the source of three-dimensional wind field data.

54 ENVIRONMENTAL SCIENCES↗

PROCESS MONITORING FOR SAFEGUARDS VIA EVENT GENERATION, INTEGRATION, AND INTERPRETATION

There is a recognized safeguards benefit from using process monitoring (PM) on nuclear facilities to complement nuclear materials accountancy. We introduce a model-based approach for PM in which the assessment regarding the state of the monitored system is conducted at a system-centric level. The proposed architecture integrates both time-driven and event-driven data integration and analysis for decision-making. While the time-driven layers of the proposed architecture encompass more traditional PM methods based on time series data and analysis, the event-driven layers encompass operation monitoring methods based on discrete event data integration and analysis. By integrating process- and operation-related information and methodologies within an unified modeling and monitoring framework that includes not only current but also past plant behaviors, the task of anomaly detection is greatly improved because this decision-making approach can benefit from not only known time-series relationships among measured signals but also from known event sequence relationships among generated events. Building from the proposed system-centric PM architecture, we briefly introduce methods that can be used to implement its different components. The application of the proposed approach is then demonstrated via simulation experiments.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Multiscale topology optimization using neural network surrogate models

In this study, we are concerned with optimization of macroscale elastic structures that are designed utilizing spatially varying microscale metamaterials. The macroscale optimization is accomplished using gradient-based nonlinear topological optimization. But instead of using density as the optimization decision variable, the decision variables are the multiple parameters that define the local microscale metamaterial. This is accomplished using single layer feedforward Gaussian basis function networks as a surrogate models of the elastic response of the microscale metamaterial. The surrogate models are trained using highly resolved continuum finite element simulations of the microscale metamaterials and hence are significantly more accurate than analytical models e.g. classical beam theory. Because the derivative of the surrogate model is important for sensitivity analysis of the macroscale topology optimization, a neural network training procedure based on the Sobolev norm is described. Since the SIMP method is not appropriate for spatially varying lattices, an alternative method is developed to enable creation of void regions. Lastly, the efficacy of this approach is demonstrated via several examples in which the optimal graded metamaterial outperforms a traditional solid structure.

42 ENGINEERING↗