Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “generalizability”

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 631 records · Page 35
All sourcesNASA NTRSDOE OSTI
NASA NTRS2023

Multiyear Dry Periods in Southern Africa

Characteristics and physical features related to low precipitation across many years in Southern Africa that lead to societal disruptions are diagnosed using observed analyses and an ensemble of historical coupled climate model simulations during 1921 to 2014. Four regions are evaluated, as identified through a hierarchical clustering algorithm applied to the Standardized Precipitation Index (SPI) during the October–April precipitation season. Although dryness spanning many October–April occurs periodically in each region, they seldom occur simultaneously, consistent with largely insignificant SPI cross-correlations between them. However, characteristics relevant to low precipitation across many years are generalizable between the four regions, including the serial persistence of October–April precipitation, the likelihood of consecutive dry October–April, and the likelihood of dry October–April in temporal extents of up to 10 consecutive such 7-month seasons. Systematic precipitation persistence is not a feature in any of the four Southern Africa regions, as serial correlations of October–April SPI are not statistically significant at any time lags. It follows that there is an exponential-folding decay in the likelihood of consecutive October–April for various SPI thresholds and that there is a large spread in the likelihood of low October–April SPI across many years. In terms of physical features, low October–April SPI in each Southern Africa region is closely related to local atmospheric circulations; however, they are not as closely related to sea surface temperatures (SSTs). These results suggest that dryness spanning many years is determined primarily by persistent local circulations related to atmospheric variability and to a lesser extent variability related to SST anomalies, including the El Niño–Southern Oscillation.

subtropical Indian Ocean dipole↗
NASA NTRS2024

Recommendations on Evidence and Process for Certification of Learning-enabled Components in Aerospace Systems

This report primarily identifies a collection of relevant and necessary evidence for assurance of machine learnt components (MLCs)—also known as learning-enabled components—integrated into aircraft systems, and gives preliminary suggestions on the elements of a certification process that invoke the identified evidence. The main focus is on feedforward neural networks that are static and trained offline through supervised learning. A brief background on the generic elements of the lifecycle of an MLC is given to contextualize the assurance considerations and, consequently, the evidence that is relevant and necessary to support certification. At the level of an MLC, those considerations relate to: (i) the consistency and correctness of MLC contributions to system functions in the context of a validated functional intent; and (ii) the absence of MLC contributions to aircraft-level failure conditions. At an ML model level, confidence in model and data properties contribute to assurance of the containing MLC, in particular: (a) generalizability and robustness of models, in the presence of inputs not previously seen during training, disturbances to inputs, and unexpected inputs; and (b) valid data, i.e., data that are at least representative, relevant, complete, and accurate. Evidence for the above span the elements of the ML lifecycle, and includes, at a minimum, lifecycle artifacts that pertain to: (1) properties of requirements capturing functional intent, safety constraints, and aspects of the intended use and operating environment; (2) model performance, model complexity and design, and algorithm choice; (3) achievement of required performance at the levels of a trained model during model development, a trained model after model development is complete, and a trained model that is transformed into an executable equivalent; (4) model implementation aspects necessary for transforming a trained model into the executable equivalent; (5) integration of the executable trained model into the containing MLC, and eventually the larger system; and, (6) lastly, the verification and validation (V&V) of each of the above. Such V&V lifecycle artifacts themselves include: aspects of coverage, e.g., of various levels of requirements by the input space of the model and the data; traceability (where applicable); application of formal methods for property specification, analysis, and checking. Examples of evidence generation methods and tools further ground the discussion on what constitutes evidence, and the contribution to assurance during certification. The identified assurance considerations and supporting evidence is not a comprehensive set. Additionally, neither what should be considered as sufficient evidence relative to the assigned criticality of an MLC, nor how criticality ought to be determined and adjusted, have been considered in this report. However, suggestions are made for potential activities of the ML lifecycle that are aimed at providing confidence that an MLC can be relied upon when integrated into its containing (aircraft) system. Those activities are proposed as candidate elements of a certification process for MLCs. The main purpose of this report to inform regulatory guidance and consensus standards that may be used to meet the safety intent of the applicable regulations.

Aviation safety↗
NASA NTRS2023

Real Options Analysis for Valuation of Climate Adaptation Pathways With Application to Transit Infrastructure

Climate change and sea level rise (SLR) are expected to increase the frequency and intensity of coastal flood events, posing risks to coastal communities and infrastructure. While regional climate adaptation investments can provide substantive flood protection, existing plans often neglect uncertainty in future climate conditions and adaptation performance, consequently neglecting the option value of flexibly implementing proposed projects. Addressing this gap, we develop and employ a generalizable real options analysis (ROA) valuation framework that considers how uncertainty in adaptation project costs, SLR, flood severity, and flood losses inform the full range of adaptation performance outcomes. We further propose and apply a novel, computationally efficient flood loss sampling algorithm to estimate the consequences of randomly arriving coastal flood events. We apply this ROA framework to assess the option value of flexibly timing adaptation investments over time, investigating an adaptation pathway proposed by the City of Boston from the perspective of the regional transit system manager. Our results suggest that flexible implementation can provide significant option value in the near-to mid-term(>30 years), with highest option values under low-probability, high consequence scenarios. Our results also suggest adaptation pathway performance in the latter half of the 21stcenturyis most sensitive to uncertainty in sea level rise, flood loss estimates, and flood frequency, underscoring the importance of uncertainty quantification in the long-term valuation of adaptation investments.

Michael V. Martello↗
NASA NTRS2024

Augmenting RANS Turbulence Models Guided by Field Inversion and Machine Learning

This report investigates the use of a data-driven approach, viz., Field Inversion and Machine Learning (FIML), to improve conventional RANS turbulence models like the Spalart-Allmaras model and the Menter SST k-ω model. One of the crucial aspects of using an ML-based approach with limited training data to produce corrections that are generalizable to a large range of flow configurations is to design appropriate “features” (inputs to the ML model). A model, based on guidance from the FIML methodology, is presented in analytical form. An additional list of potential features is provided. Although these were not used in the present correction, they were considered in the course of its development, and are included to fully document the complete process employed in the present work.

turbulence modeling↗
NASA NTRSDate not supplied

Improving Sim-to-Real Transfer in Vision-Based Robot Navigation Via Instance-Level GAN-Based Data Augmentation

Achieving robust vision-based robotic tasks requires large amounts of data, which are often difficult to obtain in real-world scenarios. Simulators and synthetic data offer a cost-effective alternative, but the visual gap between simulation and reality hinders the performance of models when deployed in real-world environments. In this paper, we present a data augmentation pipeline that integrates a foundation model (Segment Anything Model) with an unsupervised image-to-image translation model (CycleGAN) for instance-level domain transfer from simulation to reality. This pipeline enables the generation of realistic labeled data from synthetic images for training supervised machine learning models in vision-based navigation tasks. We evaluate our approach on real-world data for ego-vehicle pose estimation, a critical autonomous navigation task involving the prediction of cross-track position and heading angle relative to road center line markings. The results of our tests show that our GAN-based data augmentation pipeline significantly outperforms models trained solely on simulation data or on data processed with standard image augmentation methods for sim-to-real transfer, enhancing model robustness and generalizability in real-world scenarios. Our method provides a scalable and flexible data augmentation tool for leveraging large synthetic datasets to enhance vision-based robotic navigation tasks.

artificial intelligence↗
NASA NTRS2024

A Koopman-based residual modeling approach for the control of a soft robot arm

Soft robots are challenging to model and control due to their poorly defined kinematics and nonlinear dynamics. Recently, Koopman operator theory has been shown capable of constructing control-oriented soft robot models from data. However, building these models requires extensive data collection and they do not necessarily generalize well outside of the training observations. This paper presents a more data-efficient and generalizable approach to soft robot modeling that first identifies a physics-based Koopman model then supplements it with a data-driven residual Koopman model. The resulting combined model is linear and thus compatible with real-time model-based control techniques such as Model Predictive Control (MPC). The efficacy of the approach is demonstrated on several simulated systems and on a real soft robot arm, where it is shown to generate models that are more accurate than purely physics-based models and require less data to construct than purely data-driven models. Using a model-based controller, the soft arm is able to successfully track end effector trajectories, perform a pick-and-place task, and write on a dry-erase board, showcasing the applicability of this framework to increase the capabilities of soft robotic systems.

Daniel Bruder↗
DOE OSTI2020

Kinetics Modeling for Design of Continuous Enzymatic Hydrolysis

Enzymatic hydrolysis of cellulose to monomeric sugars continues to be a limiting step in cost-effectively producing sugar and fermentation-based biofuels from biomass. In particular, the high cost of enzymes coupled with the long time-scale of reaction pose challenges to economic viability. Performing enzymatic hydrolysis in a continuous mode with enzyme recycle may provide a path towards substantially reduced costs for sugar production, but design and analysis are complicated by a lack of suitable kinetics models. Computationally attainable models, such as fractal-based models, require knowledge of the reaction-history of the biomass, and are thus only suitable for describing the batch reactions from which they are derived. Fundamental models, while potentially more generalizable, are often too computationally intensive to use in process or reactor modeling. In this work, a phenomenological rate model is proposed based on a two-phase substrate representation. Good agreement is seen between batch and continuous enzymatic hydrolysis (CEH) experiment data, which validates the model and enables us to solve for reactor design parameters, such as CEH reactor size and stream flow rates, based on process variables like yield. This model is integrated with techno-economic analysis software to explore economic sensitivities. Important design optimizations and tradeoffs are identified and quantified, including the relative cost imposed by rate slowdown from sugar inhibition versus the cost to remove and concentrate sugars at a lower concentration. It also identifies, high-leverage avenues for further exploration, such as increasing the maximum feasible solids concentration, and sustaining high membrane flux and reliability.

09 BIOMASS FUELS↗
DOE OSTI2020

Real-time Intrusion Detection for High-bandwidth Research Networks using Unsupervised Deep Learning

Nowadays, data generated by large-scale scientific experiments are on the scale of petabytes per month. These data are transferred through dedicated high-bandwidth networks across distributed collaboration sites for processing, storage, and analysis. Like general purpose networks, research networks experience attacks. However, monitoring traffic in such high-bandwidth network environments is challenging with conventional deep packet inspection (DPI)- based network intrusion detection systems (NIDS). Recently, machine learning (ML) algorithms have shown potential in detecting anomalous patterns with better generalizability and faster speed. Compared to the general purpose networks, traffic movement in research network is less volatile and more likely to benefit from the ML-based anomaly analysis. We present ThunderSecure, an efficient AI-powered system for detecting anomalies in 100G research networks. ThunderSecure implements a high-throughput packet processing infrastructure using multi-cores and GPUs. It takes time-series statistics from a variety of network measurements and uses 2D Convolutional Neural Network (CNN) to explore both spatial and temporal dependencies in network data stream. Patterns of science data traffic are learned with a one-class neural network, which blends an Adversarial Autoencoder (AAE) with Gaussian mixture density estimation. Testing traffic flows exhibiting significant deviations from the learned baseline of normality are marked as anomalies. We trained ThunderSecure on hundreds of billions of science data packets recorded from a 100G research network at Fermi National Accelerator Laboratory. The detection performance was evaluated on traffic captured from the same research network days and weeks after the training with different types of attack flows injected. To the best of our knowledge, this is the first ML work in the area of network anomaly detection that has been validated on such extreme scale datasets. Results show that ThunderSecure can recognize science data traffic that are captured long after the training and make nearly certain detection on those with anomalous flows injected, even in the case when the anomaly-to-normal mixing ratio is 0.1%.

Gong, Qian↗
DOE OSTI2021

Quantifying Turbine-Level Risk to Golden Eagles Using a High-Fidelity Updraft Model and a Stochastic Behavioral Model

To minimize the effects of wind farms on Golden Eagle (Aquila chrysaetos) populations while enabling sustainable development of renewable energy resources, it is important to understand how eagles interact with atmospheric flows, terrain features, and anthropogenic structures. Models that predict migratory flight paths provide one tool that helps us grasp how the location of wind farms may influence interactions and impacts on migrating Golden Eagles. The current state-of-the-art in predicting migratory flight paths uses a deterministic fluid-flow analogy to predict eagle trajectory using only an orographic updraft potential computed from topographical features. This model does not take into account variables, such as thermal updrafts and time varying atmospheric conditions that are known to influence migratory behavior. In this work, we improve on the model with the objective of developing tools that advance our understanding of how atmospheric flows and terrain features affect migratory eagle behavior and their interactions with wind farms. Specifically, we 1) incorporate both orographic and thermal updraft information in simulating eagle flight paths; 2) incorporate stochasticity into eagle travel patterns to better capture the influence of exogenous factors on, and the inherent stochasticity of eagle behavior; 3) consider spatio-temporal atmospheric data at wind-farm-scale when computing updraft potential; and 4) account for how atmospheric conditions and the direction of migration change seasonally and how these changes affect eagle migratory flight behavior. We tested the model using a 50km by 50km region with 50 m resolution in the western United States. We simulated 900 independent, probabilistic eagle tracks during southerly and northerly migration, assuming eagles solely rely on orographic updrafts. The preliminary results indicate that the inclusion of finer resolution atmospheric data allows for the inclusion of realistic conditions that an eagle experiences. The stochasticity in eagle tracks provides a platform to include uncertainty in eagle decision making and help produce robust eagle presence maps. We will deploy updraft and downdraft velocities computed using a high-fidelity, wind farm scale, computational fluid dynamics solver under development at National Renewable Energy Laboratory. This work is a first step in the development of a predictive and generalizable eagle behavior model at the wind farm scale that does not rely on empirical data collection. Although the current model is intended for migratory eagles, we will extend and refine this model to inform the development of additional behavioral modes, including resident eagle behavior. This modeling approach improves our ability to understand eagle use of the landscape at a fine scale, and it is our hope that this work will ultimately help advance strategies that minimize the impact of wind development on Golden Eagle populations.

49 EE - Wind and Water Power Program - Wind (EE-4W↗
DOE OSTI2022

Anderson Acceleration for Distributed Training of Deep Learning Models

Anderson acceleration (AA) is an extrapolation technique that has recently gained interest in the deep learning (DL) community to speed-up the sequential training of DL models. However, when performed at large scale, the DL training is exposed to a higher risk of getting trapped into steep local minima of the training loss function, and standard AA does not provide sufficient acceleration to escape from these steep local minima. This results in poor generalizability and makes AA ineffective. To restore AA’s advantage to speed-up the training of DL models on large scale computing platforms, we combine AA with an adaptive moving average procedure that boosts the training to escape from steep local minima. By monitoring the relative standard deviation between consecutive iterations, we also introduce a criterion to automatically assess whether the moving average is needed. We applied the method to the following DL instantiations for image classification: (i) ResNet50 trained on the open-source CIFAR100 dataset and (ii) ResNet50 trained on the open-source ImageNet1k dataset. Numerical results obtained using up to 1,536 NVIDIA V100 GPUs on the OLCF supercomputer Summit showed the stabilizing effect of the moving average on AA for all the problems above.

Lupo Pasini, Massimiliano↗
DOE OSTI2022

DERMS Online: A New Voltage Sensitivity-Enabled Feedback Optimization Framework: Preprint

This paper proposes a distributed energy resource management system (DERMS) solution by developing a new voltage sensitivity enabled feedback optimization framework. The key idea is to adopt a measurement feedback scheme to reformulate the original nonlinear optimization into a linear programming (LP) problem via perturb and observe-based voltage sensitivity analysis. The proposed solution eliminates the dependence on load knowledge and can be implemented online thanks to an efficient open-source solver for LP problems. The proposed DERMS online platform is generalizable to deal with various types of distributed energy resources (DERs), including distributed photovoltaics (PVs), energy storage, electric vehicles, demand response, etc. Comparison results with other control methods on a realistic distribution feeder in Southern California highlight the feasibility as well as benefits for the proposed framework.

distributed energy resources management↗
DOE OSTI2022

Grid-Interactive Electric Vehicle and Building Coordination Using Coupled Distributed Control: Preprint

As an increasing number of controllable devices are introduced onto the grid, they can individually provide ancillary services in support of grid stability. However, the goals of each device differ due to their type and individual objectives, causing instances where they may conflict. To reduce the chances of these devices contributing to grid instability, these devices must effectively communicate in a cooperative manner to both meet their own needs while providing services to the grid. Previous work demonstrates that the NALD (Network Lasso-ADMM - Limited Communication - DMPC) algorithm allows coordination between two subsystems that use different control algorithms (building and charging stations) to provide services to the grid and individually optimize their performance in a specific scenario. The ideal NALD algorithm should be generalized to allow plug-and-play capabilities across devices of differing characteristics. This paper takes a step toward generalizability by updating the electric vehicle charging objective and re-defining the communication scheme compared to prior work to generalize the coordination and, as a result, improve the performance of the NALD algorithm.

ADVANCED PROPULSION SYSTEMS,ENERGY CONSERVATION, C↗
DOE OSTI2022

Redox-flow batteries employing oligomeric organic active materials and size-selective microporous polymer membranes

Intermittent energy sources, including solar and wind, require scalable, low-cost, multi-hour energy storage solutions to be effectively incorporated into the grid. Redox-flow batteries offer a solution, but suffer from rapid capacity fade and low Coulombic efficiency due to the high permeability of redox-active species across the battery's membrane. Here we show that active-species crossover can be arrested by scaling the membrane's pore size to molecular dimensions and in turn increasing the size of the active material to be above the membrane's pore-size exclusion limit. When oligomeric redox-active organic molecules were paired with microporous polymer membranes, the rate of active-material crossover was either completely blocked or slowed more than 9,000-fold compared to traditional separators at minimal cost to ionic conductivity. In the case of the latter, this corresponds to an absolute rate of ROM crossover of less than 3 μmol cm−2 day−1 (for a 1.0 M concentration gradient), which exceeds performance targets recently set forth by the battery industry. This strategy was generalizable to both high and low-potential ROMs in a variety of electrolytes, highlighting the importance of macromolecular design in implementing next-generation redox-flow batteries.

Helms, Brett A.↗
DOE OSTI2021

November Mega-AI Newsletter

Mega AI, an internal investment at Pacific Northwest National Laboratory, aims to develop next-generation artificial intelligence (AI) capabilities unique to the Department of Energy national lab complex to address research gaps in large-scale multimodal representation learning, multitask inferences, and the need for increased generalizability, rapid adaptivity, and usability of AI technologies. In this newsletter, we highlight recent developments in the research community on next-gen AI technologies focusing on massive-scale model development, deployment and evaluation, data and code availability, model interactions, and new features and capabilities that are relevant to Mega AI’s goals and science and security applications.

Volkova, Svitlana↗
DOE OSTI2022

Systems and methods for powder bed additive manufacturing anomaly detection

Detection and classification of anomalies for powder bed metal additive manufacturing. Anomalies, such as recoater blade impacts, binder deposition issues, spatter generation, and some porosities, are surface-visible at each layer of the building process. A multi-scaled parallel dynamic segmentation convolutional neural network architecture provides additive manufacturing machine and imaging system agnostic pixel-wise semantic segmentation of layer-wise powder bed image data. Learned knowledge is easily transferrable between different additive manufacturing machines. The anomaly detection can be conducted in real-time and provides accurate and generalizable results.

Scime, Luke R.↗
DOE OSTI2023

Electrochemical Reaction Engineering: A Lost (and Found) Art

The heart (or, perhaps more appropriately, the stomach) of any chemical manufacturing process is the reactor, where chemical and physical transformations occur under precisely controlled conditions. For non-electrochemical industrial processes, the design, optimization and operation of chemical reactors are extremely well developed, drawing upon the principles of chemistry, physics, calculus and economics. However, electrolytic processes face a dearth of well documented reaction engineering principles and design practices. This tutorial-style poster presentation introduces electrochemistry students and industrial practitioners to the "lost art" of reaction engineering and its organic extensions to electrochemical devices. Electrochemical reactor fundamentals and illustrative design equations are presented alongside a proposed five-step generalizable methodology for electrochemical reactor scale-up. Overall, this contribution seeks to educate electrochemistry researchers on practical means of reducing laboratory technologies to commercial practice.

electrochemistry↗
DOE OSTI2024

Navigating Urban Mobility: Evaluating Infrastructure Strategies for Enhanced Energy-Efficient Access for Micromobility

Cities and communities continue to expand pedestrian and bicycle infrastructure as part of their sustainable mobility and post-pandemic recovery plans. However, the emergence of micro-mobility (e.g., electric bicycles) has created new challenges for urban transport planning. As the popularity of micro-mobility modes has grown, so have safety concerns due to rising injuries, thus challenging many cities to come up with regulatory measures that enable efficient access while minimizing negative impacts from micro-mobility. Leveraging the Open-Source Tool (Mobility Energy Productivity metric), powered by an open-source dataset (OpenStreetMap), and enhanced through the incorporation of perceived discomfort (level-of-traffic stress), this research study focuses on evaluating the accessibility implications of infrastructure planning and regulatory measures for micro-mobility. Five scenarios pertaining to level of traffic stress, sidewalk access, traffic calming, and bike lane coverage were tested in the Denver-Aurora region in Colorado. Maximum improvements in energy-efficient access are realized when allowing sidewalks for micro-mobility use. Cities and planning agencies could leverage this information to assess sidewalk use policies for micro-mobility, while ensuring pedestrian safety and ADA access. Results indicate that expansion of bicycle lane coverage yields 11% accessibility benefits for micro-mobility compared to implementing traffic calming measures yielding 3% accessibility improvements. Further, it was observed most of the population in the Denver-Aurora region is experiencing lower accessibility in-part due to presence of a high-stress connections in the network. Although not generalizable, the use of open-source data and access calculation methodology, make this analysis reproducible and transferable to other locations.

ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2023

Enabling Interconnected Science Workflows through an Adapter Approach

The INTERSECT Software framework project aims to create an open federated library that connects, coordinates, and controls systems in the scientific domain. It features the Adapter, a flexible and extensible interface inspired by the Adapter design pattern in object-oriented programming. By utilizing Adapters, the INTERSECT SDK enables effective communication and coordination within a diverse ecosystem of systems. This adaptability facilitates the execution of complex operations within the framework, promoting collaboration and efficient workflow management in scientific research. Additionally, the generalizability of Adapters and their patterns enhances their utility in other scientific software projects and challenges.

McGaha, Jesse↗