Search NASA⌕ Search

SEARCH · Search NASA

Results for “Learning control”

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 1,333 records · Page 74

INITIAL STUDIES OF CAVITY FAULT PREDICTION AT JEFFERSON LABORATORY

The Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Laboratory is a CW recirculating linac that utilizes over 400 superconducting radio-frequency (SRF) cavities to accelerate electrons up to 12 GeV through 5-passes. Recent work has shown that, given RF signals from a cavity during a fault as input, machine learning approaches can accurately classify the fault type. In this paper we report on initial results of predicting a fault onset using only data prior to the failure event. A data set was constructed using time-series data immediately before a fault (’unstable’) and 1.5 seconds prior to a fault (’stable’) gathered from over 5,000 saved fault events. The data was used to train a binary classifier. The results gave key insights into the behavior of several fault types and provided motivation to investigate whether data prior to a failure event could also predict the type of fault. We discuss our method using a sliding window approach and report on initial results. Recent modifications to the low-level RF control system will provide access to streaming signals and we outline a path forward for leveraging deep learning on streaming data

Vidyaratne, Lasitha↗

Hybrid geological modeling: Combining machine learning and multiple-point statistics

Accurately modeling and constructing a geologically realistic subsurface model remains an outstanding problem as the morphology controls the flow behaviors. Particularly, one of the pattern-based methods, namely cross-correlation based simulation, has been proved to be an effective way to reconstruct a realistic model, at both small and large scales. However, conditioning to point data in the large-scale problems is still a crucial issue in these algorithms, since there is always a trade-off between the quality of the realizations and the degree of point data reproduction. Specifically, it is not practical to build a training image (TI) which includes all the possibilities and variabilities. Therefore, finding a pattern that can represent the point data and, at the same time, preserving the connectivities is difficult. This leads to producing highly-connected realizations with a significant mismatch or poor models with a reasonable degree of point data reproduction. To accurately reproduce the densely distributed hard data, pixel-based methods can also produce some unrealistic artifacts around the hard data. In this paper, to overcome this challenge, however, we use pattern-based methods as they often produce more disconnected geobodies when dealing with dense hard data, and proposed a hybrid algorithm using the pattern-based methods and convolutional neural network (CNN). The trained CNN model is utilized to improve the quality of conditioning to point data for the original realizations generated by the pattern-based algorithm. As such, the mismatch locations are identified, and the same regions are used in the training of CNN to mimic the procedure through which a missing region can be filled. To evaluate the performance of the proposed hybrid algorithm, it is tested on cases with different dimensions and different numbers of facies. Then, the newly improved realizations are compared with the initial realizations generated by the pattern-based algorithm. The comparison is also conducted by the flow simulation test. And it indicates that the proposed hybrid algorithm can better reproduce the point data, while the connectivities are better preserved.

58 GEOSCIENCES↗

A data-driven global soil heterotrophic respiration dataset and the drivers of its inter-annual variability

Soil heterotrophic respiration (SHR), one of the primary carbon fluxes from terrestrial ecosystems to the atmosphere, is important for carbon-climate feedbacks because of its sensitivity to available litter and soil carbon, climatic conditions, and nutrient availability. However, until recently limited SHR data were available, and most published global SHR estimates have either a short time span, coarse spatial resolution, or reply on overly-simple model formulations. To better understand and quantify the global distribution of SHR and its sensitivity to climate variability, we produced a new global SHR dataset using Random Forest algorithms, up-scaling 455 point data from the Global Soil Respiration Database (SRDB 4.0) with gridded fields of climatic, edaphic and productivity as explanatory variables. We estimated a global total SHR of 46.8 Pg C yr-1 over 1985-2013 (95% confidence interval: 38.6-56.3 Pg C yr-1), with a significant increasing trend of 0.03 Pg C yr-2 during this period. We found that the choice of soil moisture datasets contributes more to the difference among these data-driven SHR members rather than that of productivity, temperature and precipitation data sources. We also analyzed the influence of climatic variables on the inter-annual variability (IAV) of our SHR product. Water availability was the dominant driver of IAV at global scales, although the inferred sensitivity depends on the choice of the soil moisture gridded dataset. At the ecosystem scale, temperature strongly controls the IAV of SHR in tropical forests, while water availability dominates in extra-tropical forest and semi-arid regions. Our machine-learning gridded SHR dataset and outputs from process-based land surface models (TRENDYv6) show agreement for a strong association between water variability and SHR IAV at the global scale, but the two approaches lead to different temporal trend globally and different controlling variables for IAV at the ecosystem scale. Our study provides evidence for the pervasive and important role of water availability in driving SHR, indicating both a direct effect limiting decomposition rates and an indirect effect through the amount of fresh organic matter made available to SHR from productivity. In consideration of potential limitations and uncertainties remaining in our data-driven SHR datasets, we call for a more scientifically designed observation network for SHR, more observation data compilation, and increased use of deep learning methods making maximum use of observation data in hand. This will benefit process-based models, and improve our understanding of SHR response to future anomalous environmental conditions.

Yao, Yitong↗

SU-D-BRB-05: Quantum Learning for Knowledge-Based Response-Adaptive Radiotherapy

Purpose: There is tremendous excitement in radiotherapy about applying data-driven methods to develop personalized clinical decisions for real-time response-based adaptation. However, classical statistical learning methods lack in terms of efficiency and ability to predict outcomes under conditions of uncertainty and incomplete information. Therefore, we are investigating physics-inspired machine learning approaches by utilizing quantum principles for developing a robust framework to dynamically adapt treatments to individual patient’s characteristics and optimize outcomes. Methods: We studied 88 liver SBRT patients with 35 on non-adaptive and 53 on adaptive protocols. Adaptation was based on liver function using a split-course of 3+2 fractions with a month break. The radiotherapy environment was modeled as a Markov decision process (MDP) of baseline and one month into treatment states. The patient environment was modeled by a 5-variable state represented by patient’s clinical and dosimetric covariates. For comparison of classical and quantum learning methods, decision-making to adapt at one month was considered. The MDP objective was defined by the complication-free tumor control (P{sup +}=TCPx(1-NTCP)). A simple regression model represented state-action mapping. Single bit in classical MDP and a qubit of 2-superimposed states in quantum MDP represented the decision actions. Classical decision selection was done using reinforcement Q-learning and quantum searching was performed using Grover’s algorithm, which applies uniform superposition over possible states and yields quadratic speed-up. Results: Classical/quantum MDPs suggested adaptation (probability amplitude ≥0.5) 79% of the time for splitcourses and 100% for continuous-courses. However, the classical MDP had an average adaptation probability of 0.5±0.22 while the quantum algorithm reached 0.76±0.28. In cases where adaptation failed, classical MDP yielded 0.31±0.26 average amplitude while the quantum approach averaged a more optimistic 0.57±0.4, but with high phase fluctuations. Conclusion: Our results demonstrate that quantum machine learning approaches provide a feasible and promising framework for real-time and sequential clinical decision-making in adaptive radiotherapy.

60 APPLIED LIFE SCIENCES↗

Precursor Analysis Report: SQL Slammer Worm Infection of Davis-Besse Nuclear Power Plant 2003

The SQL Slammer Worm Infection of Davis-Besse Nuclear Power Plant 2003 Precursor Analysis Report leverages publicly available information about Davis-Besse’s 2003 cyber attack and catalogs anomalous observables for each technique employed in the attack. This analysis is based upon the methodology of the Cybersecurity for the Operational Technology Environment (CyOTE) program. On 25 January 2003, the SQL Slammer worm infected more than 90% of vulnerable hosts and crashed the internet in 10 to 15 minutes, making it one of the fastest spreading worms in history. SQL Slammer is a fileless, memory-resident worm that remotely exploits a stack-based buffer overflow vulnerability on local hosts to intensively scan and rapidly self-propagate across the internet. The worm infected approximately 300,000 unpatched hosts running Microsoft Structured Query Language (SQL) Server 2000 or Microsoft Desktop Engine (MSDE) 2000 with SQL Server Resolution Service. The SQL Slammer worm indirectly infected FirstEnergy’s Davis-Besse nuclear power plant by first infecting a consultant’s company network server and then propagating through an external misconfigured connection into Davis-Besse’s site network. The infection caused major network congestion, slow performance, data overloads, and the inability of local hosts to communicate with each other, which eventually caused a loss of availability and a loss of view when the Safety Parameter Display System (SPDS) and Plant Process Computer (PPC) crashed. At the time of the infection, the plant was already offline, the digital monitoring systems had redundant analog backups, and the plant control and safety functions were not affected, so there were no concerns of a safety breach. However, this incident resulted in many lessons learned and spawned important discussions about cybersecurity’s role in nuclear safety and electric power reliability regulation, policy, and guidance. Researchers and analysts identified 10 unique techniques utilized during the attack with a total of 640 observables using MITRE ATT&CK® for Industrial Control Systems. The CyOTE program assesses observables accompanying techniques used prior to the triggering event to identify opportunities to detect malicious activity. If observables accompanying the attack techniques are perceived and investigated prior to the triggering event, earlier comprehension of malicious activity can take place. Eight of the identified techniques used during Davis-Besse cyber attack were precursors to the triggering event. Analysis identified 596 observables associated with these precursor techniques, 428 of which were assessed to have an increased likelihood of being perceived in the 331 days preceding the triggering event. The response and comprehension time could have been reduced if the observables had been identified earlier. The information gathered in this report contributes to a library of observables tied to a repository of artifacts, data sources, and technique detection references for practitioners and developers to support the comprehension of indicators of attack. Asset owners and operators can use these products if they experience similar observables or to prepare for comparable scenarios.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Model-free stabilization via Extremum Seeking using a cost neural estimator

In this paper, a fully model-free architecture for vertical stabilization of thermonuclear plasmas in tokamak experimental reactors is presented. For the first time, an Extremum Seeking control algorithm is combined with neural networks to estimate the Lyapunov function to be minimized, resulting in a fully data-driven control architecture. The performance of different neural networks are compared. Specifically, Multilayer Perceptrons and Extreme Learning Machines are considered. The proposed architecture is tested in simulation to show that it can counteract relevant plasma disturbances, resulting in a significant improvement in terms of the achievable operative space compared to the Extremum Seeking algorithm, which still relies on model-based cost estimator.

42 ENGINEERING↗

Fault Detection for Dynamical Systems using Differential Geometric and Concurrent Learning Approach

This paper presents a fault identification and estimation approach that brings together the differential geometric concept of observability codistribution with data driven concurrent learning. In order to identify faults in presence of unknown disturbances, we utilize the differential geometric approach to design a coordinate transformation, to find a subspace in which the effect of disturbances and system faults can be segregated. We then use concurrent learning to estimate magnitude of the constant fault. We illustrate the approach to fault isolation for a linear helicopter dynamics and a spherical pendulum dynamics. We use Lyapunov stability analysis to show that the fault estimate by concurrent learning converges to the actual fault value, and then illustrate the design of a recovery controller.

Fault detection and diagnosis↗

High fidelity multiphysics tightly coupled model for a lead cooled fast reactor concept and application to statistical calculation of hot channel factors

A tightly coupled multiphysics code system is established using the MOOSE framework for hot channel factor (HCF) evaluation on a Lead Fast Reactor (LFR) concept. The coupled system is driven by the Griffin multiphysics coupling capability under which the MOOSE Heat Transfer module and NekRS computational fluid dynamics solver are coupled for conjugate heat transfer using the Cardinal application. The coupled capability is demonstrated on an LFR assembly model based on materials and geometry of a prototypical lead-cooled fast reactor design by Westinghouse Electric Company, LLC. Moreover, the work integrates the Multiphysics Object Oriented Simulation Environment (MOOSE) Stochastic Tools Module (STM) to perform calculations for statistical analysis of HCF. Furthermore, the coupling strategy and workflow demonstrated in this paper is not only useful for predicting accurate hot channel factors for different kinds of advanced reactors but also for other engineering applications such as control rod worth assessment, generation of high-fidelity database for Artificial intelligence (AI)/machine learning (ML) training, design optimization and multi-resolution modeling.

Cardinal↗

ESAC (EQ-SANS Assisting Chatbot): Application of large language models and retrieval-augmented generation for enhanced user experience at EQ-SANS

Neutron scattering experiments have played vital roles in exploring materials properties in the past decades. While user interfaces have been improved over time, neutron scattering experiments still require specific knowledge or training by an expert due to the complexity of such advanced instrumentation and the limited number of experiments each person may perform each year. This paper introduces an innovative chatbot application that leverages Large Language Models(LLM) and Retrieval-Augmented Generation (RAG) technologies to significantly enhance the user experience at the EQ-SANS, a small-angle neutron scattering instrument at the Spallation Neutron Source of Oak Ridge National Laboratory. Through a user-centric design approach, the EQ-SANS Assisting Chatbot (ESAC) serves as an interactive reference for users, thereby facilitating the use of the instrument by visiting scientists. By bridging the gap between the users of EQ-SANS and the control systems required to perform their experiments, the ESAC sets a new standard for interactive learning and support for the scientific community using large-scale scientific facilities.

97 MATHEMATICS AND COMPUTING↗

An AI-accelerated pathway for reproducible and stable halide perovskites

Halide perovskites (HPs) have remarkable optoelectronic properties, and in the last decade their photovoltaic power conversion efficiency and light-emitting diode efficiency have skyrocketed. Despite the surge in research on these burgeoning materials, two key challenges in the field remain: material irreproducibility and instability. Their behavior is especially dynamic in response to environmental stressors, due to complex interactions with the perovskite crystal lattice. Here, in this review, we survey the latest achievements in HP materials research accomplished with the assistance of artificial intelligence (AI), through the implementation of automated experimentation and machine learning (ML) data analysis. Automated synthesis and characterization tackle problems with material irreproducibility by systematically controlling parameters with very high precision, creating massive datasets, and allowing methodical comparisons from which unbiased conclusions can be drawn. AI can reveal otherwise unnoticed trends, inform future experiments with the highest potential information gain, and forecast future performance. The review concludes with a forward viewpoint of how human-assisted closed-loop laboratories and shared databases allow halide perovskite materials’ processing, properties, and performance to be potentially optimized with AI, accelerating the development of highly reproducible and stable optoelectronic devices.

Hering, Abigail R. [Univ. of California, Davis, CA↗

Reduced Erosion Augments Soil Carbon Storage Under Cover Crops

ABSTRACT Cover crops, a promising strategy to increase soil organic carbon (SOC) storage in croplands and mitigate climate change, have typically been shown to benefit soil carbon (C) storage from increased plant C inputs. However, input‐driven C benefits may be augmented by the reduction of C outputs induced by cover crops, a process that has been tested by individual studies but has not yet been synthesized. Here we quantified the impact of cover crops on organic C loss via soil erosion (SOC erosion) and revealed the geographical variability at the global scale. We analyzed the field data from 152 paired control and cover crop treatments from 57 published studies worldwide using meta‐analysis and machine learning. The meta‐analysis results showed that cover crops widely reduced SOC erosion by an average of 68% on an annual basis, while they increased SOC stock by 14% (0–15 cm). The absolute SOC erosion reduction ranged from 0 to 18.0 Mg C −1 ha −1 year −1 and showed no correlation with the SOC stock change that varied from −8.07 to 22.6 Mg C −1 ha −1 year −1 at 0–15 cm depth, indicating the latter more likely related to plant C inputs. The magnitude of SOC erosion reduction was dominantly determined by topographic slope. The global map generated by machine learning showed the relative effectiveness of SOC erosion reduction mainly occurred in temperate regions, including central Europe, central‐east China, and Southern South America. Our results highlight that cover crop‐induced erosion reduction can augment SOC stock to provide additive C benefits, especially in sloping and temperate croplands, for mitigating climate change.

Huang, Wenjuan [Department of Ecology, Evolution, ↗

Projective Integral Updates for High-Dimensional Variational Inference

Variational inference is an approximation framework for Bayesian inference that seeks to improve quantified uncertainty in predictions by optimizing a simplified distribution over parameters to stand in for the full posterior. Capturing model variations that remain consistent with training data enables more robust predictions by reducing parameter sensitivity. This work introduces a fixed-point optimization for variational inference that is applicable when every feasible log density can be expressed as a linear combination of functions from a given basis. In such cases, the optimizer becomes a fixed-point of projective integral updates. When the basis spans univariate quadratics in each parameter, the feasible distributions are Gaussian mean-fields and the projective integral updates yield quasi-Newton variational Bayes (QNVB). Other bases and updates are also possible. Since these updates require high-dimensional integration, this work begins by proposing an efficient quasirandom sequence of quadratures for mean-field distributions. Each iterate of the sequence contains two evaluation points that combine to correctly integrate all univariate quadratic functions and, if the mean-field factors are symmetric, all univariate cubics. More importantly, averaging results over short subsequences achieves periodic exactness on a much larger space of multivariate polynomials of quadratic total degree. The corresponding variational updates require four loss evaluations with standard (not second-order) backpropagation to eliminate error terms from over half of all multivariate quadratic basis functions. Furthermore, this integration technique is motivated by first proposing stochastic blocked mean-field quadratures, which may be useful in other contexts. A PyTorch implementation of QNVB allows for better control over model uncertainty during training than competing methods. Experiments demonstrate superior generalizability for multiple learning problems and architectures.

Gaussian mean-field↗

Multivariate Time Series Intermittent Fault Detectionin Controller Area Network CAN

Fault detection in Controller Area Network (CAN) systems is crucial for ensuring the reliability and safety of automotive and industrial applications. This study investigates and compares the effectiveness of time series classification models for supervised fault detection in CAN data. This repository contains the code and data for our benchmarking experiment aimed at detecting intermittent faults in automotive Controller Area Network (CAN) data. The goal of this project is to compare various machine learning (ML) and deep learning (DL) models using different Time Series Cross-Validation (TSCV) techniques to evaluate their effectiveness in a streaming environment for fault detection.

Hespeler, Steven [Oak Ridge National Laboratory (O↗

Enabling off-highway diesel engine downsizing and performance improvement using electrically assisted turbocharging

Internal combustion engine (ICE) downsizing through various turbocharging configurations is generally known by the powertrain design community as an effective means to reduce frictional losses, increase waste heat recovery, and improve fuel efficiency while increasing engine power density. However, often is the case that turbocharging strategies, including variable geometry turbochargers, and regulated two-stage turbochargers, incur the performance tradeoff between transient response and fuel economy (pumping losses) at high engine speeds. For off-highway vehicles having particularly transient and high-powered duty cycles, efforts to improve this tradeoff and increase operational flexibility have turned to evaluating various electrified air system architectures. In this study, a 4.5 L diesel-ICE configured with an electrically driven compressor (eBooster ® ) is placed in series with a conventional turbocharger and integrated into a 48 V mild-hybrid powertrain architecture. The objective of this powertrain configuration is to enable engine downsizing by 34%, replacing the current 6.8 L ICE platform with the hybridized 4.5 L ICE concept. The commercial 1-D simulation software GT-SUITE is used for powertrain system development and system optimization. Development and validation of the GT-SUITE model and air system controls is concurrently supported through experimental data collection. The simulation model development includes using machine learning methods for optimizing exhaust gas dilution, injection timing, and eBooster ® power to improve steady-state and transient brake-specific fuel consumption while minimizing criteria pollutant emissions. It was found that total specific fluid consumption over the standardized non-road transient engine duty cycle could be reduced by 18% over the current 6.8 L engine by using an optimized eBoosted 4.5 L engine. The hybridized 4.5 L engine concept concurrently showed sufficient transient response capability and nearly an order of magnitude reduction in duty cycle total soot production.

25 ENERGY STORAGE↗

Brush-Coated Nanoparticle Polymer Thin Films: structure-mechanical-optical properties

Our work was devoted to understanding the structure and properties of a class of thin film polymer nanocomposites (PNCs). PNCs are composed of polymer hosts into which nanoparticles (metallic nanoparticles, quantum dots, nanorods, C 60 , nanotubes) are incorporated. PNCs exhibit a diverse range of functional properties (optical, electronic, mechanical, biomedical, structural), determined in part by the chemical composition of the polymer host and the type of nanoparticle. The properties PNCs rely not only on specific functional, size-dependent, behavior of the nanoparticles, but also on the dispersion, and organizational order in some cases, inter-nanoparticle separation distances, and on relative interactions between the nanoparticles and the host. Therefore the scientific challenges associated with understanding the interrelations between the structure and function/properties of PNCs are far more complex than may be understood based only on the knowledge of the compositions of the constituents. The challenges of understanding the structure-function behavior of PNCs are further compounded by the fact that control of the dispersion of the nanoparticles within the polymer hosts is difficult; one must learn how to disperse inorganic particles within an organic host. The goal of this proposal was to develop an understanding of the connection between the structure and the thermal (glass transition), mechanical and optical properties of a specific class of PNCs. Specifically PNCs composed of polymer chain grafted gold nanoparticles within polymer hosts. A major objective was to understand how to develop basic principles that enable the fabrication of functional materials possessing optimized morphologies and combinations of materials properties.

36 MATERIALS SCIENCE↗

Updated status of the ART neutronic fast reactor tools integration to the NEAMS Workbench

The Workbench initiative was launched in FY-2017 within the Nuclear Energy Advanced Modeling and Simulation (NEAMS) Integration Product Line to facilitate the transition from conventional tools to high-fidelity tools. The Workbench provides a common user interface for model creation, real-time validation, execution, output processing, and visualization for integrated codes. The integration of the Argonne Reactor Computation (ARC) suite of codes into the NEAMS Workbench was initiated in FY-2017. The ARC codes contain both legacy codes like DIF3D and REBUS-3 that were developed with over 30 years of experience, and newer NEAMS additions like MC 2 -3 and PERSENT. The ARC integration into the NEAMS Workbench interface relies on the PyARC module which handles the pre- and post-processing of the native ARC codes input, and the runtime environment. The PyARC module together with the NEAMS Workbench interface are both released under Open Source Software licenses. This report describes the ARC capabilities available with the Workbench at the end of FY-2018, and a tutorial is provided together with the automatically-generated code documentation. Integrating the ARC codes into the Workbench benefits directly the Advanced Reactor Technology (ART) Campaign by providing a set of controlled, maintained and validated scripts to generate ARC inputs, which promotes best practices, reduces the learning curve, and facilitates project collaboration. The second benefit from this project is to improve the user experience with the ARC codes: the Workbench interface provides assistance for building an input through auto-completion, real-time validation, access to templates, and visualization. Finally, the ARC integration into the NEAMS Workbench enables new modeling capabilities, those will be extremely useful for advanced reactor design and analyses. The PyARC module facilitates and automatizes complex calculations and workflows for reactor analysis enabling geometrical perturbations, cross-section update through depletion, etc. The Dakota/PyARC coupling in the Workbench was also demonstrated to enable mathematical optimization and sensitivity analysis/uncertainty quantification (SA/UQ) techniques with ARC neutronic simulations. For demonstration purposes, the ARC codes were used through the Workbench for solving the Sodium-cooled Fast Reactor Uncertainty Analysis in Modelling (SFR-UAM) benchmark problems. The ARC codes are currently used at ANL through the Workbench by nuclear engineers for SFR core design analyses. Future efforts will focus on continuously adding modeling capabilities available with the ARC codes, training new users to continue building some experience, and on integrating high-fidelity codes such as PROTEUS into the Workbench.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Building an Adaptive Optics Demonstrator System

Adaptive optics is used to correct the wavefront of a beam of light. It originated in the field of astronomy to correct for atmospheric turbulence. This project was to construct an adaptive optics (AO) system kit from Thorlabs. An AO system consists of three parts: a deformable mirror (DM), a wavefront sensor (WFS), and control software. This system is to be used for future teaching purposes to help others learn about adaptive optics. The AO system came with instruction manuals, which were followed to build and calibrate the system. The instructions included building and aligning the system, how to install and use the software, and how to calibrate deformable mirror. A calibrated system for this project means having the deformable mirror imaged onto the wavefront sensor, with the center of the mirror on the center of the WFS. Also, a flat wavefront, which is achieved by eliminating the tip/tilt of the mirror. Eliminating the tip/tilt reduces the RMS amplitude to a value below 0.200 microns. This causes the wavefront to fold in on itself. We set up a focal plane detector behind beam splitter as a demonstration of what would be seen if this were used with a telescope. The image that should be seen on the focal plane detector is the beam with a diffraction pattern called and Airy Function. The AO system is now ready to be used to teach others about adaptive optics.

79 ASTRONOMY AND ASTROPHYSICS↗