Search NASA⌕ Search

SEARCH · Search NASA

Results for “Learning systems”

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 811 records · Page 45

Parameter study of a vehicle-scale hydrogen storage system.

Sandia National Laboratories has developed a vehicle-scale prototype hydrogen storage system as part of a Work For Others project funded by General Motors. This Demonstration System was developed using the complex metal hydride sodium alanate. For the current work, we have continued our evaluation of the GM Demonstration System to provide learning to DOE's hydrogen storage programs, specifically the new Hydrogen Storage Engineering Center of Excellence. Baseline refueling data during testing for GM was taken over a narrow range of optimized parameter values. Further testing was conducted over a broader range. Parameters considered included hydrogen pressure and coolant flow rate. This data confirmed the choice of design pressure of the Demonstration System, but indicated that the system was over-designed for cooling. Baseline hydrogen delivery data was insufficient to map out delivery rate as a function of temperature and capacity for the full-scale system. A more rigorous matrix of tests was performed to better define delivery capabilities. These studies were compared with 1-D and 2-D coupled multi-physics modeling results. The relative merits of these models are discussed along with opportunities for improved efficiency or reduced mass and volume.

08 HYDROGEN↗

Evaluation of automated stability testing in machining through closed-loop control and Bayesian machine learning

Here, this paper describes a system for automated identification of the optimal stable cutting parameters in milling through Bayesian machine learning and closed-loop control. The closed-loop control system consists of a process monitoring architecture, an analysis framework, and a feedback mechanism. The analysis framework consists of a Bayesian machine learning algorithm that learns a stability map given test results. The learned stability map is used to select parameters for stability testing using an expected improvement in the material removal rate criterion. The test parameters are communicated to the machine controller to complete the test cut through a feedback mechanism. The test cuts were monitored using an audio signal; the stability of the test cut was determined by analyzing the frequency content of the audio signal. The test result was fed back to the Bayesian learning algorithm to complete the loop. Experimental results demonstrate that the system can identify the optimal stable parameters without information about the cutting force model or the structural dynamics. The system provides a low-cost method for optimal stable parameter identification in an industrial environment.

Chatter↗

A spark‐based big data analysis framework for real‐time sentiment prediction on streaming data

Summary There are many data sources that produce large volumes of data. The Big Data nature requires new distributed processing approaches to extract the valuable information. Real‐time sentiment analysis is one of the most demanding research areas that requires powerful Big Data analytics tools such as Spark. Prior literature survey work has shown that, though there are many conventional sentiment analysis researches, there are only few works realizing sentiment analysis in real time. One major point that affects the quality of real‐time sentiment analysis is the confidence of the generated data. In more clear terms, it is a valuable research question to determine whether the owner that generates sentiment is genuine or not. Since data generated by fake personalities may decrease accuracy of the outcome, a smart/intelligent service that can identify the source of data is one of the key points in the analysis. In this context, we include a fake account detection service to the proposed framework. Both sentiment analysis and fake account detection systems are trained and tested using Naïve Bayes model from Apache Spark's machine learning library. The developed system consists of four integrated software components, ie, (i) machine learning and streaming service for sentiment prediction, (ii) a Twitter streaming service to retrieve tweets, (iii) a Twitter fake account detection service to assess the owner of the retrieved tweet, and (iv) a real‐time reporting and dashboard component to visualize the results of sentiment analysis. The sentiment classification performances of the system for offline and real‐time modes are 86.77% and 80.93%, respectively.

Kılınç, Deniz↗

A comparative study on deep learning models for condition monitoring of advanced reactor piping systems

Advanced nuclear reactors offer innovative applications due to their portability, reliability, resiliency, and high capacity factors. To operate them on a wider scale, reducing maintenance life-cycle costs while ensuring their integrity is essential. Autonomous operations in advanced nuclear reactors using augmented Digital Twin (DT) technology can serve as a cost-effective solution by increasing awareness about the system’s health. A key component of nuclear DT frameworks is the condition monitoring of safety systems, such as piping-equipment systems, which involves acquiring and monitoring the plant’s sensor data. Here, this research proposes a condition monitoring methodology utilizing deep learning algorithms, such as multilayer perceptions (MLP) and convolutional neural networks (CNNs), to detect degradation and its severity in nuclear piping-equipment systems. Sensor signals are processed to obtain the power spectral density and the Short-Time Fourier transform, and feature extraction methodologies are proposed to develop degradation-sensitive data repositories. The performance of MLP, one-dimensional (1D) CNN, and 2D CNN within the proposed condition monitoring framework is compared using a finite element model of a 3D piping system subjected to seismic loads as the application case study. Various approaches, such as dropout, k-Fold validation, regularization, and early stopping of training the network, are investigated to avoid overfitting the models to the input sensor data. The predictive capability and computational capacity of the deep learning algorithms are also compared to detect degradation in the Z-pipe system of the Experimental Breeder Reactor II (EBRII). The Z-pipe system is subjected to harmonic excitations that represent normal operating loads, such as pump-induced vibrations. The findings of the study indicate that the proposed artificial intelligence (AI)-driven condition monitoring framework demonstrates superior prediction accuracies with a 2D CNN, whereas the MLP exhibits higher computational efficiency.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

TU-A-201-00: Image Guidance Technologies and Management Strategies

Recent years have seen a widespread proliferation of available in-room image guidance systems for radiation therapy target localization with many centers having multiple in-room options. In this session, available imaging systems for in-room IGRT will be reviewed highlighting the main differences in workflow efficiency, targeting accuracy and image quality as it relates to target visualization. Decision-making strategies for integrating these tools into clinical image guidance protocols that are tailored to specific disease sites like H&N, lung, pelvis, and spine SBRT will be discussed. Learning Objectives: Major system characteristics of a wide range of available in-room imaging systems for IGRT. Advantages / disadvantages of different systems for site-specific IGRT considerations. Concepts of targeting accuracy and time efficiency in designing clinical imaging protocols.

60 APPLIED LIFE SCIENCES↗

TU-A-201-01: Introduction to In-Room Imaging System Characteristics

Recent years have seen a widespread proliferation of available in-room image guidance systems for radiation therapy target localization with many centers having multiple in-room options. In this session, available imaging systems for in-room IGRT will be reviewed highlighting the main differences in workflow efficiency, targeting accuracy and image quality as it relates to target visualization. Decision-making strategies for integrating these tools into clinical image guidance protocols that are tailored to specific disease sites like H&N, lung, pelvis, and spine SBRT will be discussed. Learning Objectives: Major system characteristics of a wide range of available in-room imaging systems for IGRT. Advantages / disadvantages of different systems for site-specific IGRT considerations. Concepts of targeting accuracy and time efficiency in designing clinical imaging protocols.

60 APPLIED LIFE SCIENCES↗

TU-A-201-02: Treatment Site-Specific Considerations for Clinical IGRT

Recent years have seen a widespread proliferation of available in-room image guidance systems for radiation therapy target localization with many centers having multiple in-room options. In this session, available imaging systems for in-room IGRT will be reviewed highlighting the main differences in workflow efficiency, targeting accuracy and image quality as it relates to target visualization. Decision-making strategies for integrating these tools into clinical image guidance protocols that are tailored to specific disease sites like H&N, lung, pelvis, and spine SBRT will be discussed. Learning Objectives: Major system characteristics of a wide range of available in-room imaging systems for IGRT. Advantages / disadvantages of different systems for site-specific IGRT considerations. Concepts of targeting accuracy and time efficiency in designing clinical imaging protocols.

60 APPLIED LIFE SCIENCES↗

Additive manufacturing of complexly shaped SiC with high density via extrusion-based technique – Effects of slurry thixotropic behavior and 3D printing parameters

Here, additive manufacturing of dense SiC parts was achieved via an extrusion-based process followed by electrical-field assisted pressure-less sintering. The aim of this research was to study the effect of the rheological behavior of SiC slurry on the printing process and quality, as well as the influence of 3D printing parameters on the dimensions of the extruded filament, which are directly related to the printing precision and quality. Different solid contents and dispersant- Darvan 821A concentrations were studied to optimize the viscosity, thixotropy and sedimentation rate of the slurry. The optimal slurry was composed of 77.5 wt% SiC, Y 2 O 3 and Al 2 O 3 powders, 0.25 wt% dispersant and 0.01 wt% defoamer. The printing parameters studied included extrusion pressure, nozzle size, layer height and printing speed; the one that had the most prominent effect on filament width and height was indicated as layer height. The nozzle inner diameter of 1.04 mm, speed of 350 mm/min, layer height of 0.7 mm and extrusion air pressure of 0.31 MPa were the optimal printing parameters. Furthermore, the relationship between the printing parameters and the filament dimensions was successfully predicted by using machine learning and grey system theory. Finally, the relative density of the printed SiC parts sintered at 1900 oC reached 94.7±1.5%.

36 MATERIALS SCIENCE↗

Kolmogorov-Arnold wavefunctions

Here, this work investigates Kolmogorov-Arnold network-based (KAN) wave-function Ansätz as viable representations for quantum Monte Carlo simulations. Through systematic analysis of one-dimensional model systems, we evaluate their computational efficiency and representational power against established methods. Our numerical experiments suggest some efficient training methods and we explore how the computational cost scales with desired precision, particle number, and system parameters. Roughly speaking, KANs seem to be 10 times cheaper computationally than other neural-network-based Ansätz . We also introduce a novel approach for handling strong short-range potentials—a persistent challenge for many numerical techniques—which generalizes efficiently to higher-dimensional, physically relevant systems with short-ranged strong potentials common in atomic and nuclear physics.

1-dimensional systems↗

Ubiquitous Traffic Volume Estimation System

The Ubiquitous Traffic Volume Estimation System uses Machine Learning (ML) algorithms to provide accurate estimates of number vehicles on the roadway (ubiquitous traffic volume) based on inputs such as commercial probe traffic data that includes average vehicle speed and probe counts, roadway characteristics, weather, time of day, and day of week, and reference volume counts from a limited number of automatic traffic recorder stations. The Ubiquitous Traffic Volume Estimation System estimates traffic volumes 24 hours a day, 7 days a week, and 365 days a year on all major corridors and freeways in Denver metropolitan area and is being expanded to the whole state of Colorado. The methodology and code developed in this effort can be easily customized to other states and cities where similar data is available.

Young, Stanley↗

Sand Thermal Energy Storage Pilot Design (Final Report)

This report summarizes work done on developing a 10-MWhe pilot of the sand-based thermal energy storage (SandTES) technology at Alabama Power’s Plant Gaston Unit 5, an operating, supercritical coal power unit. The system will be integrated to the unit, obtaining steam to heat the sand through an air-blown fluidized-bed heat exchanger, then storing the water to be reused during discharging to produce steam that will then be vented. An electrical particle heater will also be included to provide part of the heat to the sand to provide data and learnings for commercial systems that will be fully electrically heated. Hot sand is contained in one bunker, while cold sand is housed in the other bunker, and standard solids handling equipment moves the sand around. This pilot would advance the SandTES technology to Technology Readiness Level 6 and position it for commercial readiness by 2030. This work was done in two phases: Phase I performed a conceptual study that provided Association for the Advancement of Cost Engineering (AACE) Class 5 costs and estimated performance, and then Phase II, which also involved a design update, performed a more detailed pre-front-end engineering and design study that elicited AACE Class 4 costs. Work was also done to provide estimated costs for commercial applications of the technology, assess its gaps, create its technology maturation and commercialization plans, and finally perform an Environmental Information Volume for the pilot site as a first step in the National Environmental Policy Act process.

20 FOSSIL-FUELED POWER PLANTS↗

Actinide Molten Salts: A Machine-Learning Potential Molecular Dynamics Study

We know that actinide molten salts represent a class of important materials in nuclear energy. Understanding them at a molecular level is critical to proper and optimal design of relevant technological applications. Yet, owing to the complexity of electronic structure due to the 5f orbitals, computational studies of heavy elements in condensed phases using ab initio potentials to study the structure and dynamics of these elements embedded in molten salts are difficult. This lack of efficient computational protocols makes it difficult to obtain information on properties that require extensive statistical sampling like transport. To tackle this problem, we adopted a machine-learning approach to study ThCl 4 -NaCl and UCl 3 -NaCl binary systems. The machine-learning potential, with the density functional theory accuracy, allows us to obtain long molecular dynamics trajectories (ns) for large systems (10 3 atoms) at a considerably low computing cost, thereby efficiently gaining information about their bonding structures, thermodynamics, and dynamics at a range of temperature. We observed a considerable change in the coordination environments of actinide elements and their characteristic coordination-sphere lifetime. Our study also suggests that actinides in molten salts may not follow well known entropy-scaling laws.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

The Second Skin: A Wearable Sensor Suite that Enables Real-Time Human Biomechanics Tracking Through Deep Learning

Objective: Real-time determination of human kinematics and kinetics could advance biomechanics research and enable valuable applications of biofeedback and generalizable exoskeleton control. Here, this work aims to investigate a taskindependent, user-independent method for obtaining precise realtime joint state estimation across lower-body joints during a wide variety of tasks. Methods: We developed a generalizable sensing approach using a suit comprised of inertial measurement units (IMUs) and pressure insoles. With the suit, we collected a dataset of 33 tasks commonly performed during construction and hazardous waste cleanup (N = 10). We then trained deep learning user-independent, task-agnostic models to estimate joint lowerbody kinematics and dynamics using only worn sensor data. We likewise computed joint kinematics and dynamics analytically from sensor data to serve as a comparison tool for model results. Results: Our models achieved overall angle estimation root-meansquared-errors (RMSE) of 6.56±.92°, 8.60±1.01°, 7.58±.89°, and 6.00±.73° compared to 13.9±.1.3°, 15.31±1.0°, 10.76±.70°, and 7.56±.48° via analytical methods at the lower back, hip, knee, and ankle, respectively. Likewise, our models achieved overall normalized moment estimation RMSEs of .207±.069 Nm/kg, .242±.044 Nm/kg, .202±.038 Nm/kg, and .193±.034 Nm/kg compared to .306±.036 Nm/kg, .407±.021 Nm/kg, 1.18 ±.022 Nm/kg, and 1.73±.071 Nm/kg via analytical methods at the lower back, hip, knee, and ankle, respectively. Conclusion: These results are comparable to other state-of-the-art wearable sensing systems, establishing deep learning as a viable sensing approach that generalizes to new users and tasks. Significance: This work shows promise for enabling accurate real-world biomechanical data collection and enhancement of biofeedback systems and wearable robot control.

Casey, Ryan T. F. [Georgia Institute of Technology↗

Machine Learning–Based Condition Monitoring of a Circulating Water System of a Canadian Nuclear Plant

With the need to maintain long-term reliable energy using nuclear power plants, there is an underlying demand to ensure that the maintenance of plant components and systems is also done in an efficient and cost-effective manner. One way to achieve this is by moving from time-based maintenance to condition-based maintenance. The research presented in this paper focuses on applying statistical and machine-learning-based methods to capture anomalies within data for fault detection to further develop into condition monitoring. This paper focuses on system data for a circulating water system (CWS) of a pressurized heavy-water reactor for detecting anomalies. The different methodologies used for detecting and capturing anomalies in the CWS data are matrix profile, density-based spatial clustering of applications with noise (DBSCAN), and support vector machines (SVMs). Matrix profile and DBSCAN are used to distinguish between normal data and anomalous data. This paper presents a hybrid method using DBSCAN and SVM when a portion of the data is used for DBSCAN to generate clusters. This portion of data is then used to train the SVM along with the clusters generated by DBSCAN as output. SVM is then tested on unseen data as a predictive tool, which can work in real time to categorize data points as either normal or anomalous. This paper presents results that show the high accuracies of DBSCAN and SVM in capturing anomalies within the data for a CWS for fault detection. Thus, the maintenance plan would be focused on component condition rather than a time-based schedule by switching to an automated system to identify and predict faults within a CWS.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

How good are learning-based control v.s. model-based control for load shifting? Investigations on a single zone building energy system

Both model predictive control (MPC) and deep reinforcement learning control (DRL) have been presented as a way to approximate the true optimality of a dynamic programming problem, and these two have shown significant operational cost saving potentials for building energy systems. Furthermore, there is still a lack of in-depth quantitative studies on their approximation levels to the true optimality, especially in the building energy domain. To fill in the gap, this paper provides a numerical framework that enables the evaluation of the optimality levels of different controllers for building energy systems. This framework is then used to comprehensively compare the optimal control performance of both MPC and DRL controllers with given computation budgets for a single zone fan coil unit system. Note the optimality is estimated based on a user-specific selection of trade-off weights among energy costs, thermal comfort and control slew rates. Compared with the best optimality we can find through expensive optimization simulations, the best DRL agent can maximally approximate the optimality by 96.54%, which outperforms the best MPC whose optimality level is 90.11%. However, due to the stochasticity, the DRL agent is only expected to approximate the optimality by 90.42%, which is almost equivalent to the best MPC. Except for Proximal Policy Optimization (PPO), all DRL agents can have a better approximation to the optimality than the best MPC, and are expected to have better approximation than the MPC with a prediction horizon of 32 steps (15 min per step). In terms of reducing energy cost and thermal discomfort, MPC can outperform the rule-based control (RBC) by 18.47%–25.44%. DRL can be expected to outperform RBC by 18.95%–25.65% ,and the best DRL control policy can outperform RBC by 20.29%–29.72%. Although the comparison of the optimality level is performed in a perfect setting, e.g., MPC assumes perfect models, and DRL assumes a perfect offline training process and online deployment process, this can shed insight on their capabilities of approximating to the original dynamic programming problem.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Deep Learning of Dark Energy Spectroscopic Instrument Mock Spectra to Find Damped Lyα Systems

We have updated and applied a convolutional neural network (CNN) machine-learning model to discover and characterize damped Ly α systems (DLAs) based on Dark Energy Spectroscopic Instrument (DESI) mock spectra. We have optimized the training process and constructed a CNN model that yields a DLA classification accuracy above 99% for spectra that have signal-to-noise ratios (S/N) above 5 per pixel. The classification accuracy is the rate of correct classifications. This accuracy remains above 97% for lower S/N ≈1 spectra. This CNN model provides estimations for redshift and H i column density with standard deviations of 0.002 and 0.17 dex for spectra with S/N above 3 pixel -1 . Also, this DLA finder is able to identify overlapping DLAs and sub-DLAs. Further, the impact of different DLA catalogs on the measurement of baryon acoustic oscillations (BAO) is investigated. The cosmological fitting parameter result for BAO has less than 0.61% difference compared to analysis of the mock results with perfect knowledge of DLAs. This difference is lower than the statistical error for the first year estimated from the mock spectra: above 1.7%. We also compared the performances of the CNN and Gaussian Process (GP) models. Our improved CNN model has moderately 14% higher purity and 7% higher completeness than an older version of the GP code, for S/N > 3. Both codes provide good DLA redshift estimates, but the GP produces a better column density estimate by 24% less standard deviation. A credible DLA catalog for the DESI main survey can be provided by combining these two algorithms.

79 ASTRONOMY AND ASTROPHYSICS↗

Engine Combustion System Optimization Using Computational Fluid Dynamics and Machine Learning: A Methodological Approach

Gasoline compression ignition (GCI) engines are considered an attractive alternative to traditional spark-ignition and diesel engines. Here, a Machine Learning-Grid Gradient Ascent (ML-GGA) approach was developed to optimize the performance of internal combustion engines. ML offers a pathway to transform complex physical processes that occur in a combustion engine into compact informational processes. The developed ML-GGA model was compared with a recently developed Machine Learning-Genetic Algorithm (ML-GA). Detailed investigations of optimization solver parameters and variable limit extension were performed in the present ML-GGA model to improve the accuracy and robustness of the optimization process. Detailed descriptions of the different procedures, optimization tools, and criteria that must be followed for a successful output are provided here. The developed ML-GGA approach was used to optimize the operating conditions (case 1) and the piston bowl design (case 2) of a heavy-duty diesel engine running on a gasoline fuel with a research octane number (RON) of 80. The ML-GGA approach yielded >2% improvements in the merit function, compared with the optimum obtained from a thorough computational fluid dynamics (CFD) guided system optimization. The predictions from the ML-GGA approach were validated with engine CFD simulations. This study demonstrates the potential of ML-GGA to significantly reduce the time needed for optimization problems, without loss in accuracy compared with traditional approaches.

33 ADVANCED PROPULSION SYSTEMS↗