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 487 records · Page 27

Time Dependent Data Mining in RAVEN

RAVEN is a generic software framework to perform parametric and probabilistic analysis based on the response of complex system codes. The goal of this type of analyses is to understand the response of such systems in particular with respect their probabilistic behavior, to understand their predictability and drivers or lack of thereof. Data mining capabilities are the cornerstones to perform such deep learning of system responses. For this reason static data mining capabilities were added last fiscal year (FY 15). In real applications, when dealing with complex multi-scale, multi-physics systems it seems natural that, during transients, the relevance of the different scales, and physics, would evolve over time. For these reasons the data mining capabilities have been extended allowing their application over time. In this writing it is reported a description of the new RAVEN capabilities implemented with several simple analytical tests to explain their application and highlight the proper implementation. The report concludes with the application of those newly implemented capabilities to the analysis of a simulation performed with the Bison code.

97 MATHEMATICS AND COMPUTING↗

Robust Learning of Dynamic Interactions for Enhancing Power System Resilience (Final Scientific Report)

The overall goal of the project is to leverage robust graphical learning and phase measurement unit (PMU) data to learn the dynamic interactions of electrical grid components in order to improve the power system resilience. Specifically, the learned dynamic interaction graphs are utilized to detect known and unknown anomalous patterns in power systems, identify high-risk operational conditions, and enable risk-based cascading mitigation.

24 POWER TRANSMISSION AND DISTRIBUTION↗

MO-AB-210-00: Diagnostic Ultrasound Imaging Quality Control and High Intensity Focused Ultrasound Therapy Hands-On Workshop

The goal of this ultrasound hands-on workshop is to demonstrate advancements in high intensity focused ultrasound (HIFU) and to demonstrate quality control (QC) testing in diagnostic ultrasound. HIFU is a therapeutic modality that uses ultrasound waves as carriers of energy. HIFU is used to focus a beam of ultrasound energy into a small volume at specific target locations within the body. The focused beam causes localized high temperatures and produces a well-defined regions of necrosis. This completely non-invasive technology has great potential for tumor ablation and targeted drug delivery. At the workshop, attendees will see configurations, applications, and hands-on demonstrations with on-site instructors at separate stations. The involvement of medical physicists in diagnostic ultrasound imaging service is increasing due to QC and accreditation requirements. At the workshop, an array of ultrasound testing phantoms and ultrasound scanners will be provided for attendees to learn diagnostic ultrasound QC in a hands-on environment with live demonstrations of the techniques. Target audience: Medical physicists and other medical professionals in diagnostic imaging and radiation oncology with interest in high-intensity focused ultrasound and in diagnostic ultrasound QC. Learning Objectives: Learn ultrasound physics and safety for HIFU applications through live demonstrations Get an overview of the state-of-the art in HIFU technologies and equipment Gain familiarity with common elements of a quality control program for diagnostic ultrasound imaging Identify QC tools available for testing diagnostic ultrasound systems and learn how to use these tools List of supporting vendors for HIFU and diagnostic ultrasound QC hands-on workshop: Philips Healthcare Alpinion Medical Systems Verasonics, Inc Zonare Medical Systems, Inc Computerized Imaging Reference Systems (CIRS), Inc. GAMMEX, Inc., Cablon Medical BV Steffen Sammet: NIH/NCI grant 5R25CA132822, NIH/NINDS grant 5R25NS080949 and Philips Healthcare research grant C32.

60 APPLIED LIFE SCIENCES↗

MO-AB-210-01: Ultrasound Imaging and Therapy-Hands On Workshop

The goal of this ultrasound hands-on workshop is to demonstrate advancements in high intensity focused ultrasound (HIFU) and to demonstrate quality control (QC) testing in diagnostic ultrasound. HIFU is a therapeutic modality that uses ultrasound waves as carriers of energy. HIFU is used to focus a beam of ultrasound energy into a small volume at specific target locations within the body. The focused beam causes localized high temperatures and produces a well-defined regions of necrosis. This completely non-invasive technology has great potential for tumor ablation and targeted drug delivery. At the workshop, attendees will see configurations, applications, and hands-on demonstrations with on-site instructors at separate stations. The involvement of medical physicists in diagnostic ultrasound imaging service is increasing due to QC and accreditation requirements. At the workshop, an array of ultrasound testing phantoms and ultrasound scanners will be provided for attendees to learn diagnostic ultrasound QC in a hands-on environment with live demonstrations of the techniques. Target audience: Medical physicists and other medical professionals in diagnostic imaging and radiation oncology with interest in high-intensity focused ultrasound and in diagnostic ultrasound QC. Learning Objectives: Learn ultrasound physics and safety for HIFU applications through live demonstrations Get an overview of the state-of-the art in HIFU technologies and equipment Gain familiarity with common elements of a quality control program for diagnostic ultrasound imaging Identify QC tools available for testing diagnostic ultrasound systems and learn how to use these tools List of supporting vendors for HIFU and diagnostic ultrasound QC hands-on workshop: Philips Healthcare Alpinion Medical Systems Verasonics, Inc Zonare Medical Systems, Inc Computerized Imaging Reference Systems (CIRS), Inc. GAMMEX, Inc., Cablon Medical BV Steffen Sammet: NIH/NCI grant 5R25CA132822, NIH/NINDS grant 5R25NS080949 and Philips Healthcare research grant C32.

60 APPLIED LIFE SCIENCES↗

MO-AB-210-02: Ultrasound Imaging and Therapy-Hands On Workshop

The goal of this ultrasound hands-on workshop is to demonstrate advancements in high intensity focused ultrasound (HIFU) and to demonstrate quality control (QC) testing in diagnostic ultrasound. HIFU is a therapeutic modality that uses ultrasound waves as carriers of energy. HIFU is used to focus a beam of ultrasound energy into a small volume at specific target locations within the body. The focused beam causes localized high temperatures and produces a well-defined regions of necrosis. This completely non-invasive technology has great potential for tumor ablation and targeted drug delivery. At the workshop, attendees will see configurations, applications, and hands-on demonstrations with on-site instructors at separate stations. The involvement of medical physicists in diagnostic ultrasound imaging service is increasing due to QC and accreditation requirements. At the workshop, an array of ultrasound testing phantoms and ultrasound scanners will be provided for attendees to learn diagnostic ultrasound QC in a hands-on environment with live demonstrations of the techniques. Target audience: Medical physicists and other medical professionals in diagnostic imaging and radiation oncology with interest in high-intensity focused ultrasound and in diagnostic ultrasound QC. Learning Objectives: Learn ultrasound physics and safety for HIFU applications through live demonstrations Get an overview of the state-of-the art in HIFU technologies and equipment Gain familiarity with common elements of a quality control program for diagnostic ultrasound imaging Identify QC tools available for testing diagnostic ultrasound systems and learn how to use these tools List of supporting vendors for HIFU and diagnostic ultrasound QC hands-on workshop: Philips Healthcare Alpinion Medical Systems Verasonics, Inc Zonare Medical Systems, Inc Computerized Imaging Reference Systems (CIRS), Inc. GAMMEX, Inc., Cablon Medical BV Steffen Sammet: NIH/NCI grant 5R25CA132822, NIH/NINDS grant 5R25NS080949 and Philips Healthcare research grant C32.

60 APPLIED LIFE SCIENCES↗

MO-AB-210-03: Workshop [ Advancements in high intensity focused ultrasound ]

The goal of this ultrasound hands-on workshop is to demonstrate advancements in high intensity focused ultrasound (HIFU) and to demonstrate quality control (QC) testing in diagnostic ultrasound. HIFU is a therapeutic modality that uses ultrasound waves as carriers of energy. HIFU is used to focus a beam of ultrasound energy into a small volume at specific target locations within the body. The focused beam causes localized high temperatures and produces a well-defined regions of necrosis. This completely non-invasive technology has great potential for tumor ablation and targeted drug delivery. At the workshop, attendees will see configurations, applications, and hands-on demonstrations with on-site instructors at separate stations. The involvement of medical physicists in diagnostic ultrasound imaging service is increasing due to QC and accreditation requirements. At the workshop, an array of ultrasound testing phantoms and ultrasound scanners will be provided for attendees to learn diagnostic ultrasound QC in a hands-on environment with live demonstrations of the techniques. Target audience: Medical physicists and other medical professionals in diagnostic imaging and radiation oncology with interest in high-intensity focused ultrasound and in diagnostic ultrasound QC. Learning Objectives: Learn ultrasound physics and safety for HIFU applications through live demonstrations Get an overview of the state-of-the art in HIFU technologies and equipment Gain familiarity with common elements of a quality control program for diagnostic ultrasound imaging Identify QC tools available for testing diagnostic ultrasound systems and learn how to use these tools List of supporting vendors for HIFU and diagnostic ultrasound QC hands-on workshop: Philips Healthcare Alpinion Medical Systems Verasonics, Inc Zonare Medical Systems, Inc Computerized Imaging Reference Systems (CIRS), Inc. GAMMEX, Inc., Cablon Medical BV Steffen Sammet: NIH/NCI grant 5R25CA132822, NIH/NINDS grant 5R25NS080949 and Philips Healthcare research grant C32.

60 APPLIED LIFE SCIENCES↗

The 'Room within a Room' Concept for Monitored Warhead Dismantlement

Over the past 10 years, US and UK experts have engaged in a technical collaboration with the aim of improving scientific and technological abilities in support of potential future nuclear arms control and non-proliferation agreements. In 2011 a monitored dismantlement exercise provided an opportunity to develop and test potential monitoring technologies and approaches. The exercise followed a simulated nuclear object through a dismantlement process and looked to explore, with a level of realism, issues surrounding device and material monitoring, chain of custody, authentication and certification of equipment, data management and managed access. This paper focuses on the development and deployment of the ‘room-within-a-room’ system, which was designed to maintain chain of custody during disassembly operations. A key challenge for any verification regime operating within a nuclear weapon complex is to provide the monitoring party with the opportunity to gather sufficient evidence, whilst protecting sensitive or proliferative information held by the host. The requirement to address both monitoring and host party concerns led to a dual function design which: • Created a controlled boundary around the disassembly process area which could provide evidence of unauthorised diversion activities. • Shielded sensitive disassembly operations from monitoring party observation. The deployed room-within-a-room was an integrated system which combined a number of chain of custody technologies (i.e. cameras, tamper indicating panels and enclosures, seals, unique identifiers and radiation portals) and supporting deployment procedures. This paper discusses the bounding aims and constraints identified by the monitoring and host parties with respect to the disassembly phase, the design of the room-within-a-room system, lessons learned during deployment, conclusions and potential areas of future work. Overall it was agreed that the room-within-a-room approach was effective but the individual technologies used to create the system deployed during this exercise required further development.

Nuclear Weapon, Dismantlement, Chain of Custody, C↗

Use of Machine Learning on PMU Data for Transmission System Fault Analysis

Synchrophasor technology has been used for monitoring, control, and protection of bulk power system for over 10 years. Deployment of phasor measurement units (PMUs) in the USA power system has surpassed 3000 units installed in the transmission substations as stand-alone intelligent electronic devices (IEDs) or as a software add-on to other devices such as digital protective relays (DPRs) or digital fault recorders (DFRs). By now, thousands of terabytes of PMU data may have been captured and stored by various transmission system operators (TSOs) and independent system operators (ISOs). This creates an opportunity to deploy advanced machine learning (ML) techniques to detect and classify faults recorded by PMUs automatically to be used by the system operators for rapid, critical decision-making when manual analysis of the past or unfolding events is not feasible. In this paper we offer a brief background on how the automated fault analysis may be done using DPR and/or DFR data, and compare some of the legacy approaches to the new ML approaches in the context of the system-wide PMU recordings. We then offer insights from developing practical ML solutions that have been applied on field recordings captured by close to 450 PMUs from all three US interconnections (Western, Eastern and ERCOT) over two years (2016-2017). We identify and illustrate ML challenges we addressed: inaccurate data, data with scarce and temporally imprecise fault labels, data recorded by PMUs sparsely located at substations resulting in the fault records taken afar from the ends of the faulted lines, data containing only positive sequence values, and data taken at different voltage levels. We then illustrate the ML model results for fault analysis under different application scenarios. The novelty of this study is not only in the design, implementation, and performance analysis of the ML algorithms, but also in the use of advanced fault modelling and simulation approaches to improve the training results when developing supervised ML models for fault detection and classification. Extensive simulations of faults were conducted on a 14-bus power system to create a training dataset with over 1400 accurately labelled faults. This dataset was applied to enhance the accuracy of fault detection and classification of machine learning-based models trained with small number of labelled faults in large datasets recorded in the grid interconnections ranging from 5,000 to 70,000 buses.

Synchrophasors, Machine Learning, Fault Analysis, ↗

An Augmented 3D iPad Mobile Application for Communication, Collaboration, and Learning (CCL) of Building MEP Systems

The goal of this research is to gain better understanding of how BIM models can be used to create mobile 3D game environments for some serious purpose other than entertainment. This project was motivated by the perspective that the capability of BIM creating realistic 3D virtual game environments for collaboration, communication and learning (CCL). The wide dissemination of mobile devices is driving many computer applications toward the direction of ubiquitous mobile accessibility. Considering the popularity of the iPad mobile device, a BIM model of an institutional building was created for this 3D mobile game application on iPad. This game application allows users to visualize and interact with the HVAC BIM components. In this paper, we reported our preliminary findings in developing a BIM-based iPad 3D game application for CCL activities in architecture, engineering, and construction (AEC). Technical barriers of using BIM models to create mobile 3D CCL applications were also discussed. The detailed system architecture and implementation framework were presented in this paper. Differences between the developed program and the existing BIM mobile programs were elaborated to distinguish the developed program from the off-the-shelf BIM mobile programs, such as Vella and Autodesk Inventor mobile viewer. The preliminary results indicated that fairly complex BIM models including multiple building systems can be imported and created through 3D virtual-built environments to conducted CCL activities in the AEC domain.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Learning Delayed Influences of Biological Systems

Boolean networks are widely used model to represent gene interactions and global dynamical behavior of gene regulatory networks. To understand the memory effect involved in some interactions between biological components, it is necessary to include delayed influences in the model. In this paper, we present a logical method to learn such models from sequences of gene expression data. This method analyzes each sequence one by one to iteratively construct a Boolean network that captures the dynamics of these observations. To illustrate the merits of this approach, we apply it to learning real data from bioinformatic literature. Using data from the yeast cell cycle, we give experimental results and show the scalability of the method. We show empirically that using this method we can handle millions of observations and successfully capture delayed influences of Boolean networks.

60 APPLIED LIFE SCIENCES↗

A Sequential Model Predictive and Deep Reinforcement Learning-Based Controller for Distribution System Outage Mitigation under Hurricane Events

This paper proposes a proactive outage mitigation framework for power distribution networks to withstand hurricane-induced disruptions. It leverages Model Predictive Control (MPC) to identify safe lines for proactive switching during hurricanes, minimizing the risk of cascading failures and voltage violations. The switching strategies optimized by MPC are sequentially integrated with a Deep Reinforcement Learning agent using the Advantage Actor-Critic algorithm, enabling dynamic line switching to maximize connected buses and minimize voltage violations in real time. Using a probabilistic hurricane model, the framework predicts line failures and adapts to varying conditions to enhance grid resilience. Simulations on the IEEE 123-bus system demonstrate its effectiveness in maintaining high connectivity and minimizing disruptions. Real-time testing with an RTDS confirms the practicality and reliability of the proposed approach.

Selim, Alaa [University of Connecticut]↗

Black and gray box learning of amplitude equations: Application to phase field systems

Here, we present a data-driven approach to learning surrogate models for amplitude equations and illustrate its application to interfacial dynamics of phase field systems. In particular, we demonstrate learning effective partial differential equations describing the evolution of phase field interfaces from full phase field data. We illustrate this on a model phase field system, where analytical approximate equations for the dynamics of the phase field interface (a higher-order eikonal equation and its approximation, the Kardar-Parisi-Zhang equation) are known. For this system, we discuss data-driven approaches for the identification of equations that accurately describe the front interface dynamics. When the analytical approximate models mentioned above become inaccurate, as we move beyond the region of validity of the underlying assumptions, the data-driven equations outperform them. In these regimes, going beyond black box identification, we explore different approaches to learning data-driven corrections to the analytically approximate models, leading to effective gray box partial differential equations.

42 ENGINEERING↗

Online Learning and Pricing for Service Systems with Reusable Resources

We consider a price-based revenue management problem with finite reusable resources over a finite time horizon T. Customers arrive following a price-dependent Poisson process, and each customer requests one unit of c homogeneous reusable resources. If there is an available unit, the customer gets served within a price-dependent exponentially distributed service time; otherwise, the customer waits in a queue until the next available unit. In this paper, we assume that the firm does not know how the arrival and service rates depend on posted prices, and thus it makes adaptive pricing decisions in each period based only on past observations to maximize the cumulative revenue. Given a discrete price set with cardinality P, we propose two online learning algorithms, termed batch upper confidence bound (BUCB) and batch Thompson sampling (BTS), and prove that the cumulative regret upper bound is O˜(√PT) , which matches the regret lower bound. In establishing the regret, we bound the transient system performance upon price changes via a novel coupling argument, and also generalize bandits to accommodate subexponential rewards. Here, we also extend our approach to models with balking and reneging customers and discuss a continuous price setting. Our numerical experiments demonstrate the efficacy of the proposed BUCB and BTS algorithms.

97 MATHEMATICS AND COMPUTING↗

Data-Driven Distribution System Coordinated PV Inverter Control Using Deep Reinforcement Learning

The deployment of distributed solar photovoltaic (PV) systems has increased consistently over the past decades. High penetrations of PVs could cause a series of adverse grid impacts, such as voltage violations. The recent development of smart inverter technologies rises the incentives of developing PV control solutions that regulate the inverter output power and seeking the optimization on system operational objectives. This paper proposes a data-driven control solution based on deep reinforcement learning (DRL) to optimize PV inverters for voltage regulation. The proposed solution can minimize PV real power curtailment while maintaining network voltage at an acceptable range. Comparison results between the proposed DRL control algorithms with deep deterministic policy gradient (DDPG) and volt-var control on a real feeder in west Colorado highlight the advantage of the proposed framework in controlling the system voltage while minimizing the PV real power curtailment.

deep reinforcement learning↗

The value of human data annotation for machine learning based anomaly detection in environmental systems

Anomaly detection is the process of identifying unexpected data samples in datasets. Automated anomaly detection is either performed using supervised machine learning models, which require a labelled dataset for their calibration, or unsupervised models, which do not require labels. While academic research has produced a vast array of tools and machine learning models for automated anomaly detection, the research community focused on environmental systems still lacks a comparative analysis that is simultaneously comprehensive, objective, and systematic. This knowledge gap is addressed for the first time in this study, where 15 different supervised and unsupervised anomaly detection models are evaluated on 5 different environmental datasets from engineered and natural aquatic systems. To this end, anomaly detection performance, labelling efforts, as well as the impact of model and algorithm tuning are taken into account. As a result, our analysis reveals the relative strengths and weaknesses of the different approaches in an objective manner without bias for any particular paradigm in machine learning. Most importantly, our results show that expert-based data annotation is extremely valuable for anomaly detection based on machine learning.

54 ENVIRONMENTAL SCIENCES↗

Dynamic Model Agnostic Reliability Evaluation of Machine-Learning Models Integrated in Instrumentation & Control Systems

In recent years, the field of data-driven neural network-based machine learning (ML) algorithms has grown significantly and spurred research in its applicability to instrumentation and control systems. While they are promising in operational contexts, the trustworthiness of such algorithms is not adequately assessed. Failures of ML-integrated systems are poorly understood; the lack of comprehensive risk modeling can degrade the trustworthiness of these systems. In recent reports by the National Institute for Standards and Technology, trustworthiness in ML is a critical barrier to adoption and will play a vital role in intelligent systems' safe and accountable operation. Thus, in this work, we demonstrate a real-time model-agnostic method to evaluate the relative reliability of ML predictions by incorporating out-of-distribution detection on the training dataset. It is well documented that ML algorithms excel at interpolation (or near-interpolation) tasks but significantly degrade at extrapolation. This occurs when new samples are "far" from training samples. The method, referred to as the Laplacian distributed decay for reliability (LADDR), determines the difference between the operational and training datasets, which is used to calculate a prediction's relative reliability. LADDR is demonstrated on a feedforward neural network-based model used to predict safety significant factors during different loss-of-flow transients. LADDR is intended as a "data supervisor" and determines the appropriateness of well-trained ML models in the context of operational conditions. Ultimately, LADDR illustrates how training data can be used as evidence to support the trustworthiness of ML predictions when utilized for conventional interpolation tasks.

97 MATHEMATICS AND COMPUTING↗

SU-E-T-397: Include Organ Deformation Into Dose Calculation of Prostate Brachytherapy

Purpose: Prostate brachytherapy is an important curative treatment for patients with localized prostate cancer. In brachytherapy, rectal balloon is generally needed to adjust for unfavorable prostate position for seed placement. However, rectal balloon causes prostate deformation, which is not accounted for in dosimetric planning. Therefore, it is possible that brachytherapy dosimetry deviates significantly from initial plan when prostate returns to its non-deformed state (after procedure). The goal of this study is to develop a method to include prostate deformation into the treatment planning of brachytherapy dosimetry. Methods: We prospectively collected ultrasound images of prostate pre- and post- rectal balloon inflation from thirty five consecutive patients undergoing I-125 brachytherapy. Based on the cylinder coordinate systems, we learned the initial coordinate transformation parameters between the manual segmentations of both deformed and non-deformed prostates of each patient in training set. With the nearest-neighbor interpolation, we searched the best transformation between two coordinate systems to maximum the mutual information of deformed and non-deformed images. We then mapped the implanted seeds of five selected patients from the deformed prostate into non-deformed prostate. The seed position is marked on original pre-inflation US image and it is imported into VariSeed software for dose calculation. Results: The accuracy of image registration is 87.5% as quantified by Dice Index. The prostate coverage V100% dropped from 96.5±0.5% of prostate deformed plan to 91.9±2.6% (p<0.05) of non-deformed plan. The rectum V100% decreased from 0.44±0.26 cc to 0.10±0.18 cc (p<0.05). The dosimetry of the urethra showed mild change but not significant: V150% changed from 0.05±0.10 cc to 0.14±0.15 cc (p>0.05) and D1% changed from 212.9±37.3 Gy to 248.4±42.8 Gy (p>0.05). Conclusion: We have developed a deformable image registration method that allows dosimetry calculation in the non-deformed prostate – which may be used to more accurately reflect “true” dosimetry during planning.

60 APPLIED LIFE SCIENCES↗

Method and system for analyzing gas leak based on machine learning

Embodiments of the present invention provide a system for estimating a location of a gas leak, based on machine learning from forward gas concentration data provided by an analog or scale model including a gas source. The system improves significantly over previous systems by providing high quality, physically accurate forward modeling data inexpensively. During operation, the system configures an aerosol source at a first location to emit a gaseous aerosol. The system then configures a laser source to illuminate the aerosol with a laser sheet. The system may then obtain an image of a reflection of the laser sheet from the aerosol. The system may then analyze the image to quantify a three-dimensional concentration distribution of the aerosol. The system may then estimate, based on solving an inverse problem and an observed second gas concentration, a second location of a second gas source.

03 NATURAL GAS↗