Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Continual learning”

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 937 records · Page 52

Building a quality culture in the Office of Space Flight: Approach, lessons learned and implications for the future

The purpose of this paper is to describe the approach and lessons learned by the Office of Space Flight (OSF), National Aeronautics and Space Administration (NASA), in its introduction of quality. In particular, the experience of OSF Headquarters is discussed as an example of an organization within NASA that is considering both the business and human elements of the change and the opportunities the quality focus presents to improve continuously. It is hoped that the insights shared will be of use to those embarking upon similar cultural changes. The paper is presented in the following parts: the leadership challenge; background; context of the approach to quality; initial steps; current initiatives; lessons learned; and implications for the future.

Roberts, C. Shannon↗

Grid-Interactive Multi-Zone Building Control Using Reinforcement Learning with Global-Local Policy Search

In this paper, we develop a grid-interactive multi-zone building controller based on a deep reinforcement learning (RL) approach. The controller is designed to facilitate building operation during normal conditions and demand response events, while ensuring occupants comfort and energy efficiency. We leverage a continuous action space RL formulation, and devise a two-stage global-local RL training framework. In the first stage, a global fast policy search is performed using a gradient-free RL algorithm. In the second stage, a local fine-tuning is conducted using a policy gradient method. In contrast to the state-of-the-art model predictive control (MPC) approach, the proposed RL controller does not require complex computation during real-time operation and can adapt to nonlinear building models. We illustrate the controller performance numerically using a five-zone commercial building.

30 DIRECT ENERGY CONVERSION↗

Grid-Interactive Multi-Zone Building Control Using Reinforcement Learning with Global-Local Policy Search: Preprint

In this paper, we develop a grid-interactive multi-zone building controller based on a deep reinforcement learning (RL) approach. The controller is designed to facilitate building operation during normal conditions and demand response events, while ensuring occupants comfort and energy efficiency. We leverage a continuous action space RL formulation, and devise a two-stage global-local RL training framework. In the first stage, a global fast policy search is performed using a gradient-free RL algorithm. In the second stage, a local fine-tuning is conducted using a policy gradient method. In contrast to the state-of-the-art model predictive control (MPC) approach, the proposed RL controller does not require complex computation during real-time operation and can adapt to non-linear building models. We illustrate the controller performance numerically using a five-zone commercial building.

30 DIRECT ENERGY CONVERSION↗

Lessons Learned With Risk Management: A Systems Engineer's Perspective

Risk management is a communications device that, when executed as an essential task, enables systems engineering to effectively balance risk across the project. Developing and baselining risks is an essential continuous task to ensure top project concerns both from bottom up and top down are being mitigated. Risk management provides the opportunity to avoid the consequence of the risk when mitigation steps start early enough. Just discussing risk with all the project flight elements during development, even if no risks are open, provides an excellent communication opportunity between systems engineering and those elements, ensuring concerns and worries have a platform for discussion. A well-managed risk identification process will identify concerns that are serious but not being clearly communicated, and it will enable mitigation of those potential problems before they cause a failure. Effective risk management requires considerable time and effort, but that effort will save time and money across the development. Risk management must be frequent enough to be useful and in depth enough to bring out emerging issues. It also requires a trusting relationship between the lead systems engineer and element and/or subsystem leads. The discussions need to be with the right number of individuals (typically a handful) and the right duration in time (typically an hour a month). Outside of these risk working groups, there is a formal management process to input, status, and disposition risks, and a monthly Risk Management Board meeting where key project stakeholders are informed. This paper provides good guidance on effective risk management from a systems engineering perspective and provides project lessons learned from the NASA spaceflight missions NICER, Landsat 9, LRO, and OSIRIS-REx to demonstrate the effectiveness of risk management.

Lessons Learned↗

Lessons Learned With Risk Management: A Systems Engineer’s Perspective

Risk management is a communications device that, when executed as an essential task, enables systems engineering to effectively balance risk across the project. Developing and baselining risks is an essential continuous task to ensure top project concerns both from bottom up and top down are being mitigated. Risk management provides the opportunity to avoid the consequence of the risk when mitigation steps start early enough. Just discussing risk with all the project flight elements during development, even if no risks are open, provides an excellent communication opportunity between systems engineering and those elements, ensuring concerns and worries have a platform for discussion. A well-managed risk identification process will identify concerns that are serious but not being clearly communicated, and it will enable mitigation of those potential problems before they cause a failure. Effective risk management requires considerable time and effort, but that effort will save time and money across the development. Risk management must be frequent enough to be useful and in depth enough to bring out emerging issues. It also requires a trusting relationship between the lead systems engineer and element and/or subsystem leads. The discussions need to be with the right number of individuals (typically a handful) and the right duration in time (typically an hour a month). Outside of these risk working groups, there is a formal management process to input, status, and disposition risks, and a monthly Risk Management Board meeting where key project stakeholders are informed. This paper provides good guidance on effective risk management from a systems engineering perspective and provides project lessons learned from the NASA spaceflight missions NICER, Landsat 9, LRO, and OSIRIS-REx to demonstrate the effectiveness of risk management.

Lessons Learned↗

Idaho National Laboratory Integrated Safety Management System FY 2013 Effectiveness Review and Declaration Report

Idaho National Laboratory (INL) performed an Annual Effectiveness Review of the Integrated Safety Management System (ISMS), per 48 Code of Federal Regulations (CFR) 970.5223 1, “Integration of Environment, Safety and Health into Work Planning and Execution.” The annual review assessed Integrated Safety Management (ISM) effectiveness, provided feedback to maintain system integrity, and identified target areas for focused improvements and assessments for Fiscal Year (FY) 2014. Results of the FY 2013 annual effectiveness review demonstrate that the INL’s ISMS program is “Effective” and continually improving and shows signs of being significantly strengthened. Although there have been unacceptable serious events in the past, there has also been significant attention, dedication, and resources focused on improvement, lessons learned and future prevention. BEA’s strategy of focusing on these improvements includes extensive action and improvement plans that include PLN 4030, “INL Sustained Operational Improvement Plan, PLN 4058, “MFC Strategic Excellence Plan,” PLN 4141, “ATR Sustained Excellence Plan,” and PLN 4145, “Radiological Control Road to Excellence,” and the development of LWP 20000, “Conduct of Research.” As a result of these action plans, coupled with other assurance activities and metrics, significant improvement in operational performance, organizational competence, management oversight and a reduction in the number of operational events is being realized. In short, the realization of the fifth core function of ISMS (feedback and continuous improvement) and the associated benefits are apparent.

99 GENERAL AND MISCELLANEOUS↗

WE-B-BRC-02: Risk Analysis and Incident Learning

Prospective quality management techniques, long used by engineering and industry, have become a growing aspect of efforts to improve quality management and safety in healthcare. These techniques are of particular interest to medical physics as scope and complexity of clinical practice continue to grow, thus making the prescriptive methods we have used harder to apply and potentially less effective for our interconnected and highly complex healthcare enterprise, especially in imaging and radiation oncology. An essential part of most prospective methods is the need to assess the various risks associated with problems, failures, errors, and design flaws in our systems. We therefore begin with an overview of risk assessment methodologies used in healthcare and industry and discuss their strengths and weaknesses. The rationale for use of process mapping, failure modes and effects analysis (FMEA) and fault tree analysis (FTA) by TG-100 will be described, as well as suggestions for the way forward. This is followed by discussion of radiation oncology specific risk assessment strategies and issues, including the TG-100 effort to evaluate IMRT and other ways to think about risk in the context of radiotherapy. Incident learning systems, local as well as the ASTRO/AAPM ROILS system, can also be useful in the risk assessment process. Finally, risk in the context of medical imaging will be discussed. Radiation (and other) safety considerations, as well as lack of quality and certainty all contribute to the potential risks associated with suboptimal imaging. The goal of this session is to summarize a wide variety of risk analysis methods and issues to give the medical physicist access to tools which can better define risks (and their importance) which we work to mitigate with both prescriptive and prospective risk-based quality management methods. Learning Objectives: Description of risk assessment methodologies used in healthcare and industry Discussion of radiation oncology-specific risk assessment strategies and issues Evaluation of risk in the context of medical imaging and image quality E. Samei: Research grants from Siemens and GE.

60 APPLIED LIFE SCIENCES↗

Development of Gamma Background Radiation Digital Twin with Machine Learning Algorithms: Application of Unsupervised Machine Learning to Detection of Anomalies and Nuisances in Gamma Background Radiation Environmental Screening Data

Environmental screening of gamma radiation consists of detecting weak nuisance and anomaly signal in the presence of strong and highly varying background. In a typical scenario, a mobile detector-spectrometer continuously measures gamma radiation spectra in short, e.g., one-second, signal acquisition intervals. The measurement data is a 2D matrix, where one dimension is gamma ray energy, and the other dimension is the number of measurements or total time. In principle, gamma radiation sources can be detected and identified from the measured data by their unique spectral lines. Detecting sources from data measured in a search scenario is difficult due to the highly varying background because of naturally occurring radioactive material (NORM), and low signal-to-noise ratio (S/N) of spectral signal measured during one-second acquisition intervals. The objective of this work is to explore unsupervised machine learning (ML) algorithms for development of a digital twin of gamma radiation background, and for detection and identification of weak nuisances and anomalies events in the presence of highly fluctuating background. In one segment of work, we developed a gamma background estimation model using a Longshort term memory (LSTM) network for one-step CPS time series prediction. The LSTM model was validated with two data sets of measurements from two independent NaI detectors positioned on a mobile platform. The data sets contained background radiation only and no orphan isotope sources. The LSTM model was constructed and tested using data from one of the detectors. Performance of the LSTM model was validate through one-step prediction of CPS time series of another NaI detector without re-training. This approach allows to create a digital twin for nuclear background estimation. Using LSTM, it could be possible to detect a source through subtraction of the estimated counts from the measured background. In another segment of work, we investigated detection of gamma emitting sources in the presence of complex background using unsupervised machine learning. Spectral lines of isotopes are difficult to observe in one-second measurements. Averaging over the entire measurement campaign data set reveals spectral lines of most common background isotopes. Spectral lines of orphan sources, which might appear only in a few measurements during the campaign, will be washed out if averaging is performed over the entire measurement data set. The approach we have explored consists of extracting one-second measurements containing weak spectral features through data clustering. Averaging one-second spectra in a cluster should reveal the presence of anomaly sources. We created two ML models using K-means clustering and Neural Network Self-organizing Map (SOM). Performance of these ML models was benchmarked using search data. One data set contained 137 Cs source, and another dataset contained 131 I source.

54 ENVIRONMENTAL SCIENCES↗

Learning the Low Frequency Earthquake Activity on the Central San Andreas Fault

Abstract Low frequency earthquakes (LFEs) originating below the central San Andreas Fault are associated with slow‐slip beneath the seismogenic zone within the more ductile portion of the crust. Monitoring efforts over 15 years detected >1 million LFEs. We train a gradient boosted tree model using statistical features describing the seismic waveforms to estimate the hourly LFE event count. The burst‐like LFE behavior is reproduced, while lower amplitudes are predicted during the most active periods. The hourly event counts are up to 18% greater than the catalog. The ability to continuously monitor LFE activity provides insight to when geodetic measurements of slow slip are possible, without the need for developing a computational‐intensive template‐matching catalog. Similar waveform statistical features are found between detecting LFEs and tremors, which provides additional evidence tremors are composed of LFEs. The approach extracts information contained in continuous seismic waveforms that might benefit detecting precursory signals.

58 GEOSCIENCES↗

On fast simulation of dynamical system with neural vector enhanced numerical solver

The large-scale simulation of dynamical systems is critical in numerous scientific and engineering disciplines. However, traditional numerical solvers are limited by the choice of step sizes when estimating integration, resulting in a trade-off between accuracy and computational efficiency. To address this challenge, we introduce a deep learning-based corrector called Neural Vector (NeurVec), which can compensate for integration errors and enable larger time step sizes in simulations. Our extensive experiments on a variety of complex dynamical system benchmarks demonstrate that NeurVec exhibits remarkable generalization capability on a continuous phase space, even when trained using limited and discrete data. NeurVec significantly accelerates traditional solvers, achieving speeds tens to hundreds of times faster while maintaining high levels of accuracy and stability. Moreover, NeurVec’s simple-yet-effective design, combined with its ease of implementation, has the potential to establish a new paradigm for fast-solving differential equations based on deep learning.

97 MATHEMATICS AND COMPUTING↗

NASA/Mir Phase 1: A Lesson Learned in Long Duration Mission Planning and Operations

The NASA/Mir Phase I Program is a precursor to the assembly and operations of the International Space Station (ISS). The Phase I Program covers a four-year period and encompasses over ten shuttle flights to the Mir, culminating with the flight of STS-91 in June 1998. The Mir-Orbiting Station (Mir-OS) provides an excellent environment, both onboard and on the ground, for learning and studying the characteristics associated with crew-tended operations on long duration missions. By the end of the Phase I program, NASA astronauts will have accumulated over two years of continuous operations experience working and living onboard the Mir-OS. Ground controllers, working in Russia and at remote sites in the United States, will have gained valuable experience in the generation, maintenance, and coordination of the information and products required to support the long duration crew members onboard the Mir-OS. The experience gained by the crew and ground support personnel in the area of long duration mission operations has resulted in a number of factors that need to be considered in the definition and implementation of the processes and products for the ISS. While there are many areas in which experience has been gained, this paper focuses on the lessons learned regarding the level of detail and flexibility of crew timelines. Provided in this paper is an overview of the characteristics associated with long duration mission timeline development and execution, a summary of applicable lessons learned identified and documented through participation in the NASA/Mir Phase I Program, a detailed discussion of "Gross Timelines" as implementation which addresses the lessons learned. Also discussed are the challenges that must be overcome in getting the ISS program to recognize, evaluate, and learn from the crew and ground controller experience gained through participation in the NASA/Mir Phase I Program.

Hagopian, Jeff↗

Learning instrument invariant characteristics for generating high-resolution global coral reef maps

Coral reefs are one of the most biologically complex and diverse ecosystems within the shallow marine environment. Unfortunately, these underwater ecosystems are threatened by a number of anthropogenic challenges, including ocean acidification and warming, overfishing, and the continued increase of marine debris in oceans. This requires a comprehensive assessment of the world's coastal environments, including a quantitative analysis on the health and extent of coral reefs and other associated marine species, as a vital Earth Science measurement. However, limitations in observational and technological capabilities inhibit global sustained imaging of the marine environment. Harmonizing multimodal data sets acquired using different remote sensing instruments presents additional challenges, thereby limiting the availability of good quality labeled data for analysis. In this work, we develop a deep learning model for extracting domain invariant features from multimodal remote sensing imagery and creating high-resolution global maps of coral reefs by combining various sources of imagery and limited hand-labeled data available for certain regions. This framework allows us to generate, for the first time, coral reef segmentation maps at 2-meter resolution, which is a significant improvement over the kilometer-scale state-of-the-art maps. Additionally, this framework doubles accuracy and IoU metrics over baselines that do not account for domain invariance.

Domain Adaptation↗

Inference offers a metric to constrain dynamical models of neutrino flavor transformation

The multimessenger astrophysics of compact objects presents a vast range of environments where neutrino flavor transformation may occur and may be important for nucleosynthesis, dynamics, and a detected neutrino signal. Development of efficient techniques for surveying flavor evolution solution spaces in these diverse environments, which augment and complement existing sophisticated computational tools, could leverage progress in this field. To this end we continue our exploration of statistical data assimilation (SDA) to identify solutions to a small-scale model of neutrino flavor transformation. SDA is a machine learning formula wherein a dynamical model is assumed to generate any measured quantities. Specifically, we use an optimization formulation of SDA wherein a cost function is extremized via the variational method. Regions of state space in which the extremization identifies the global minimum of the cost function will correspond to parameter regimes in which a model solution can exist. Our example study seeks to infer the flavor transformation histories of two monoenergetic neutrino beams coherently interacting with each other and with a matter background. We require that the solution be consistent with measured neutrino flavor fluxes at the point of detection, and with constraints placed upon the flavor content at various locations along their trajectories, such as the point of emission, and the locations of the Mikheyev-Smirnov-Wolfenstein resonances. We show how the procedure efficiently identifies solution regimes and rules out regimes where solutions are infeasible. Overall, results in this work intimate the promise of this “variational annealing” methodology to efficiently probe an array of fundamental questions that traditional numerical simulation codes render difficult to access.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

HIDES: Hybrid Intrusion Detector for Energy Systems

The establishment of a resilient electric grid accompanied by a secure communications network is an ongoing battle as advanced persistent threats continue to exploit existing vulnerabilities in legacy supervisory control and data acquisition system (SCADA) infrastructure. Traditional intrusion detection systems (IDSs) lack consistent performance because of the continuously evolving attack surface of SCADA systems. These shortcomings can be overcome by integrating logical system behavior, protocol-specific knowledge, and data-based learning to develop a comprehensive IDS solution. In this paper, we present a Hybrid Intrusion Detector for Energy Systems by integrating a network-based IDS, state-of-the-art machine learning-based IDS, and model-based IDS to detect unknown and stealthy cyberattacks targeting the SCADA networks. The proposed IDS uses synchrophasor measurements and cyber logs to learn patterns of different scenarios based on spatiotemporal behaviors of power systems. As a proof of concept, we implement and validate the proposed IDS by leveraging resources available at the National Renewable Energy Laboratory's Energy Systems Integration Facility test bed. Experimental results show promising performance in detecting cyberattacks while providing realtime visualization of power system measurements and cyber logs.

machine-learning intrusion detection system↗

Optimizing Patient-Specific Medication Regimen Policies Using Wearable Sensors in Parkinson’s Disease

Effective treatment of Parkinson’s disease (PD) is a continual challenge for healthcare providers, and providers can benefit from leveraging emerging technologies to supplement traditional clinic care. We develop a data-driven reinforcement learning (RL) framework to optimize PD medication regimens through wearable sensors. We leverage a data set of n = 26 PD patients who wore wrist-mounted movement trackers for two separate six-day periods. Using these data, we first build and validate a simulation model of how individual patients’ movement symptoms respond to medication administration. We then pair this simulation model with an on-policy RL algorithm that recommends optimal medication types, timing, and dosages during the day while incorporating human-in-the-loop considerations on medication administration. The results show that the RL-prescribed medication regimens outperform physicians’ medication regimens, despite physicians having access to the same data as the RL agent. To validate our results, we assess our wearable-based RL medication regimens using n = 399 PD patients from the Parkinson’s Progression Markers Initiative data set. We show that the wearable-based RL medication regimens would lead to significant symptom improvement for these patients, even more so than training RL policies directly from this data set. In doing so, we show that RL models from even small data sets of wearable data can offer novel, generalizable clinical insights and medication strategies, which may outperform those derived from larger data sets without wearable data.

60 APPLIED LIFE SCIENCES↗

ESTEEM: A Novel Framework for Qualitatively Evaluating and Visualizing Spatiotemporal Embeddings in Social Media

Analyzing and visualizing large amounts of social media communications and contrasting short-term conversation changes over time and geo-locations is extremely important for commercial and government applications. Earlier approaches for large-scale text stream summarization used dynamic topic models and trending words. Instead, we rely on text embeddings – low-dimensional word representations in a continuous vector space where similar words are embedded nearby each other. This paper presents ESTEEM,1 a novel tool for visualizing and evaluating spatiotemporal embeddings learned from streaming social media texts. Our tool allows users to monitor and analyze query words and their closest neighbors with an interactive interface. We used state-of- the-art techniques to learn embeddings and developed a visualization to represent dynamically changing relations between words in social media over time and other dimensions. This is the first interactive visualization of streaming text representations learned from social media texts that also allows users to contrast differences across multiple dimensions of the data.

social media dynamics↗

Multi-Gas Sensors for Enhanced Reliability of SOFC Operation

GE Research, in partnership with SUNY Polytechnic Institute (SUNY Poly), designed, built, and tested gas sensors for in situ monitoring of H 2 and CO anode tail gases produced with on-site steam reforming in solid oxide fuel cell (SOFC) systems. The knowledge of the H 2 /CO ratio of these anode tail gases should allow accurate determination and control of the efficiency of the reforming process in the SOFC system and should deliver a lower operating cost for SOFC customers. The project objectives were to achieve multi-gas monitoring capability with a single multivariable sensor, and to sustain this performance in the presence of gaseous interferences and a potential poison for the sensor. The duration of the project was 24 months with the project structure that included three technical tasks such as (1) development of design rules of photonic nanostructures for H 2 and CO gas detection, (2) laboratory validation of photonic nanostructures for selective H 2 and CO gas detection, and (3) validation of photonic nanostructures for initial stability and poison-resistance against H 2 S. To build multivariable sensors for selective H 2 and CO gas detection in the presence of interferences, we expanded our earlier knowledge of multi-gas sensors into new fabrication and functionalization methodologies as well as into new methodologies for the spectral data analysis of multi-gas responses. We have advanced our design rules of the three-dimensional (3D) photonic nanostructures that allowed detection of H 2 and CO at high temperatures as individual gases and as their mixtures and rejection of interferences such as CO 2 , H 2 O, CH 4 , and other hydrocarbons for SOFC applications. Our advanced design rules should be attractive for building the new generation of cost-effective industrial sensors. Stability and poison resistance of our 3D photonic nanostructures was tested in the laboratory conditions. While initially we utilized conventional machine learning data analysis tools, we have found that they were unable to correct for the sensor drift. Thus, we have implemented new methods of machine learning for the analysis of our spectral data. These learnings pave the way to move the future studies into advanced testing of effects of interferences, aging and field tests. In future, our work will continue to advance our sensing designs to operate in conditions with known and unknown interferences by implementing nanostructures with enhanced spectral diversity of responses to gaseous species of interest and interferences. Our systematic reduction of technical risks in this completed project and in future studies will ensure transition of this sensing technology to commercialization.

03 NATURAL GAS↗

A microbial survey of the International Space Station (ISS)

Background: Modern advances in sequencing technology have enabled the census of microbial members of many natural ecosystems. Recently, attention is increasingly being paid to the microbial residents of human-made, built ecosystems, both private (homes) and public (subways, office buildings, and hospitals). Here, we report results of the characterization of the microbial ecology of a singular built environment, the International Space Station (ISS). This ISS sampling involved the collection and microbial analysis (via 16S rDNA PCR) of 15 surfaces sampled by swabs onboard the ISS. This sampling was a component of Project MERCCURI (Microbial Ecology Research Combining Citizen and University Researchers on ISS). Learning more about the microbial inhabitants of the “buildings” in which we travel through space will take on increasing importance, as plans for human exploration continue, with the possibility of colonization of other planets and moons. Results: Sterile swabs were used to sample 15 surfaces onboard the ISS. The sites sampled were designed to be analogous to samples collected for (1) the Wildlife of Our Homes project and (2) a study of cell phones and shoes that were concurrently being collected for another component of Project MERCCURI. Sequencing of the 16S rDNA genes amplified from DNA extracted from each swab was used to produce a census of the microbes present on each surface sampled. We compared the microbes found on the ISS swabs to those from both homes on Earth and data from the Human Microbiome Project. Conclusions: While significantly different from homes on Earth and the Human Microbiome Project samples analyzed here, the microbial community composition on the ISS was more similar to home surfaces than to the human microbiome samples. The ISS surfaces are species-rich with 1,036–4,294 operational taxonomic units (OTUs per sample). There was no discernible biogeography of microbes on the 15 ISS surfaces, although this may be a reflection of the small sample size we were able to obtain.

International space station↗