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 523 records · Page 29

Cataloguing Past Incidents in Transporting Spent Nuclear Fuel

Shipments of spent nuclear fuel (SNF) and high level radioactive waste (HLW) have a history of more than 50 years, during which tens of thousands of tons of material have been transported in tens of thousands of shipments worldwide. The vast majority of these have been completed without incident. This safety record has been made possible by establishing robust regulatory requirements for the casks used in transport, studying past shipments and incidents, and improving equipment and procedures to lessen the probability and impacts of future events. The authors recently collected a wide array of publicly available information on transportation incidents involving SNF and HLW worldwide. This presented a challenge, as no international or US databases currently exist which are dedicated to SNF and HLW transportation incidents. Thus, sources for this research included national database reports from nations that publish them; technical reports and archival journal papers focused on the solutions to particular transportation problems; and news reports of SNF and HLW transportation incidents, including transportation accidents and protests of shipments. Based on the survey of past transportation incidents, of which several representative descriptions are included in this paper, the authors conclude that (1) the casks in which SNF and HLW are shipped are able to contain their radioactive cargoes in severe accidents because of their robust construction and their being in compliance with regulatory requirements; and (2) most incidents in transporting SNF and HLW have been minor ones, primarily resulting from procedural compliance deficiencies. Furthermore, recent incidents suggest that public information and engagement challenges outweigh technical challenges in SNF and HLW transportation, since the most significant challenges surrounding shipments in recent years have been the results of public demonstrations and protests. This paper discusses some SNF and HLW transportation incidents, as well as the process used in cataloguing them. In general, there have been few transportation incidents worldwide in the history of transporting SNF and HLW. The transportation of SNF and HLW has an excellent safety record; by learning from past incidents, shippers of SNF and HLW can continue to make safe shipments while avoiding mistakes made in the past. Despite the safe and successful record of SNF and HLW transportation, information on its record is not easy to find. Few incidents are detailed as thoroughly as the truck accident that occurred in 1971. Several anti-nuclear organizations publish pamphlets that make reference to transportation accidents involving radioactive material and imply that transporting SNF and HLW is an intrinsically dangerous activity whose history is fraught with accidents. The absence of a clear, complete, official record on transportation incidents allows these claims to continue to be unanswered and uncorrected. For these reasons, it would be beneficial to develop a database detailing incidents involving SNF and HLW internationally. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

The Pixel Anomaly Detection Tool : a user-friendly GUI for classifying detector frames using machine-learning approaches

Data collection at X-ray free electron lasers has particular experimental challenges, such as continuous sample delivery or the use of novel ultrafast high-dynamic-range gain-switching X-ray detectors. This can result in a multitude of data artefacts, which can be detrimental to accurately determining structure-factor amplitudes for serial crystallography or single-particle imaging experiments. Here, a new data-classification tool is reported that offers a variety of machine-learning algorithms to sort data trained either on manual data sorting by the user or by profile fitting the intensity distribution on the detector based on the experiment. This is integrated into an easy-to-use graphical user interface, specifically designed to support the detectors, file formats and software available at most X-ray free electron laser facilities. The highly modular design makes the tool easily expandable to comply with other X-ray sources and detectors, and the supervised learning approach enables even the novice user to sort data containing unwanted artefacts or perform routine data-analysis tasks such as hit finding during an experiment, without needing to write code.

47 OTHER INSTRUMENTATION↗

Network-Scale Ubiquitous Volume Estimation Using Tree-Based Ensemble Learning Methods

Currently ubiquitous volume data for roadway networks remains the key missing dimension in traffic operations. Most volume data are average annual daily traffic (AADT) measures derived from the Highway Performance Monitoring System (HPMS). Although methods to factor the AADT to hourly averages for typical day of week exist, actual volume data is limited to a sparse collection of locations in which volumes are continuously recorded. This paper/poster explores the use of state-of-art machine learning techniques to estimate accurate volume measures that span the highway network providing ubiquitous coverage in space, and point-in-time measures for a specific date and time. Three tree-based ensemble learning models, random forest (RF), gradient boost machine (GBM), and extreme gradient boost (XGBoost), were tested for volume estimation by learning from combined dataset of commercial probe data provided by TomTom, the FHWA's Travel Monitoring Analysis System (TMAS) data, and other infrastructure attributes such as number of lanes, speed limit, and weather. The methods were tested on major corridors and freeways in the metropolitan area of Denver. All three machine learning methods were able to provide hourly volume estimates 24 hours a day, 7 days a week, and 365 days a year with around 18% mean absolute error to true volume and about 5% of error with respect to roadway capacity. The low error measures allow the potential application by transportation agencies.

33 ADVANCED PROPULSION SYSTEMS↗

Optimizing Perovskite Thin‐Film Parameter Spaces with Machine Learning‐Guided Robotic Platform for High‐Performance Perovskite Solar Cells

Abstract Simultaneously optimizing the processing parameters of functional thin films remains a challenge. The design and utilization of a fully automated platform called SPINBOT is presented for the engineering of solution‐processed functional thin films. The SPINBOT is capable of performing experiments with high sampling variability through the unsupervised processing of hundreds of substrates with exceptional experimental control. Through the iterative optimization process enabled by the Bayesian optimization (BO) algorithm, the SPINBOT explores an intricate parameter space, continuously improving the quality and reproducibility of the produced thin films. This machine learning (ML)‐guided reliable SPINBOT platform enables the acceleration of the optimization process of perovskite solar cells via a simple photoluminescence characterization of films. As a result, this study arrives at an optimal film that, when processed into a solar cell in an ambient atmosphere, immediately yields a champion power conversion efficiency (PCE) of 21.6% with satisfactory performance reproducibility. The unsealed devices retain 90% of their initial efficiency after 1100 h of continuous operation at 60–65 °C under metal‐halide lamps. It is anticipated that the integration of robotic platforms with the intelligent algorithm will facilitate the widespread adoption of effective autonomous experimentation to address the evolving needs and constraints within the materials science research community.

14 SOLAR ENERGY↗

Designing Specification Languages for Process Control Systems: Lessons Learned and Steps to the Future

Previously, we defined a blackbox formal system modeling language called RSML (Requirements State Machine Language). The language was developed over several years while specifying the system requirements for a collision avoidance system for commercial passenger aircraft. During the language development, we received continual feedback and evaluation by FAA employees and industry representatives, which helped us to produce a specification language that is easily learned and used by application experts. Since the completion of the PSML project, we have continued our research on specification languages. This research is part of a larger effort to investigate the more general problem of providing tools to assist in developing embedded systems. Our latest experimental toolset is called SpecTRM (Specification Tools and Requirements Methodology), and the formal specification language is SpecTRM-RL (SpecTRM Requirements Language). This paper describes what we have learned from our use of RSML and how those lessons were applied to the design of SpecTRM-RL. We discuss our goals for SpecTRM-RL and the design features that support each of these goals.

Leveson, Nancy G.↗

Network Scale Ubiquitous Volume Estimation Using Tree-Based Ensemble Learning Methods

Currently real-time, ubiquitous volume data for roadway networks remains the key missing dimension in traffic operations. Most volume data are average annual daily traffic (AADT) measures derived from the Highway Performance Monitoring System (HPMS). Although methods to factor the AADT to hourly averages for typical day of week exist, actual volume data is limited to a sparse collection of locations in which volumes are continuously recorded. This paper explores the use of state-of-art machine learning techniques to estimate accurate real-time volume measures that span the highway network providing ubiquitous coverage in space, and point-in-time measures for a specific date and time. Three tree-based ensemble learning models, random forest (RF), gradient boost machine (GBM), and extreme gradient boost (XGBoost), were tested for volume estimation by learning from combined dataset of commercial probe data provided by TomTom, the FHWA's Travel Monitoring Analysis System (TMAS) data, and other infrastructure attributes such as number of lanes, speed limit, and weather. The methods were tested on major corridors and freeways in the metropolitan area of Denver. All three machine learning methods were able to provide hourly volume estimates 24 hours a day, 7 days a week, and 365 days a year with around 18% mean absolute error to true volume and about 5% of error with respect to roadway capacity. The low error measures allow the potential application by transportation agencies.

33 ADVANCED PROPULSION SYSTEMS↗

TU-A-BRE-01: The Relative Biological Effectiveness of Proton Beams Relative to Photon Beams

Proton therapy patients receive a 10% lower physical dose than the dose administered using photons, i.e. the proton relative biological effectiveness (RBE) is 1.1 in comparison to high-energy photons. The use of a generic, spatially invariant RBE within tumor targets and normal tissue structures disregards a large body of evidence indicating that proton RBE tends to increase with increasing linear energy transfer (LET). Because the doseaveraged proton LET in the distal edge of a spread out Bragg peak (SOBP) is larger than the LET in the plateau region or proximal edge of a SOBP, the use of a spatially invariant RBE is not well justified from a mechanistic point of view. On the other hand, the available clinical data on local tumor control rates and early or late side effects do not provide strong evidence against the continued use of a constant and spatially invariant clinical RBE. The only potential downside to the ongoing use of a constant RBE of 1.1 seems to be that we are missing a potential opportunity to enhance the therapeutic ratio, i.e., design proton therapy treatments in ways that exploit, rather than mitigate, spatial variations in proton RBE. Speakers in this symposium will: 1-review the laboratory and clinical evidence for and against the continued use of a spatially invariant RBE of 1.1, 2-examine some of the putative mechanisms connecting spatial variations in particle LET to estimates of the proton RBE at the molecular, cellular and tissue levels 3-assess the possible clinical significance of incorporating models for spatial variations in proton RBE into treatment planning systems. 4-discuss treatment planning and delivery techniques that will exploit the spatial variations of RBE within proton beams. Learning Objectives: To review laboratory and clinical evidence for and against the continued use of a constant RBE of 1.1 To understand major mechanisms connecting proton LET to RBE at the molecular, cellular and tissue levels. To quantify the potential opportunities and potential pitfalls of neglecting spatial variations in proton RBE in treatment plans To understand the minimum requirements required to design proton treatment plans that benefit from a spatially changing RBE and discuss some of the technological challenges they may require.

60 APPLIED LIFE SCIENCES↗

labquake_future_prediction

The labquake_future_prediction code is a collection of python modules and scripts that serves as supporting information for the article “Predicting future laboratory fault friction through deep learning” for publication in the journal of “Geophysical Research Letters”. It is designed to predict laboratory fault slips in the immediate future by scanning continuous acoustic emission (AE) waveforms recorded in laboratory biaxial shear experiments. The predictions are made with a deep learning model based on convolutional encoder-decoder (CED) models and the Transformer model primarily developed for Natural Language Processing (NLP). The deep learning model is trained with the tensorflow package using publicly available laboratory data sets in standard binary file format in numpy. The utility functions for reading data files, configuring model hyperparameters, constructing the CED and Transformer models, training and testing of the models are defined in python module files. The workflow of training the models for labquake future predictions and the multiple GPU’s rapid model hyperparameter optimization as described in the journal article, are demonstrated in accompanying python script files and Jupyter notebooks.

Wang, Kun↗

Modular machine learning-based elastoplasticity: Generalization in the context of limited data

The development of highly accurate constitutive models for materials that undergo path-dependent processes continues to be a complex challenge in computational solid mechanics. Challenges arise both in considering the appropriate model assumptions and from the viewpoint of data availability, verification, and validation. Recently, data-driven modeling approaches have been proposed that aim to establish stress-evolution laws that avoid user-chosen functional forms by relying on machine learning representations and algorithms. However, these approaches not only require a significant amount of data but also need data that probes the full stress space with a variety of complex loading paths. Furthermore, they rarely enforce all necessary thermodynamic principles as hard constraints. Hence, they are in particular not suitable for low-data or limited-data regimes, where the first arises from the cost of obtaining the data and the latter from the experimental limitations of obtaining labeled data, which is commonly the case in engineering applications. In this work, we discuss a hybrid framework that can work on a variable amount of data by relying on the modularity of the elastoplasticity formulation where each component of the model can be chosen to be either a classical phenomenological or a data-driven model depending on the amount of available information and the complexity of the response. The method is tested on synthetic uniaxial data coming from simulations as well as cyclic experimental data for structural materials. The discovered material models are found to not only interpolate well but also allow for accurate extrapolation in a thermodynamically consistent manner far outside the domain of the training data. This ability to extrapolate from limited data was the main reason for the early and continued success of phenomenological models and the main shortcoming in machine learning-enabled constitutive modeling approaches. Training aspects and details of the implementation of these models into Finite Element simulations are discussed and analyzed.

42 ENGINEERING↗

Assembling a Cyber Range to Evaluate Artificial Intelligence / Machine Learning (AI/ML) Security Tools

In this case study, we will describe the design and assembly of a cyber security test range we have built at Oak Ridge National Laboratory in Oak Ridge, TN, USA. The range is designed to provide a flexible environment to evaluate cyber security tools—particularly those involving AI/ML—in a way that provides realistic environments and where we can control the experiments to determine the strengths and weaknesses of the tools. We have designed in the ability to repeat the evaluations, so additional tools can be evaluated and compared at a later time. The system is one that can be scaled up or down for experiment sizes. At the time of the conference we will have completed two full-scale, national, government challenges on this range. These challenges are evaluating the performance and operating costs for AI/ML-based cyber security tools for application into large, government-sized environments. These evaluations will be described, in order to provide motivation and context for various design decisions and adaptations we have made. The first challenge measured end-point security tools against 100K malware samples chosen across a range of types. The second is network detection of attempted penetration and exploitations with varying levels of covertness in a high-volume, business network. The scale of each of these challenges is requiring us to create automation systems to repeat the experiments identically for each tool. Preventing there being easy signs of malicious activity for the AI/ML tools to focus on has been a particularly interesting and challenging aspect of designing and executing these challenge events. After the events, the range continues to be used for other research such as adversarial machine learning where the repeatability, scale, and automation required for the national challenge events become essential elements for research.

Nichols, Jeff↗

Lessons Learned in International Safeguards - Implementation of Safeguards at the Rokkasho Reprocessing Plant

The focus of this report is lessons learned at the Rokkasho Reprocessing Plant (RRP). However, the subject of lessons learned for application of international safeguards at reprocessing plants includes a cumulative history of inspections starting at the West Valley (New York, U.S.A.) reprocessing plant in 1969 and proceeding through all of the efforts over the years. The RRP is the latest and most challenging application the International Atomic Energy Agency has faced. In many ways the challenges have remained the same, timely inspection and evaluation with limited inspector resources, with the continuing realization that planning and preparations can never start early enough in the life cycle of a facility. Lessons learned over the years have involved the challenges of using ongoing advances in technology and dealing with facilities with increased throughput and continuous operation. This report will begin with a review of historical developments and lessons learned. This will provide a basis for a discussion of the experiences and lessons learned from the implementation of international safeguards at RRP.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

MODIS Aerosol Optical Depth Bias Adjustment Using Machine Learning Algorithms

To monitor the earth atmosphere and its surface changes, satellite based instruments collect continuous data. While some of the data is directly used, some others such as aerosol properties are indirectly retrieved from the observation data. While retrieved variables (RV) form very powerful products, they don't come without obstacles. Different satellite viewing geometries, calibration issues, dynamically changing atmospheric and earth surface conditions, together with complex interactions between observed entities and their environment affect them greatly. This results in random and systematic errors in the final products.

Albayrak, Arif↗

Thermal Infrared Sensor (TIRS) Instrument Thermal Subsystem Design and Lessons Learned

The Thermal Infrared Sensor (TIRS) is one of two instruments on the Landsat Data Continuity Mission (LDCM), which is scheduled to launch in February of 2013. The TIRS instrument was officially added to the mission later in the flow, which led to a highly aggressive schedule that became one of the main drivers during instrument development. The thermal subsystem design of the TIRS Sensor Unit is comprised of five thermal zones which range in temperature from less than 43 Kelvin to 330 Kelvin. Most zones are proportional heater controlled, and all are within a volume of 35 cu.ft. A two-stage cryocooler is used to cool the "cold stage" including three QWIP detectors to less than 43 Kelvin, and cool the "warm stage" to 105 Kelvin. The excess power dissipation from the cryocooler is rejected via ammonia transport heat pipes to a dedicated Cryocooler Radiator with embedded ammonia heat pipes. The cryogenic subsystem includes a series of shells used to radiatively and conductively isolate the cold stage from the warmer surroundings. The Optical System (telescope) is passively cooled to 180-190 Kelvin using a "thermal link" (comprised of a Flexible Conductive Thermal Strap and an APG Bar) which couples the telescope stage to a dedicated radiator with embedded ethane heat pipes. The Scene Select Mechanism, which is responsible for moving the Scene Select Mirror to three distinct positions (including Nadir, Space, and On-board Black Body Calibrator pointing), runs nominally at 278 Kelvin and is thermally isolated from the cryogenic thermal zones. The On-board Black Body Calibrator requires a dedicated radiator which allows for a temperature range of 260-330 Kelvin at the Source. The detectors are powered by the FPE Box, which is mounted to the nadir external surface of the composite honeycomb structure. There are two additional electronics boxes which are wet-mounted directly to the spacecraft shear panel, the Main Electronics Box and Cryocooler Electronics Box; thermal control of these boxes is the responsibility of Orbital Sciences Corporation, the spacecraft developer. The TIRS thermal subsystem design was successfully verified during months of testing campaign, from component & subsystem level to two instrument-level thermal vacuum tests. The Instrument, despite an aggressive schedule, was delivered to the spacecraft vendor in February of 2012 and is currently undergoing the final stages of spacecraft environmental testing in preparation for launch.

Otero, Veronica↗

Semi-supervised learning approach for crack detection and identification in advanced gas-cooled reactor graphite bricks - 111

One of the life-limiting components of an Advanced Gas cooled Reactor (AGR) is its graphite core. The bricks present in the core undergo radiolytic oxidation throughout their lifetime which causes graphite weight loss and irradiation which can result in some of the bricks developing cracks. Understanding the nature and extent of brick cracking within the core is key to ensuring continued and extended operation of the AGR fleet. A semi-supervised machine learning classification algorithm is proposed as a method for improving the detection of cracked graphite bricks, by combining the labels derived from infrequent, detailed inspections of the core, with unlabeled, more frequent monitoring measurements taken during refueling operations. Semi-supervised machine learning, which is an emerging field in nuclear power condition monitoring, is the combination of ideas from both supervised and unsupervised machine learning whereby the data that is used to train the algorithm is a combination of labeled and unlabeled data. This paper introduces the initial research that has been undertaken in creating a semi-supervised self-training algorithm to detect the presence of graphite brick cracks and then proceeds to show that there is an improvement in the classification of graphite bricks using a semi-supervised machine learning classifier compared to supervised machine learning classifiers. This improved classification performance is encouraging as it does not require time consuming and costly human analysis to obtain extra learning information from available data. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Making or Breaking a Rover: System Engineering Parameters On-Board the Mars 2020 Perseverance Rover

On February 18, 2021, Perseverance, NASA’s Jet Propulsion Laboratory’s (JPL’s) Mars 2020 Rover, successfully landed on Mars with all systems nominal, despite the risk surrounding the over 200,000 internal flight parameters that had to be properly configured. The Perseverance team defines these parameters as software variables that are configurable, commandable and retrievable from Earth. In 2015, the Mars 2020 project leaders focused on improving systems engineering of parameters based on their experiences from parameter management on previous Mars rovers (Curiosity, Opportunity, Spirit, and Pathfinder) and parameter failures of past missions, such as the mission-ending parameter of the Mars Climate Orbiter. The new rigorous development process allowed for efficient certification and effective implementation of the parameters, allowing the rover to approach and land on the red planet (the most challenging phase of the mission) with zero parameter issues. Although successful, the Perseverance team learned many lessons for how to better manage parameters for the continued surface operations of the Mars 2020 mission and future missions. This paper will discuss eight parameter-management topics for the Perseverance Mission. The first is parameter definition: how we define parameters on our mission, where they are physically located on the vehicle, and why we have so many of them. The second topic is the updated parameter flight software module from Curiosity, including details on the 99% reduction in parameter commands, new bulk configuration capabilities, and improved parameter traceability. The third topic is parameter selection for different mission phases; this includes improving and tweaking our preferred parameter settings until they become certification candidates and managing parameter configurations based on test venue throughout the mission life cycle. The fourth topic is our flight certification process; this includes certification of flight values for four different epochs in the mission: Launch, Entry Decent and Landing (EDL) - 6days, Landing + 5 Sols (Martian Days, still on Cruise Flight Software), and once are on Surface Flight Software (FSW). The fifth topic covers in-flight command implementation, along with details on testing, validation, and verification of those commands. In the sixth section, we will explain our use of open-source management tools, including how we used GitHub for version control and management approvals. The seventh topic will describe the ground tools used in operations, including capabilities of the in-house built tool called Parasol. The eighth and final topic will dig into lessons learned for improving parameter management in the future of this mission and others.

Roth, Brian↗

Making or Breaking a Rover- Systems Engineering Parameters On-Board the Mars 2020 Perseverance Rover

On February 18, 2021, Perseverance, NASA’s Jet Propulsion Laboratory’s (JPL’s) Mars 2020 Rover, successfully landed on Mars with all systems nominal, despite the risk surrounding the over 200,000 internal flight parameters that had to be properly configured. The Perseverance team defines these parameters as software variables that are configurable, commandable and retrievable from Earth. In 2015, the Mars 2020 project leaders focused on improving systems engineering of parameters based on their experiences from parameter management on previous Mars rovers (Curiosity, Opportunity, Spirit, and Pathfinder) and parameter failures of past missions, such as the mission-ending parameter of the Mars Climate Orbiter. The new rigorous development process allowed for efficient certification and effective implementation of the parameters, allowing the rover to approach and land on the red planet (the most challenging phase of the mission) with zero parameter issues. Although successful, the Perseverance team learned many lessons for how to better manage parameters for the continued surface operations of the Mars 2020 mission and future missions. This paper will discuss eight parameter-management topics for the Perseverance Mission. The first is parameter definition: how we define parameters on our mission, where they are physically located on the vehicle, and why we have so many of them. The second topic is the updated parameter flight software module from Curiosity, including details on the 99% reduction in parameter commands, new bulk configuration capabilities, and improved parameter traceability. The third topic is parameter selection for different mission phases; this includes improving and tweaking our preferred parameter settings until they become certification candidates and managing parameter configurations based on test venue throughout the mission life cycle. The fourth topic is our flight certification process; this includes certification of flight values for four different epochs in the mission: Launch, Entry Decent and Landing (EDL) - 6days, Landing + 5 Sols (Martian Days, still on Cruise Flight Software), and once are on Surface Flight Software (FSW). The fifth topic covers in-flight command implementation, along with details on testing, validation, and verification of those commands. In the sixth section, we will explain our use of open-source management tools, including how we used GitHub for version control and management approvals. The seventh topic will describe the ground tools used in operations, including capabilities of the in-house built tool called Parasol. The eighth and final topic will dig into lessons learned for improving parameter management in the future of this mission and others.

Roth, Brian↗

Uncertainty-Aware, Structure-Preserving Machine Learning Approach for Domain Shift Detection From Nonlinear Dynamic Responses of Structural Systems

Complex structural systems deployed for aerospace, civil, or mechanical applications must operate reliably under varying operational conditions. Structural health monitoring (SHM) systems help ensure the reliability of these systems by providing continuous monitoring of the state of the structure. SHM relies on synthesizing measured data with a predictive model to make informed decisions about structural states. However, these models—which may be thought of as a form of a digital twin—need to be updated continuously as structural changes (e.g., due to damage) arise. We propose an uncertainty-aware machine learning model that enforces distance preservation of the original input state space and then encodes a distance-aware mechanism via a Gaussian process (GP) kernel. The proposed approach leverages the spectral-normalized neural GP algorithm to combine the flexibility of neural networks with the advantages of GP, subjected to structure-preserving constraints, to produce an uncertainty-aware model. This model is used to detect domain shift due to structural changes that cannot be observed directly because they may be spatially isolated (e.g., inside a joint or localized damage). This work leverages detection theory to detect domain shift systematically given statistical features of the prediction variance produced by the model. The proposed approach is demonstrated on a nonlinear structure being subjected to damage conditions. In conclusion, it is shown that the proposed approach is able to rely on distances of the transformed input state space to predict increased variance in shifted domains while being robust to normative changes.

Algorithms↗