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 901 records · Page 50

Project VISION (Very Intensive Scientific Intercurricular On-Site Education

Project VISION (Very Intensive Scientific Intercurricular On-Site Education) is a joint effort among NASA/John F. Kennedy Space Center, Florida International University, Universidad del Turabo, Miami-Dade County Public Schools and the Caguas/Gurabo Public Schools in Puerto Rico. The project's main mission is to institutionalize change among the elementary and middle school science and math teachers at participating schools so that their students receive continuously enriched instruction in the principles of science and math through the use of hands-on and minds-on experiments called learning modules. These leaming modules incorporate the national science and math education standards provided by the National Committee on Science Education Standards and Assessments and the National Council of Teachers of Mathematics, respectively. The use of learning modules that require hands-on and minds-on activities in a classroom setting garners great enthusiasm and motivation on the part of the target students for the understanding of the lesson's underlying math and science principles. With this enthusiasm and motivation, comes acceptance, attention, participation, discipline, acquiescence, and collaboration. Additionally, the use of hands-on activities may also require learning through a gamut of senses. Not only can the student use his/her eyes and ears during these activities, but most times, they can also use their senses of touch, smell, and taste, as well as intuition. Learning is, therefore, achieved using most or all the human senses. The combination of motivation/enthusiasm and the use of multiple senses creates an ideal environment conducive to leaming at a profound level.

Roig, Gustavo A.↗

Clinical Core Competency Training for NASA Flight Surgeons

Introduction: The cohort of NASA flight surgeons (FS) is a very accomplished group with varied clinical backgrounds; however, the NASA Flight Surgeon Office has identified that the extremely demanding schedule of this cohort prevents many of these physicians from practicing clinical medicine on a regular basis. In an effort to improve clinical competency, the NASA FS Office has dedicated one day a week for the FS to receive clinical training. Each week, an FS is assigned to one of five clinical settings, one being medical patient simulation. The Medical Operations Support Team (MOST) was tasked to develop curricula using medical patient simulation that would meet the clinical and operational needs of the NASA FS Office. Methods: The MOST met with the Lead FS and Training Lead FS to identify those core competencies most important to the FS cohort. The MOST presented core competency standards from the American Colleges of Emergency Medicine and Internal Medicine as a basis for developing the training. Results: The MOST identified those clinical areas that could be best demonstrated and taught using medical patient simulation, in particular, using high fidelity human patient simulators. Curricula are currently being developed and additional classes will be implemented to instruct the FS cohort. The curricula will incorporate several environments for instruction, including lab-based and simulated microgravity-based environments. Discussion: The response from the NASA FS cohort to the initial introductory class has been positive. As a result of this effort, the MOST has identified three types of training to meet the clinical needs of the FS Office; clinical core competency training, individual clinical refresher training, and just-in-time training (specific for post-ISS Expedition landings). The MOST is continuing to work with the FS Office to augment the clinical training for the FS cohort, including the integration of Web-based learning.

Polk, J. D.↗

Execution of the Spitzer In-orbit Checkout and Science Verification Plan

The Spitzer Space Telescope is an 85-cm telescope with three cryogenically cooled instruments. Following launch, the observatory was initialized and commissioned for science operations during the in-orbit checkout (IOC) and science verification (SV) phases, carried out over a total of 98.3 days. The execution of the IOC/SV mission plan progressively established Spitzer capabilities taking into consideration thermal, cryogenic, optical, pointing, communications, and operational designs and constraints. The plan was carried out with high efficiency, making effective use of cryogen-limited flight time. One key component to the success of the plan was the pre-launch allocation of schedule reserve in the timeline of IOC/SV activities, and how it was used in flight both to cover activity redesign and growth due to continually improving spacecraft and instrument knowledge, and to recover from anomalies. This paper describes the adaptive system design and evolution, implementation, and lessons learned from IOC/SV operations. It is hoped that this information will provide guidance to future missions with similar engineering challenges

operations↗

Grease-Resistant O Rings for Joints in Solid Rocket Motors

There is a continuing effort to develop improved O rings for sealing joints in solid-fuel rocket motors. Following an approach based on the lessons learned in the explosion of the space shuttle Challenger, investigators have been seeking O-ring materials that exhibit adequate resilience for effective sealing over a broad temperature range: What are desired are O rings that expand far and fast enough to maintain seals, even when metal sealing surfaces at a joint move slightly away from each other shortly after ignition and the motor was exposed to cold weather before ignition. Other qualities desired of the improved O rings include adequate resistance to ablation by hot rocket gases and resistance to swelling when exposed to hydrocarbon-based greases used to protect some motor components against corrosion. Five rubber formulations two based on a fluorosilicone polymer and three based on copolymers of epichlorohydrin with ethylene oxide were tested as candidate O-ring materials. Of these, one of the epichlorohydrin/ethylene oxide formulations was found to offer the closest to the desired combination of properties and was selected for further evaluation.

Harvey, Albert R.↗

Use of Continuous Integration Tools for Application Performance Monitoring

High performance computing systems are becom- ing increasingly complex, both in node architecture and in the multiple layers of software stack required to compile and run applications. As a consequence, the likelihood is increasing for application performance regressions to occur as a result of routine upgrades of system software components which interact in complex ways. The purpose of this study is to evaluate the effectiveness of continuous integration tools for application performance monitoring on HPC systems. In addition, this paper also describes a prototype system for application perfor- mance monitoring based on Jenkins, a Java-based continuous integration tool. The monitoring system described leverages several features in Jenkins to track application performance results over time. Preliminary results and lessons learned from monitoring applications on Cray systems at the Oak Ridge Leadership Computing Facility are presented.

Vergara Larrea, Veronica G↗

Power Grid Planning with Higher Renewable Share

Power grid planning with a higher renewable share got worldwide attention since the declining cost of renewable energy technologies and greenhouse gas reduction goals continue to drive a transition to a cleaner power grid. Energy professionals from developing countries working to achieve renewable power goals want to learn about NREL's state-of-the-art research efforts in renewable power and power grid planning. Supporting these efforts, the United States Energy Association, Female Leaders in Energy initiative facilitates mid-career female energy professionals from South East Asia to learn from energy professionals from the USA. This presentation gives an overview of NREL scientific research studies on power grid planning with a high renewable share to mid-career female energy professionals.

ENERGY PLANNING, POLICY, AND ECONOMY,POWER TRANSMI↗

Model-Based GN and C Simulation and Flight Software Development for Orion Missions beyond LEO

For Orion missions beyond low Earth orbit (LEO), the Guidance, Navigation, and Control (GN&C) system is being developed using a model-based approach for simulation and flight software. Lessons learned from the development of GN&C algorithms and flight software for the Orion Exploration Flight Test One (EFT-1) vehicle have been applied to the development of further capabilities for Orion GN&C beyond EFT-1. Continuing the use of a Model-Based Development (MBD) approach with the Matlab®/Simulink® tool suite, the process for GN&C development and analysis has been largely improved. Furthermore, a model-based simulation environment in Simulink, rather than an external C-based simulation, greatly eases the process for development of flight algorithms. The benefits seen by employing lessons learned from EFT-1 are described, as well as the approach for implementing additional MBD techniques. Also detailed are the key enablers for improvements to the MBD process, including enhanced configuration management techniques for model-based software systems, automated code and artifact generation, and automated testing and integration.

Odegard, Ryan↗

Demonstration of Machine Learning-Based Model-Independent Stabilization of Source Properties in Synchrotron Light Sources

Synchrotron light sources, arguably among the most powerful tools of modern scientific discovery, are presently undergoing a major transformation to provide orders of magnitude higher brightness and transverse coherence enabling the most demanding experiments. In these experiments, overall source stability will soon be limited by achievable levels of electron beam size stability, presently on the order of several microns, which is still 1–2 orders of magnitude larger than already demonstrated stability of source position and current. Until now source size stabilization has been achieved through corrections based on a combination of static predetermined physics models and lengthy calibration measurements, periodically repeated to counteract drift in the accelerator and instrumentation. In this paper, we now demonstrate for the first time how the application of machine learning allows for a physics- and model-independent stabilization of source size relying only on previously existing instrumentation. Such feed-forward correction based on a neural network that can be continuously online retrained achieves source size stability as low as 0.2 μm (0.4%) rms, which results in overall source stability approaching the subpercent noise floor of the most sensitive experiments.

43 PARTICLE ACCELERATORS↗

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

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

60 APPLIED LIFE SCIENCES↗

Hybrid Entanglement Distribution between Remote Microwave Quantum Computers Empowered by Machine Learning

Superconducting microwave circuits with Josephson junctions, the major platform for quantum computing, can only reach the full capability when connected. This requires an efficient protocol to distribute microwave entanglement. While quantum computers typically use discrete-variable (DV) methods for information encoding, the entire continuous-variable (CV) degree of freedom in electromagnetic fields must be utilized to achieve the highest entanglement distribution rate. Here, we propose a hybrid protocol to resolve the incompatibility between DV microwave quantum computers and CV quantum communications. CV microwave entanglement is distributed using optical swapping of optical-microwave entanglement pairs. To interface with DV microwave quantum computers, we further design a hybrid circuit to simultaneously convert and distill high-quality DV entanglement from noisy CV entanglement. The hybrid circuit is trained with machine-learning algorithms, ensuring high entanglement fidelity and generation rate. In conclusion, our work not only provides a practical method to realize efficient quantum links for superconducting microwave quantum computers, but also opens avenues to bridge the gap between DV and CV quantum systems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Structural integrity of future aging airplanes

A multitude of design considerations is involved in ensuring the structural integrity of Boeing jet transports that have common design concepts validated by extensive analyses, tests, and three decades of service. As airplanes approach their design service objectives, the incidences of fatigue and corrosion may become widespread. Continuing airworthiness of the aging jet fleet requires diligent performance from the manufacturer, the airlines, and airworthiness authorities. Aging fleet support includes timely development of supplemental structural inspection documents applicable to selected older airplanes, teardown inspections of high-time airframes retired from service, fatigue testing of older airframes, and structural surveys of more than 130 airplanes operated throughout the world. Lessons learned from these activities are incorporated in service bulletin recommendations, production line modifications, and design manual updates. An overview of traditional Boeing fleet support activities and the anticipated benefits for future generations of commercial airplanes based on the continuous design improvement process are presented.

Aging Aircraft↗

Identifying Light-Duty Vehicle Travel from Large-Scale Multimodal Wearable GPS Data with Novelty Detection Algorithms

Identifying travel mode within travel survey data sets, especially light-duty vehicle (LDV) travel, is foundational, though nontrivial, to travel behavior analysis and fuel consumption estimation. Current travel mode detection approaches require well-sampled and balanced data sets with ground truth travel mode labels. They are rarely applied and validated on large-scale, real-world data sets, which may not satisfy the data requirements. This paper proposes an LDV travel mode detection model as a supplement to current travel mode detection methods, for the case when the training set is highly (and/or completely) unbalanced, to the extent that classical machine-learning approaches become difficult or impossible to deploy. The proposed model uses a novelty detection technique-one-class support vector machines (OCSVMs)-and a novel exhaustive feature extraction (EFE) technique on continuous time series data (i.e., Global Positioning System [GPS] speed profiles) for single-mode trip trajectories. Training and validation of the model are conducted on a large-scale, real-world data set. The proposed method accurately identifies LDV trips from a broad set of multimodal trips by leveraging a wealth of preexisting in-vehicle GPS travel data. Additional sensitivity analysis sheds light on the optimal training size, which will benefit applications limited by highly imbalanced data. The paper also discusses performance comparison with regular machine-learning approaches, the model's robustness, and the potential to extend the proposed model to multimodal prediction.

47 OTHER INSTRUMENTATION↗

Adaptive Data-Driven Model Predictive Control for Heat Pipe Microreactors

To establish a technical basis for self-regulating microreactors, a model predictive control (MPC) system is investigated to proactively respond to anomalies and disturbances in anticipation of potential deviations from operating setpoints. Due to the difficulty of developing a physics-based surrogate model that can accurately match plant data in various operating conditions, machine learning algorithms are used in MPC, which allow for learning from both simulation and operation data, thus efficiently describing the targeted transient with arbitrary accuracy. However, one of the biggest concerns in applying ML algorithms like artificial neural networks (ANNs) is that the predictive capabilities of ANN are limited by training data. If there are gaps between the training and target domain, the accuracy of an ANN can degrade significantly when it is used to predict unseen data. To improve the predictive capability of ANN and enable a confident use of data-driven MPCs outside the training data, this study proposes an adaptive data-driven MPC framework. The system will monitor the discrepancy between plant responses and surrogate predictions, fine-tune the ANN-based surrogate when a large discrepancy is detected, and continue MPC operation with updated surrogates. The framework is demonstrated on a point kinetic model for microreactors. The hyperparameters of the update strategy, including layers to update, error thresholds, learning rate discount, and number of data points used for fine-tuning, are optimized so the simulated microreactor is able to follow changes in setpoint with the smallest of deviations.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Adaptable Data Driven Model Predictive Control for Heat Pipe Microreactors

To establish a technical basis for self-regulating microreactors, a model predictive control (MPC) system is investigated to proactively respond to anomalies and disturbances in anticipation of potential deviations from operating setpoints. Due to the difficulty of developing a physics-based surrogate model that can accurately match plant data in various operating conditions, machine learning algorithms are used in MPC, which allow for learning from both simulation and operation data, thus efficiently describing the targeted transient with arbitrary accuracy. However, one of the biggest concerns in applying ML algorithms like artificial neural networks (ANNs) is that the predictive capabilities of ANN are limited by training data. If there are gaps between the training and target domain, the accuracy of an ANN can degrade significantly when it is used to predict unseen data. To improve the predictive capability of ANN and enable a confident use of data-driven MPCs outside the training data, this study proposes an adaptive data-driven MPC framework. The system will monitor the discrepancy between plant responses and surrogate predictions, fine-tune the ANN-based surrogate when a large discrepancy is detected, and continue MPC operation with updated surrogates. The framework is demonstrated on a point kinetic model for microreactors. The hyperparameters of the update strategy, including layers to update, error thresholds, learning rate discount, and number of data points used for fine-tuning, are optimized so the simulated microreactor is able to follow changes in setpoint with the smallest of deviations.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

MIONet: Learning Multiple-Input Operators via Tensor Product

As an emerging paradigm in scientific machine learning, neural operators aim to learn operators, via neural networks, that map between infinite-dimensional function spaces. Several neural operators have been recently developed. However, all the existing neural operators are only designed to learn operators defined on a single Banach space; i.e., the input of the operator is a single function. Here, for the first time, we study the operator regression via neural networks for multiple-input operators defined on the product of Banach spaces. We first prove a universal approximation theorem of continuous multiple-input operators. We also provide a detailed theoretical analysis including the approximation error, which provides guidance for the design of the network architecture. Based on our theory and a low-rank approximation, we propose a novel neural operator, MIONet, to learn multiple-input operators. MIONet consists of several branch nets for encoding the input functions and a trunk net for encoding the domain of the output function. Here, we demonstrate that MIONet can learn solution operators involving systems governed by ordinary and partial differential equations. In our computational examples, we also show that we can endow MIONet with prior knowledge of the underlying system, such as linearity and periodicity, to further improve accuracy.

97 MATHEMATICS AND COMPUTING↗

Defect detection in atomic-resolution images via unsupervised learning with translational invariance

Abstract Crystallographic defects can now be routinely imaged at atomic resolution with aberration-corrected scanning transmission electron microscopy (STEM) at high speed, with the potential for vast volumes of data to be acquired in relatively short times or through autonomous experiments that can continue over very long periods. Automatic detection and classification of defects in the STEM images are needed in order to handle the data in an efficient way. However, like many other tasks related to object detection and identification in artificial intelligence, it is challenging to detect and identify defects from STEM images. Furthermore, it is difficult to deal with crystal structures that have many atoms and low symmetries. Previous methods used for defect detection and classification were based on supervised learning, which requires human-labeled data. In this work, we develop an approach for defect detection with unsupervised machine learning based on a one-class support vector machine (OCSVM). We introduce two schemes of image segmentation and data preprocessing, both of which involve taking the Patterson function of each segment as inputs. We demonstrate that this method can be applied to various defects, such as point and line defects in 2D materials and twin boundaries in 3D nanocrystals.

36 MATERIALS SCIENCE↗

Occlusion-Perturbed Deep Learning for Probabilistic Solar Forecasting via Sky Images

Solar forecasting is shifting to the probabilistic paradigm due to the inherent uncertainty within the solar resource. Input uncertainty quantification is one of the widely used and best-performing ways to model solar uncertainty. However, compared to other sources of inputs, such as numerical weather prediction models, pure sky image-based probabilistic solar forecasting lags behind. In this research, an occlusion-perturbed convolutional neural network, named the PSolarNet, is developed. The PSolarNet provides very short-term deterministic forecasts, forecast scenarios, and probabilistic forecasts of the global horizontal irradiance from sky image sequences. Case studies based on 6 years of open-source data show that the developed PSolarNet is able to generate accurate 10-minute ahead deterministic forecasts with a 5.62% normalized root mean square error, realistic and diverse forecast scenarios with a 0.966 average correlation with the actual time series, and reliable and sharp probabilistic forecasts with a 2.77% normalized continuous ranked probability score.

Bayesian model averaging↗

Long-term missing value imputation for time series data using deep neural networks

We present an approach that uses a deep learning model, in particular, a MultiLayer Perceptron, for estimating the missing values of a variable in multivariate time series data. We focus on filling a long continuous gap (e.g., multiple months of missing daily observations) rather than on individual randomly missing observations. Our proposed gap filling algorithm uses an automated method for determining the optimal MLP model architecture, thus allowing for optimal prediction performance for the given time series. We tested our approach by filling gaps of various lengths (three months to three years) in three environmental datasets with different time series characteristics, namely daily groundwater levels, daily soil moisture, and hourly Net Ecosystem Exchange. We compared the accuracy of the gap-filled values obtained with our approach to the widely used R-based time series gap filling methods ImputeTS and mtsdi. The results indicate that using an MLP for filling a large gap leads to better results, especially when the data behave nonlinearly. Thus, our approach enables the use of datasets that have a large gap in one variable, which is common in many long-term environmental monitoring observations.

97 MATHEMATICS AND COMPUTING↗