Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Online 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 559 records · Page 31

SU-F-J-49: IGRT Credentialing in NCTN Trials

Purpose: To make Image Guided Radiation Therapy (IGRT) credentialing a more unified, consistent and efficient process across the entire National Clinical Trial Network (NCTN). Methods: IGRT plays a role in several advanced NCTN trials. Previously an institution had to be IGRT credentialed for each protocol. When institutions were allowed to use previous credentials for new protocols it was limited to the same disease site as the original credentialing. The credentialing was analyzed by the physics PI of the protocol. We consulted with several of these physicists to determine what is important to consider when reviewing submissions and to learn ways to apply credentialing more broadly. Results: For trials open in 2016, IGRT credentialing can be simplified to cover either boney anatomy or soft tissue. This revised credentialing will cover all disease sites based on the type of anatomy, unless otherwise stated within the protocol. Institutions will submit will complete an online questionnaire about their IGRT procedures. Boney anatomy requirements will include submission of data from 2 sequential fraction of both a patient aligned with boney anatomy and pelvic patient. Soft tissue will require similar submissions for a patient aligned using soft tissue and a pelvic patient. Institutions will only be required to submit the pelvic patient once. Data should be in DICOM format and includes planning CT set, RT structure set, RT plan file, RT dose file, localization images and spatial registration file (if available). Reviews will be done by IROC-Houston staff who will continue to provide feedback to the sites. Conclusion: This revised IGRT credentialing process will bring consistency, a savings in time and effort for both the IROC Houston QA office and to those institutions wanting to be credentialed to participate in NCTN Trials. Sponsored by NIH/NCI CA10953.

60 APPLIED LIFE SCIENCES↗

A Learning-based Supervisory Control Architecture for Electric Vehicle Charging System Paired with Energy Storages

A Machine Learning-based predictive model is developed for optimal dispatching energy storage system integrated with Electric Vehicle battery charger. The model is effectively trained using transfer learning algorithm and successfully validated via measurement data to achieve high fidelity and accuracy. The study findings provide possibilities for future development of the presented approach to implement computational heavy predictive algorithms with multi-step predictions in emerging modular bidirectional charger topologies to realize vehicle-to-grid (V2G) technology. The online tuning can be explored using the new measurements obtained during the operation, henceforth, enhancing the controller performance for integrating multiple DERs into distribution power networks that shows huge interest for the utilities and regulatory agencies. Multiple operation scenatios have been evaluated and simulation results are discussed to verify intended performance of the proposed control solution.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Real World Connections Through Videoconferences

The Learning Technologies Project (LTP) is a partner in the National Aeronautics and Space Administration's (NASA's) educational technology program unit, an electronic community center that fosters interaction, collaboration, and sharing among educators, learners, and scientists. The goal of the NASA Glenn Research Center's Learning Technologies Project is to increase students' interest and proficiency in mathematics, science, and technology through the use of computing and communications technology and by using NASA's mission in aerospace technology as a theme. The primary components are: (1) Beginner's Guide to Aeronautics, including interactive simulation packages and teacher-created online activities. (2) NASA Virtual Visits, videoconferences (with online pre-post-conference activities) connecting students and teachers to NASA scientists and researchers.

Peterson, Ruth↗

Machine Learning Design of Perovskite Catalytic Properties

Abstract Discovering new materials that efficiently catalyze the oxygen reduction and evolution reactions is critical for facilitating the widespread adoption of solid oxide fuel cell and electrolyzer (SOFC/SOEC) technologies. Here, machine learning (ML) models are developed to predict perovskite catalytic properties critical for SOFC/SOEC applications, including oxygen surface exchange, oxygen diffusivity, and area specific resistance (ASR). The models are based on trivial‐to‐calculate elemental features and are more accurate and dramatically faster than the best models based on ab initio‐derived features, potentially eliminating the need for ab initio calculations in descriptor‐based screening. The model of ASR enables temperature‐dependent predictions, has well calibrated uncertainty estimates and online accessibility. Use of temporal cross‐validation reveals the model to be effective at discovering new promising materials prior to their initial discovery, demonstrating the model can make meaningful predictions. Using the SHapley Additive ExPlanations (SHAP) approach, detailed discussion of different approaches of model featurization is provided for ML property prediction. Finally, the model is used to screen more than 19 million perovskites to develop a list of promising cheap, earth‐abundant, stable, and high performing materials, and find some top materials contain mixtures of less‐explored elements (e.g., K, Bi, Y, Ni, Cu) worth exploring in more detail.

25 ENERGY STORAGE↗

Characterization and automated optimization of laser-driven proton beams from converging liquid sheet jet targets

Compact, stable, and versatile laser-driven ion sources hold great promise for applications ranging from medicine to materials science and fundamental physics. While single-shot sources have demonstrated favorable beam properties, including the peak fluxes necessary for several applications, high-repetition-rate operation will be necessary to generate and sustain the high average flux needed for many of the most exciting applications of laser-driven ion sources. Further, to navigate through the high-dimensional space of laser and target parameters toward experimental optima, it is essential to develop ion acceleration platforms compatible with machine learning techniques and capable of autonomous real-time optimization. Here, we present a multi-Hz ion acceleration platform employing a liquid sheet jet target. We characterize the laser-plasma interaction and the laser-driven proton beam across a variety of key parameters governing the interaction using an extensive suite of online diagnostics. We also demonstrate real-time, closed-loop optimization of the ion beam maximum energy by tuning the laser wave front using a Bayesian optimization scheme. This approach increased the maximum proton energy by 11% compared to a manually optimized wave front by enhancing the energy concentration within the laser focal spot, demonstrating the potential for closed-loop optimization schemes to tune future ion accelerators for robust high-repetition-rate operation.

Glenn, G. D. [SLAC National Accelerator Laboratory↗

Identifiers for the 21st century: How to design, provision, and reuse persistent identifiers to maximize utility and impact of life science data

In many disciplines, data are highly decentralized across thousands of online databases (repositories, registries, and knowledgebases). Wringing value from such databases depends on the discipline of data science and on the humble bricks and mortar that make integration possible; identifiers are a core component of this integration infrastructure. Drawing on our experience and on work by other groups, we outline 10 lessons we have learned about the identifier qualities and best practices that facilitate large-scale data integration. Specifically, we propose actions that identifier practitioners (database providers) should take in the design, provision and reuse of identifiers. We also outline the important considerations for those referencing identifiers in various circumstances, including by authors and data generators. While the importance and relevance of each lesson will vary by context, there is a need for increased awareness about how to avoid and manage common identifier problems, especially those related to persistence and web-accessibility/resolvability. We focus strongly on web-based identifiers in the life sciences; however, the principles are broadly relevant to other disciplines.

59 BASIC BIOLOGICAL SCIENCES↗

Dynamic and Responsive Distributed Energy Resource Education Solutions for Building, Fire, and Safety Department Officials (Final Technical Report)

From April 2021 through March 2024, the Interstate Renewable Energy Council (IREC) led a collaborative project to develop a free online clearinghouse of educational resources about solar photovoltaics (PV), energy storage systems (ESS), electric vehicle supply equipment (EVSE), and grid-interactive efficient building (GEB) technologies. Two websites—the Clean Energy Clearinghouse and CleanEnergyTraining.org—housed over 70 educational resources. Over the course of the three-year project, 154,272 unique visitors accessed the learning materials. Learner feedback was overwhelmingly positive. Even through the end of the project, there was sustained demand for education and communication. A primary innovation of the project was to drive multiple complementary audiences to the same place. Building owners, designers, installation contractors and developers, authorities having jurisdiction (AHJs), and fire service personnel all benefit from a shared understanding of clean energy technologies, including safety and code-related requirements. When considering the impact on the target audience, the project team worked with partners and advisors to inform resource creation and delivery in such a way as to address key motivational factors of the target audience and compel each user to seek additional information on the topic and return to the Clean Energy Clearinghouse website as their central location for more information. Resources were intentionally developed to be concise—five to 15 minutes—and accessible, meaning not overly technical. Providing basic information demystified the technologies and invited the professional to explore additional learning opportunities. Awardee and partner collaboration was key to project success. IREC facilitated collaboration among the other Topic 2 awardees, Southface and New Buildings Institute (NBI). The three awardees shared relevant information gained through discovery and validation questionnaires that informed product development and reduced duplication of effort by coordinating the development of complementary, and not competing, educational resources. Inspired by this collaboration, IREC brought on additional partners even in the final year of the project. Five regional energy efficiency organizations were part of the project, which expanded the connection between efficiency and distributed energy resources. We also included resources on the Clearinghouse that were developed through other federally funded projects, such as the Buildings Energy Efficiency Frontiers & Innovation Technologies (BENEFIT) program. The website was developed with the learner in mind, and not solely the funding source. Feedback from stakeholders throughout the project, and especially in its final year, indicated the need for continued education and facilitated communication among stakeholders to further the safe and widespread adoption of clean energy.

14 SOLAR ENERGY↗

Kernelized approaches to streaming compression of scientific data

In this paper three algorithms are developed for the streaming compression of scientific data. The algorithms presented are reliant on the theory of vector-valued reproducing kernel Hilbert spaces and operator valued kernel. Further, the scientific data is modeled as a snapshot of time dependent vector field F(x, t) over a manifold M and the recovery of the data is framed as a learning problem. These processes are then appropriately modified and ana lyzed for the streaming scenario in which data is generated without the ability to revisit past entries.

97 MATHEMATICS AND COMPUTING↗

Phasor-Measurement-Unit-Based Data Analytics Using Digital Twin and PhasorAnalytics Software

A major objective of this project was to apply GE’s commercial machine learning and data analytics toolsets to large-scale, real-world, anonymized Phasor Measurement Unit (PMU) datasets in order to extract signatures, correlated and/or causal factors, and precursor patterns associated with significant power system phenomena. The project had a particular emphasis on extraction of insights relevant to asset health monitoring, real-time load modeling and cybersecurity monitoring. Additionally, the team was directed to undertake a comprehensive data quality analysis for the provided datasets and encouraged to estimate the ‘machine-learning readiness’ of the datasets by documenting any major obstacles to the application of commercial machine learning algorithms. To accomplish the aforementioned objectives, the project team’s work centered around the identification of key event signatures and application of the identified event signatures for event detection and event classification. The industry-validated, semi-supervised machine learning strategy employed for event signature identification involved several major tasks, including data-preprocessing, generation of an overabundance of features, normal data identification, normality modeling, and event signature identification through a methodical, quantitative ranking of features in order of relevance to each studied event type. Throughout the project, data quality issues and mitigation techniques were investigated. In this report, insights are provided regarding the readiness of the provided synchrophasor datasets for application of machine learning and data analytics. The methodologies employed for this technical strategy are summarized in this report. With regards to data preprocessing and feature generation, the provided Training and Test Datasets were ingested into GE’s big data environment. Subsequently, the team applied bad data cleansing and data imputation scripts, event detection scripts, and application programming interfaces (APIs) to the datasets for convenient data access. The project team completed development and validation of dozens of physics-based, statistics-based and transformation-based feature functions used for the extraction of over 60 synchrophasor features. Using a new parallel feature generation technology developed on this project, over 60 features have been rapidly generated for the full two years’ worth of Training and Test Dataset data associated with both the Eastern and Western interconnects. Even accommodating for temporal down-sampling inherent to the feature extraction procedure, this parallel feature generation activity resulted in a massive feature set with a storage requirement approximately equal to that of the raw training dataset itself. With regards to normal data identification and normality modeling, a normality model was built using the feature data extracted from the Training Dataset and iteratively refined subsequent to incremental adjustments and expansions of the Training Dataset feature data. With respect to event characterization and signature identification, an event signature identification pipeline was developed and used in conjunction with the normality model to identify over 15 event signatures for key event categories within the Training Dataset. The identified event signatures were used to characterize hundreds of key events in terms of relative severity, duration, and location of the event. An investigation was undertaken to identify correlated and causal factors involved in transformer events. A separate investigation into temporal trends in ring-down analysis results was undertaken to determine possible associations between system dynamics and various other factors such as loading, season or year. To validate the identified event signatures, additional work was undertaken to develop signature-based anomaly detection and classification tools suitable for convenient application to the synchrophasor datasets. The anomaly detection and classification tools, suitable for online application, were then applied to the entirety of the Eastern Interconnect Training and Test Datasets. Performance of the event detection and classification tools was evaluated upon receipt of the Test Dataset event logs (i.e., the labels for events contained in the Test Dataset), and promising results were obtained despite several challenges (documented herein) associated with application of supervised or semi-supervised machine learning methods to large-scale, anonymized datasets. Finally, the detection and classification tools were used to detect, classify, and characterize thousands of new events not included in the original event logs provided by the DOE within both the Training and Test Datasets.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Improving Predictions Under Uncertainty of Material Plasma Device Operations

Understanding the properties of materials when exposed to various plasma temperatures and fluxes is essential to the building and operating of fusion reactors. The Material Plasma Exposure eXperiment (MPEX) is an instrument currently being developed by the Department of Energy (DOE) for this purpose. MPEX is expected to come online in stages over the next five years. Proto-MPEX, the predecessor to MPEX, operated from 2014 to 2021, and was designed to understand the generation of plasma temperatures and fluxes at orders of magnitude below what will be obtained by MPEX. This work uses the recently developed stochastic neural network (SNN), a machine learning technique capable of operating under uncertainty to provide a surrogate model for the Proto-MPEX device. We demonstrate that SNN outperforms Bayesian neural network (BNN), a standard in the field of machine learning with uncertainty. The development of a robust surrogate of the Proto-MPEX will aid in the commissioning and operation of the MPEX device.

Archibald, Rick↗

Block-Structured Operator Inference for Coupled Multiphysics Model Reduction

This work presents a block-structured formulation of Operator Inference as a way to learn structured reduced-order models for multiphysics systems. The approach specifies the governing equation structure for each physics component and the structure of the coupling terms. Once the multiphysics structure is specified, the reduced-order model is learned from snapshot data following the nonintrusive Operator Inference methodology. In addition to preserving physical system structure, which in turn permits preservation of system properties such as stability and second-order structure, the block-structured approach has the advantages of reducing the overall dimensionality of the learning problem and admitting tailored regularization for each physics component. The numerical advantages of the block-structured formulation over a monolithic Operator Inference formulation are demonstrated for aeroelastic analysis, which couples aerodynamic and structural models. For the benchmark test case of the AGARD 445.6 wing, block-structured Operator Inference provides an average 20% online prediction speedup over monolithic Operator Inference across subsonic and supersonic flow conditions in both the stable and fluttering parameter regimes while preserving the accuracy achieved with monolithic Operator Inference.

42 ENGINEERING↗

Biomass Harmonization and SAR Analysis with the Multi-mission Algorithm and Analysis Platform (MAAP)

The Multi‐mission Algorithm and Analysis Platform (MAAP) is a collaborative effort between NASA and the European Space Agency (ESA) to support above ground biomass (AGB) research in an open science framework. MAAP brings together relevant data, algorithms, and computing capabilities in a common cloud environment to address the challenges of sharing and processing data from field, airborne and satellite measurements. MAAP was publicly released in October 2021, providing computing capabilities co-located with the data, a collaborative coding and analysis environment, and a set of interoperable tools and algorithms developed to support the estimation and visualization of data. MAAP has allowed scientists from both North America and Europe to collaborate on the generation and analysis/visualization of data derived from multiple, discipline-adjacent missions in an open, collaborative environment that has reached beyond traditional scientific investigation. MAAP has been used to support multiple scientific activities. To date, existing LiDAR data from multiple platforms has been calibrated with field measurements and combined for more comprehensive and accurate estimates of above ground biomass AGB; these LiDAR platforms include airborne (e.g. LVIS), the International Space Station (NASA’s Global Ecosystem Dynamics Investigation (GEDI), and satellites (e.g. ICESat-2). The current challenge is to effectively and seamlessly combine the aforementioned LiDAR-based data with new data sources such as P-band RADAR from ESA’s upcoming BIOMASS mission, existing ESA Sentinel-1 C-band SAR, and the 30 PB/yr of high cadence global coverage L-band SAR data from the upcoming NASA-ISRO SAR (NISAR) mission. Recent analysis using MAAP merged ICESat-2 and optical data (Harmonized Landsat Sentinel) produced the most comprehensively precise estimate of boreal-wide AGB to date. Another effort using MAAP is the production and open distribution of global comparisons of AGB map estimates, including from ICESat-2 and GEDI, to bolster stakeholder uptake for policy applications. These map estimates will feed into the Intergovernmental Panel on Climate Change (IPCC) database, likely aiding the next Global Carbon Stocktake of the UNFCCC. Furthermore, the biomass retrieval intercomparison exercise BRIX-2 could benefit from the MAAP providing standardized test cases (based on airborne campaign and spaceborne data) allowing the community to develop and apply retrieval algorithms based on these test cases, while forthcoming SAR data training curricula could also use the MAAP as a teaching and learning platform. The MAAP is meeting the challenges inherent in international, open science collaboration and large scale computing with a platform that is entirely open source and cloud native, using open standards for data access, manipulation, protocols, and formats. The MAAP data system consists of a dedicated data store whose data is indexed in an online catalog conforming to established metadata, application programmatic interfaces (APIs), and service interface standards, using an implementation of the open sourced NASA Common Metadata Repository. Federation of user identities allows users from either NASA or ESA to access and consume services from the other using a unified metadata catalog for the data utilized across the ESA and NASA MAAP platforms. Similarly, we are exploring how to increase interoperability to achieve a common approach to packaging, orchestrating and executing algorithms, with interoperable access to data for subsetting, fast browse, and cloud-optimized access, all using interoperable standards such as those from the Open Geospatial Consortium (OGC). Designed for interoperability, ESA and NASA utilize a common architecture for the software platform. It provides a cloud-based algorithm development environment (ADE) that enables scientists to develop algorithms collaboratively with access to the MAAP data catalog as well as other data archives. MAAP provides an Eclipse Che-based ADE supporting both Python and R languages, popular in this biomass community. Algorithms developed and containerized within the ADE can be deployed to run to thousands of computational nodes in the MAAP’s data processing system (DPS), dramatically speeding up processing and giving scientists a rapid, iterative turnaround of results. NASA’s implementation of the DPS is based on the Hybrid Science Data System (HySDS) framework, used by NASA flight projects to produce Earth science standard products.

cloud computing↗

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

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

24 POWER TRANSMISSION AND DISTRIBUTION↗

Review of Particle Physics

Abstract The Review summarizes much of particle physics and cosmology. Using data from previous editions, plus 2,143 new measurements from 709 papers, we list, evaluate, and average measured properties of gauge bosons and the recently discovered Higgs boson, leptons, quarks, mesons, and baryons. We summarize searches for hypothetical particles such as supersymmetric particles, heavy bosons, axions, dark photons, etc. Particle properties and search limits are listed in Summary Tables. We give numerous tables, figures, formulae, and reviews of topics such as Higgs Boson Physics, Supersymmetry, Grand Unified Theories, Neutrino Mixing, Dark Energy, Dark Matter, Cosmology, Particle Detectors, Colliders, Probability and Statistics. Among the 120 reviews are many that are new or heavily revised, including a new review on Machine Learning, and one on Spectroscopy of Light Meson Resonances. The Review is divided into two volumes. Volume 1 includes the Summary Tables and 97 review articles. Volume 2 consists of the Particle Listings and contains also 23 reviews that address specific aspects of the data presented in the Listings. The complete Review (both volumes) is published online on the website of the Particle Data Group (pdg.lbl.gov) and in a journal. Volume 1 is available in print as the PDG Book. A Particle Physics Booklet with the Summary Tables and essential tables, figures, and equations from selected review articles is available in print, as a web version optimized for use on phones, and as an Android app.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Envision Charlotte Project Final Report

The Envision Charlotte Project (EC Project), supported by partners at the Center for Climate and Energy Solutions, the University of North Carolina at Charlotte’s Center for Sustainably Integrated Buildings and Sites and the City of Charlotte, has successfully demonstrated that existing buildings can benefit from operations improvements and low-cost efficiency upgrades, and that behavioral change campaigns can be effectively designed for measurability, longevity and impact. Support from this grant gave Envision Charlotte and its partners the resources and flexibility to demonstrate a variety of approaches to improve coordination with stakeholders in the implementation of programs to reduce the consumption of energy, water and waste. The project team also gained insights into the motivations and abilities that influence stakeholder engagement in sustainability initiatives at multiple levels. Importantly, as this report describes, strong partnerships between nonprofits, industry partners and research institutions can produce tangible impacts and replicable results in building efficiency. Further, building operators working alongside student researchers can form productive relationships that can advance building energy management goals. Key Findings of the EC Project This project focused on advancing solutions to energy efficiency in U.S. commercial buildings. The Building Re-Tuning™ Training (BRT) developed by the U.S. DOE’s Pacific Northwest National Lab (PNNL), designed to detect energy savings opportunities and implement improvements with resources and online tools, provided the starting point for the curriculum developed by stakeholders in the EC Project. Although the project partners encountered some issues with data consistency, the overall project outcomes fell short of the energy savings stated in the original proposal, and the total participating building targets were lower than anticipated, important findings, outcomes and lessons learned were achieved. EC Project grantees observed that the PNNL BRT, as currently written, was not compatible with the skills of average frontline mechanical personnel in the Charlotte region where the grant program was performed. Frontline staff are not, typically, energy managers, engineers or building commissioning agents, but they can play a key role in building performance. As detailed in this report, the grantees therefore focused on changes to the BRT that improved the level of training and competency of the frontline workers in this region. Among the technical observations related to improving the BRT that were observed in the EC Project: centralized facilities management often has an impact on building performance; operators can be overly reliant on service contracts; individual behaviors matter, and technology cannot be viewed as a crutch; current Building Automation Systems (BAS) are often too complex to support building efficiency optimization, but BAS coupled with appropriate training can support better outcomes; the “perfectly optimized” building is not a near-term possibility, but better education, especially of executives and managers, on the role of behavioral intervention is a promising direction the building science community can and should pursue; a lack of standards and protocols, and competition between proprietary software from service contracts, remain challenges for this industry. These technical observations, in turn, informed the EC Project’s overall findings, which were: Accomplishing building efficiency is a process of education for all involved in the building design, construction and operation supply chain. Building operators, as a professional class, have an opportunity for further training and education that will lead to better on-the-job performance. An updated Building Re-Tuning™ Training curriculum, based on the findings of this report, could be an avenue for these operators’ professional development.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

The Best Ever Alarm System Toolkit

Learning from our experience with the standard Experimental Physics and Industrial Control System (EPICS) alarm handler (ALH) as well as a similar intermediate approach based on script-generated operator screens, we developed the Best Ever Alarm System Toolkit (BEAST). It is based on Java and Eclipse on the Control System Studio (CSS) platform, using a relational database (RDB) to store the configuration and log actions. It employs a Java Message Service (JMS) for communication between the modular pieces of the toolkit, which include an Alarm Server to maintain the current alarm state, an arbitrary number of Alarm Client user interfaces (GUI), and tools to annunciate alarms or log alarm related actions. Web reports allow us to monitor the alarm system performance and spot deficiencies in the alarm configuration. The Alarm Client GUI not only gives the end users various ways to view alarms in tree and table, but also makes it easy to access the guidance information, the related operator displays and other CSS tools. It also allows online configuration to be simply modified from the GUI. Coupled with a good "alarm philosophy" on how to provide useful alarms, we can finally improve the configuration to achieve an effective alarm system.

43 PARTICLE ACCELERATORS↗

Station Program Note Pull Automation

Upon commencement of my internship, I was in charge of maintaining the CoFR (Certificate of Flight Readiness) Tool. The tool acquires data from existing Excel workbooks on NASA's and Boeing's databases to create a new spreadsheet listing out all the potential safety concerns for upcoming flights and software transitions. Since the application was written in Visual Basic, I had to learn a new programming language and prepare to handle any malfunctions within the program. Shortly afterwards, I was given the assignment to automate the Station Program Note (SPN) Pull process. I developed an application, in Python, that generated a GUI (Graphical User Interface) that will be used by the International Space Station Safety & Mission Assurance team here at Johnson Space Center. The application will allow its users to download online files with the click of a button, import SPN's based on three different pulls, instantly manipulate and filter spreadsheets, and compare the three sources to determine which active SPN's (Station Program Notes) must be reviewed for any upcoming flights, missions, and/or software transitions. Initially, to perform the NASA SPN pull (one of three), I had created the program to allow the user to login to a secure webpage that stores data, input specific parameters, and retrieve the desired SPN's based on their inputs. However, to avoid any conflicts with sustainment, I altered it so that the user may login and download the NASA file independently. After the user has downloaded the file with the click of a button, I defined the program to check for any outdated or pre-existing files, for successful downloads, to acquire the spreadsheet, convert it from a text file to a comma separated file and finally into an Excel spreadsheet to be filtered and later scrutinized for specific SPN numbers. Once this file has been automatically manipulated to provide only the SPN numbers that are desired, they are stored in a global variable, shown on the GUI, and transferred over to a new Excel worksheet for comparison. I managed to get my application to acquire the CSWG (Computer Safety Working Group) and the SPNWG (Space Station Working Group) SPN's with just two mouse clicks for each pull, as opposed to several from the original process. When all three pulls are performed, an Excel sheet containing all three different results will be generated for the user to compare and determine which SPN's will be presented or reviewed the following month. The experience from this internship has been spectacular. As a high school senior who will begin attending college in the fall, this internship has been both educationally and occupationally beneficial. The internship has allowed me the opportunities to learn new programming languages, effectively network with NASA personnel from a variety of departments at JSC, and allowed me to learn new professional skills and etiquette. My internship at NASA's Johnson Space Center has further motivated me to pursue a Master's degree in Software Engineering and strive for a prosperous career with NASA as a civil servant.

Delgado, Ivan↗

Bayesian Inference with Latent Hamiltonian Neural Networks (L-HNNs)

When sampling for Bayesian inference, one popular approach is to use Hamiltonian Monte Carlo (HMC) and the No-U-Turn Sampler (NUTS). However, HMC and NUTS can require numerous numerical gradients of the target density and can prove slow in practice. We propose Hamiltonian neural networks (HNNs) with HMC and NUTS for solving Bayesian inference problems [1, 2]. Once trained, HNNs do not require gradients of the target density while sampling. Moreover, they satisfy important properties such as perfect time reversibility and Hamiltonian conservation, making them well suited for use within HMC and NUTS because stationarity can be shown. We also propose an HNN extension called latent HNNs (L-HNNs), which predict latent variable outputs. Compared to HNNs, L-HNNs offer improved expressivity and a reduction in integration errors. Finally, we propose employing L-HNNs in NUTS with an online error monitoring scheme to prevent degeneracy of the sampling in regions of low probability density. We demonstrate L-HNNs in NUTS with online error monitoring by using several example cases involving complex, heavy-tailed, and high local curvature probability densities. Overall, L-HNNs in NUTS with online error monitoring satisfactorily inferred these probability densities. Compared to traditional NUTS, L-HNNs in NUTS with online error monitoring improved the effective sample size (ESS) per gradient by an order of magnitude.

97 MATHEMATICS AND COMPUTING↗