Search NASA⌕ Search

SEARCH · Search NASA

Results for “Operator 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 1,495 records · Page 83
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

Machine Learning-Based Anomaly Detection for Load Forecasting Under Cyberattacks

Accurate load forecasting can create both economic and reliability benefits for power system operators. However, the cyberattack on load forecasting may mislead operators to make unsuitable operational decisions for the electricity delivery. To effectively and accurately detect these cyberattacks, this paper develops a machine learning-based anomaly detection (MLAD) methodology. First, load forecasts provided by neural networks are used to reconstruct the benchmark and scaling data by using the k-means clustering. Second, the cyberattack template is estimated by the naive Bayes classification based on the cumulative distribution function and statistical features of the scaling data. Finally, the dynamic programming is utilized to calculate both the occurrence and parameter of one cyberattack on load forecasting data. Here, a widely used symbolic aggregation approximation method is compared with the developed MLAD method. Numerical simulations on the publicly load data show that the MLAD method can effectively detect cyberattacks for load forecasting data with relatively high accuracy. Also, the robustness of MLAD is verified by thousands of attack scenarios based on Monte Carlo simulation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Optimization and Multimachine Learning Algorithms to Predict Nanometal Surface Area Transfer Parameters for Gold and Silver Nanoparticles

Interactions between gold metallic nanoparticles and molecular dyes have been well described by the nanometal surface energy transfer (NSET) mechanism. However, the expansion and testing of this model for nanoparticles of different metal composition is needed to develop a greater variety of nanosensors for medical and commercial applications. In this study, the NSET formula was slightly modified in the size-dependent dampening constant and skin depth terms to allow for modeling of different metals as well as testing the quenching effects created by variously sized gold, silver, copper, and platinum nanoparticles. Overall, the metal nanoparticles followed more closely the NSET prediction than for Förster resonance energy transfer, though scattering effects began to occur at 20 nm in the nanoparticle diameter. To further improve the NSET theoretical equation, an attempt was made to set a best-fit line of the NSET theoretical equation curve onto the Au and Ag data points. An exhaustive grid search optimizer was applied in the ranges for two variables, 0.1≤C≤2.0 and 0≤α≤4, representing the metal dampening constant and the orientation of donor to the metal surface, respectively. Three different grid searches, starting from coarse (entire range) to finer (narrower range), resulted in more than one million total calculations with values C=2.0 and α=0.0736. The results improved the calculation, but further analysis needed to be conducted in order to find any additional missing physics. With that motivation, two artificial intelligence/machine learning (AI/ML) algorithms, multilayer perception and least absolute shrinkage and selection operator regression, gave a correlation coefficient, R2, greater than 0.97, indicating that the small dataset was not overfitting and was method-independent. This analysis indicates that an investigation is warranted to focus on deeper physics informed machine learning for the NSET equations.

Demers, Steven M. E. (ORCID:0000000192213246)↗
DOE OSTI2022

The L-CAPE Project at FNAL

The controls system at FNAL records data asynchronously from several thousand Linac devices at their respective cadences, ranging from 15Hz down to once per minute. In case of downtimes, current operations are mostly reactive, investigating the cause of an outage and labeling it after the fact. However, as one of the most upstream systems at the FNAL accelerator complex, the Linac’s foreknowledge of an impending downtime as well as its duration could prompt downstream systems to go into standby, potentially leading to energy savings. The goals of the Linac Condition Anomaly Prediction of Emergence (L-CAPE) project that started in late 2020 are (1) to apply data-analytic methods to improve the information that is available to operators in the control room, and (2) to use machine learning to automate the labeling of outage types as they occur and discover patterns in the data that could lead to the prediction of outages. We present an overview of the challenges in dealing with time-series data from 2000+ devices, our approach to developing an ML-based automated outage labeling system, and the status of augmenting operations by identifying the most likely devices predicting an outage.

43 PARTICLE ACCELERATORS↗
NASA NTRS2000

Experimental Verification of Electric Drive Technologies Based on Artificial Intelligence Tools

In this report, a fully integrated prototype of a flight servo control system is successfully developed and implemented using brushless dc motors. The control system is developed by the fuzzy logic theory, and implemented with a multilayer neural network. First, a neural network-based architecture is introduced for fuzzy logic control. The characteristic rules and their membership functions of fuzzy systems are represented as the processing nodes in the neural network structure. The network structure and the parameter learning are performed simultaneously and online in the fuzzy-neural network system. The structure learning is based on the partition of input space. The parameter learning is based on the supervised gradient decent method, using a delta adaptation law. Using experimental setup, the performance of the proposed control system is evaluated under various operating conditions. Test results are presented and discussed in the report. The proposed learning control system has several advantages, namely, simple structure and learning capability, robustness and high tracking performance and few nodes at hidden layers. In comparison with the PI controller, the proposed fuzzy-neural network system can yield a better dynamic performance with shorter settling time, and without overshoot. Experimental results have shown that the proposed control system is adaptive and robust in responding to a wide range of operating conditions. In summary, the goal of this study is to design and implement-advanced servosystems to actuate control surfaces for flight vehicles, namely, aircraft and helicopters, missiles and interceptors, and mini- and micro-air vehicles.

Rubaai, Ahmed↗
DOE OSTI2018

Deep Learning for Enhancing Fault Tolerant Capabilities of Scientific Workflows

Deep Learning for Enhancing Fault TolerantCapabilities of Scientific WorkflowsAlok Singh, Ilkay AltintasSan Diego Supercomputer Center, UCSDLa Jolla, CA, USA{a1singh, ialtintas}@ucsd.eduMalachi Schram, Nathan TallentPacific Northwestern National LaboratoryRichland, WA, USA{malachi.schram, nathan.tallent}@pnnl.govAbstract-In the history of Computer Science, the act of ‘delegation’ has been the greatest multiplier of society’s problem-solving ability. A scientist working on detecting anomalies in a phenomenon, does not need to re-invent matrix multiplication techniques to solve her problem. Scientific workflows provide ultimate ‘delegation’ mechanism -where a domain scientist can completely forget the specifics of ‘how’ her program will execute on a large cluster in an efficient and cost-effective manner and can instead focus on the mathematical formulation and theoretical robustness of her solution. We present here an approach that directly aims to make the execution of Scientific Workflows more reliable, robust and efficient. We aim that the work presented in this paper will propel the larger effort, from the scientific workflow community, of making scientific workflow execution as simple, efficient and robust as a JOIN operation in a modern database. Specifically, we apply Deep Learning techniques to develop a mechanism that forecasts the final state (success or failure) of a dynamic job in a large-scale particle physics experiment, with minimal data gathering, and as early as possible in job’s life cycle. The key advantage of having a predictive mechanism to identify and anticipate failure-prone jobs is the potential for designing intelligent Fault Tolerance mechanisms to handle anomalous events. We achieve a 14% improvement in computational resources utilization, and an overall classification accuracy of 85% on real tasks executed in a High Energy Physics Computing workflow.To the best of our knowledge, this is the most exhaustive and first of its kind study of neural network architectures in context of a real-dataset profiled from a large-scalescientific workflow.

deep learning↗
DOE OSTI2022

A Systematic Framework for Projecting the Future Cost of Offshore Wind Energy

Offshore wind costs are expected to decline rapidly in the short and medium term future as the industry grows and gains experience in manufacturing, installing, and operating commercial scale projects. Estimating the future costs of offshore wind energy is critical for evaluating the technology's economic performance, how it can fit into a broader clean energy economy, and how R&D investment can be allocated to advance the technology. We present a newly developed approach for forecasting these costs which focuses on an empirically-derived learning rate for capital costs and prescribed cost and performance improvements for operational costs and capacity factor. We establish baseline costs for a series of reference fixed-bottom and floating projects in 2021 and project cost trajectories to 2035, presenting both an average cost trajectory as well as describing the range of potential future costs associated with site-specific cost variations and uncertainty in the estimate of the learning rate. We also conduct sensitivity analyses showing the impact of different global deployments by 2035 and variations in the prescribed operational costs and capacity factors. The results show that fixed-bottom and floating offshore wind capital costs could decrease to around $\$$2,400/kW and $\$$3,300/kW by 2035, respectively, with ranges of $\$$2,100/kW - $\$$2,750/kW for fixed-bottom projects and $\$$2,850/kW - $\$$5,500/kW for floating projects. The levelized cost of energy of fixed-bottom and floating wind projects could decrease to $\$$53.1/MWh and $\$$63.9/MWh by 2035, with ranges of $\$$48.4/MWh - $\$$59.7/MWh for fixed-bottom projects and $\$$46.5/MWh - $\$$99.9/MWh for floating projects. By presenting the uncertainty associated with the forecast we provide a transparent description of the spectrum of potential cost trajectories for offshore wind.

17 WIND ENERGY↗
NASA NTRS1998

Operational Concepts for Uninhabited Tactical Aircraft

This paper describes experiences with five remotely piloted flight research vehicle projects in the developmental flight test phase. These projects include the Pathfinder, Perseus B, Altus, and X-36 aircraft and the Highly Maneuverable Aircraft Technology (HiMAT). Each of these flight projects was flown at the NASA Dryden Flight Research Center. With the exception of the HiMAT, these projects are a part of the Flight Research Base Research and Technology (R&T) Program of the NASA Aeronautics and Space Transportation Technology Enterprise. Particularly with respect to operational interfaces between the ground-based pilot or operator, this paper draws from those experiences, then provides some rationale for extending the lessons learned during developmental flight research to the possible situations involved in the developmental flights proceeding deployed uninhabited tactical aircraft (UTA) operations. Two types of UTA control approaches are considered: autonomous and remotely piloted. In each of these cases, some level of human operator or pilot control blending is recommended. Additionally, "best practices" acquired over years of piloted aircraft experience are drawn from and presented as they apply to operational UTA.

Deets, Dwain A.↗
NASA NTRS1998

Operational Concepts for Uninhabited Tactical Aircraft

This paper describes experiences with five remotely piloted flight research vehicle projects in the developmental flight test phase. These projects include the Pathfinder, Perseus B, Altus, and X-36 aircraft and the Highly Maneuverable Aircraft Technology (HiMAT). Each of these flight projects was flown at the NASA Dryden Flight Research Center. With the exception of the HIMAT, these projects are a part of the Flight Research Base Research and Technology (R&T) Program of the NASA Aeronautics and Space Transportation Technology Enterprise. Particularly with respect to operational interfaces between the ground-based pilot or operator, this paper draws from those experiences, then provides some rationale for extending the lessons learned during developmental flight research to the possible situations involved in the developmental flights proceeding deployed uninhabited tactical aircraft (UTA) operations. Two types of UTA control approaches are considered: autonomous and remotely piloted. In each of these cases, some level of human operator or pilot control blending is recommended. Additionally, "best practices" acquired over years of piloted aircraft experience are drawn from and presented as they apply to operational UTA.

Deets, Dwain A.↗
NASA NTRS2022

Flight Rule Design, Implementation, Verification, and Validation for the Psyche Mission

NASA Jet Propulsion Lab (JPL)’s upcoming mission Psyche will begin its journey to the asteroid (16) Psyche in late 2022 in an effort to better understand its origins and, in turn, better understand our own. Operating the spacecraft safely will require the dedicated efforts of a small team that understands the spacecraft’s operational constraints, as well as a set of powerful spacecraft models designed to catch command errors that can pose risks to mission success. One of the responsibilities of the operations team is to ensure adherence to a set of Flight Rules written by spacecraft and instrument experts that are designed to mitigate these risks. Psyche’s innovations in Flight Rule design principles and advancements in the tools and processes used to implement and check Flight Rules are discussed. A comparison of Psyche’s approach to Flight Rules to other JPL missions will provide lessons learned for future missions that must perform constraint checking during operations. Flight Rule development faces several major challenges. First, flight rule developers must work with Subject Matter Experts (SME) to write the rules in a way that captures the intent of the constraint in a straightforward, enforceable manner. Second, software implementers must correctly interpret flight rules into code so that it meets the original intent of the SME. Finally, a means must be provided for SMEs to validate flight rule implementations without requiring them to understand the underlying software. Innovative processes intended to efficiently close the loop between stakeholders and software developers are described, such as the use of test-driven development to provide stakeholders with easy-to-review implementations. New guidelines for flight rule writing, designed to address these challenges, are described for future missions to adopt and build upon. Psyche Mission System has a variety of new and heritage tools that improve in the Flight Rule validation and checking process. Psyche developed a powerful, new tool called RandSEQ and made significant improvements to Octopusjam, two valuable tools that aid the development of Flight Rule unit tests. Advancements in the models and processes for performing sequence validation with SEQuence GENerator (SEQGEN), the primary, high-heritage tool used for automated flight rule checks on Psyche, are described. The development of new software and the advancements to existing software put Psyche at the forefront of Flight Rule technology. All missions must perform detailed constraint checking, so a comparison of Psyche’s approach to some of these items to the approaches taken by other missions such as Dawn, M2020, and Europa Clipper is done, specifically to examine SME-developer communication, tools used, and development process. Lessons learned from this comparison will be provided.

Weise, Tim↗
NASA NTRS2019

Prediction of Pushback Times and Ramp Taxi Times for Departures at Charlotte Airport

When optimizing the takeoff sequence and schedule for departures at busy airports, it is important to accurately predict the taxi times from gate to runway because those are used to calculate the earliest possible takeoff times. Several airports like Charlotte Douglas International Airport show relatively long taxi times inside the ramp area with large variations, with respect to the travel times in the airport movement area. Also, the pushback process times have not been accurately modeled so far mainly due to the lack of accurate data. The recent deployment of the integrated arrival, departure, and surface traffic management system at Charlotte airport by NASA enables more accurate flight data in the airport surface operations to be obtained. Taking advantage of this system, actual pushback times and ramp taxi times from historical flight data at this airport are analyzed. Based on the analysis, a simple, data-driven prediction model is introduced for estimating pushback times and ramp transit times of individual departure flights. To evaluate the performance of this prediction model, several machine learning techniques are also applied to the same dataset. The prediction results show that the data-driven prediction model is as good as the machine learning algorithms when comparing various prediction performance metrics.

airport surface operations↗
NASA NTRS2016

NASA's Earth Science Flight Program Meets the Challenges of Today and Tomorrow

NASA's Earth science flight program is a dynamic undertaking that consists of a large fleet of operating satellites, an array of satellite and instrument projects in various stages of development, a robust airborne science program, and a massive data archiving and distribution system. Each element of the flight program is complex and present unique challenges. NASA builds upon its successes and learns from its setbacks to manage this evolving portfolio to meet NASA's Earth science objectives. NASA fleet of 16 operating missions provide a wide range of scientific measurements made from dedicated Earth science satellites and from instruments mounted to the International Space Station. For operational missions, the program must address issues such as an aging satellites operating well beyond their prime mission, constellation flying, and collision avoidance with other spacecraft and orbital debris. Projects in development are divided into two broad categories: systematic missions and pathfinders. The Earth Systematic Missions (ESM) include a broad range of multi-disciplinary Earth-observing research satellite missions aimed at understanding the Earth system and its response to natural and human-induced forces and changes. Understanding these forces will help determine how to predict future changes, and how to mitigate or adapt to these changes. The Earth System Science Pathfinder (ESSP) program provides frequent, regular, competitively selected Earth science research opportunities that accommodate new and emerging scientific priorities and measurement capabilities. This results in a series of relatively low-cost, small-sized investigations and missions. Principal investigators whose scientific objectives support a variety of studies lead these missions, including studies of the atmosphere, oceans, land surface, polar ice regions, or solid Earth. This portfolio of missions and investigations provides opportunity for investment in innovative Earth science that enhances NASA's capability for better understanding the current state of the Earth system. ESM and ESSP projects often involve partnerships with other US agencies and/or international organizations. This adds to the complexity of mission development, but allows for a greater scientific return on NASA's investments. The Earth Science Airborne Science Program provides manned and unmanned aircraft systems that further science and advance the use of satellite data. NASA uses these assets worldwide in campaigns to investigate extreme weather events, observe Earth system processes, obtain data for Earth science modeling activities, and calibrate instruments flying aboard Earth science spacecraft. The Airborne Science Program has six dedicated aircraft and access to many other platforms. The Earth Science Multi-Mission Operations program acquires, preserves, and distributes observational data from operating spacecraft to support Earth Science research focus areas. The Earth Observing System Data and Information System (EOSDIS), which has been in operations since 1994, primarily accomplishes this. EOSDIS acquires, processes, archives, and distributes Earth Science data and information products. The archiving of NASA Earth Science information happens at eight Distributed Active Archive Centers (DAACs) and four disciplinary data centers located across the United States. The DAACs specialize by topic area, and make their data available to researchers around the world. The DAACs currently house over 9 petabytes of data, growing at a rate of 6.4 terabytes per day. NASA's current Earth Science portfolio is responsive to the National Research Council (NRC) 2007 Earth Science Decadal Survey and well as the 2010 NASA Response to President Obama's Climate Plan. As the program evolves into the future it will leverage the lessons learned from the current missions in operations and development, and plan for adjustments to future objectives in response to the anticipated 2017 NRC Decadal Survey.

Flight Program↗
DOE OSTI2024

Deep Reinforcement Learning for Microgrid Cost Optimization Considering Load Flexibility

This paper proposes a novel Soft-Actor-Critic (SAC) based Deep Reinforcement Learning (DRL) method for optimizing the cost of microgrid operation by leveraging load flexibility. The proposed SAC-DRL method is designed to coordinate the control of distributed energy resources (DERs) and flexible load, addressing practical energy billing formation by power distribution utilities. Key contributions include an innovative reward function to mitigate sparse reward challenges and a mixed control strategy for discrete and continuous variables, ensuring radial network topology and minimizing power loss. We evaluate the proposed method on the model of a real microgrid located in Southern California, U.S.. The SAC-DRL model is tested to demonstrate its efficacy in reducing grid dependence, optimizing resource use, and minimizing costs. The results highlight the potential of DRL in modern energy systems, offering a sustainable and economically efficient solution for energy management in microgrids.

deep reinforcement learning↗
DOE OSTI2021

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

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

30 DIRECT ENERGY CONVERSION↗
NASA NTRS2005

Operational behavioral health and performance resources for international space station crews and families

The Behavioral Health and Performance Section (BHP) at NASA Johnson Space Center provides direct and indirect psychological services to the International Space Station (ISS) astronauts and their families. Beginning with the NASA-Mir Program, services available to the crews and families have gradually expanded as experience is gained in long-duration flight. Enhancements to the overall BHP program have been shaped by crewmembers' personal preferences, family requests, specific events during the missions, programmatic requirements, and other lessons learned. The BHP program focuses its work on four areas: operational psychology, behavioral medicine, human-to-system interface, and sleep and circadian. Within these areas of focus are psychological and psychiatric screening for astronaut selection as well as many resources that are available to the crewmembers, families, and other groups such as crew surgeon and various levels of management within NASA. Services include: preflight, in flight, and postflight preparation; training and support; resources from a Family Support Office; in-flight monitoring; clinical care for astronauts and their families; and expertise in the workload and work/rest scheduling of crews on the ISS. Each of the four operational areas is summarized, as are future directions for the BHP program.

Space Flight↗
NASA NTRS2022

Revisiting the Ground Magnetic Field Perturbations Challenge: A Machine Learning Perspective

Forecasting ground magnetic field perturbations has been a long-standing goal of the space weather community. The availability of ground magnetic field data and its potential to be used in geomagnetically induced current studies, such as risk assessment, have resulted in several forecasting efforts over the past few decades. One particular community effort was the Geospace Environment Modeling (GEM) challenge of ground magnetic field perturbations that evaluated the predictive capacity of several empirical and first principles models at both mid- and high-latitudes in order to choose an operative model. In this work, we use three different deep learning models-a feed-forward neural network, a long short-term memory recurrent network and a convolutional neural network-to forecast the horizontal component of the ground magnetic field rate of change (dB H /dt) over 6 different ground magnetometer stations and to compare as directly as possible with the original GEM challenge. We find that, in general, the models are able to perform at similar levels to those obtained in the original challenge, although the performance depends heavily on the particular storm being evaluated. We then discuss the limitations of such a comparison on the basis that the original challenge was not designed with machine learning algorithms in mind.

Victor A. Pinto↗
DOE OSTI2020

Attack-resilient algorithms and testbed federation for wide-area protection and control in smart grid

Today’s energy infrastructure is undergoing a massive transformation across all generation, transmission, and distribution systems to provide reliability, efficiency, and sustainability to the power system network. During the past 10 years, there has been a remarkable enhancement in developing close-loop wide-area protection and control (WAPAC) applications, such as automatic generation control (AGC), wide-area protection scheme (WAPS), synchrophasor-based wide-area voltage control system (WAVCS), etc., in the energy management system (EMS) to provide an adequate situational awareness to control center operators during physical disturbances and natural events. The rapid digitalization, unencrypted communication, and numerous data-sharing devices have increased digital access points that have exposed the grid network to countless cyberattacks pertinent to the WAPAC cybersecurity. Given the amount of complex grid network coupled with legacy infrastructures, it is untrustworthy to completely rely on information technology (IT)-based conventional security measures to defend against evolving cyberattacks. Hence, there is a compelling need to go beyond the traditional paradigm of ‘security by obscurity’ and ‘bolt-on’ security measures, and develop a suite of innovative security solutions at multiple layers to transform the current “fault-resilient” grid into “fault and attack-resilient” grid of the future. This dissertation specifically focuses on developing application-specific attack-resilient solutions for critical WAPAC applications in the smart grid. The first part of the study, Attack-Resilient Wide-Area Protection Scheme, presents a cyber-physical attack-resilient system (CPARS) for WAPS, where two anomaly detection methods are proposed to detect multi-level cyberattacks followed by the rules-based intrusion mitigation system (RIMS) to initiate effective mitigation strategies, tailored to these attacks, to restore the normal grid operation after disturbances. The first anomaly detection component, Machine learning-based Anomaly Detection for Centralized WAPS, discusses a novel architecture and methodology for developing a machine-learning-based anomaly detection system (ADS) for the centralized wide-area protection system (CWAPS) by applying the variational mode decomposition (VMD) technique and decision tree (DT) algorithms using cyber logs and synchrophasor measurements. The second component, Multi-Agent-based Anomaly Detection for Decentralized WAPS, presents a two-level hierarchical multi-agent-based decentralized WAPS against the system-aware stealthy cyber-attacks. Finally, for the RIMS, the state transition diagram-based mitigation measures are defined based on the grid operation states and experimental testing and evaluation are presented in a cyber-physical testbed environment. The second part of the study, Attack-Resilient Wide-Area Voltage Control System, addresses the existing vulnerabilities in the wide-area voltage control system (WAVCS), presents the methodology for performing an impact analysis during data integrity attacks, and proposes a data-driven attackresilient system (DARS) by incorporating machine learning-based ADS and rules-based anomaly mitigation system (RAMS) to detect these attacks and provide necessary mitigation actions to maintain the transient voltage stability after disturbances. Finally, the third part, Networked Federation Testbed (NEFTSec) for Smart Grid Cybersecurity, focuses on developing an interconnected federated testbed through a common network like the internet, describes its conceptual architectures, and also presents its design components for experimental testing, validation, and evaluation. Further, it presents the co-simulation interface algorithm (CIA) to facilitate the geographically-dispersed real-time distributed simulation (GD-RTDS) during the cyber-physical federation. In general, this dissertation presents innovative attack-resilient solutions for WAPAC applications that are experimentally tested and evaluated using the PowerCyber testbed and NEFTSec platform to accelerate the transition of state-of-the-art research works to the real-field deployments.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Graph Analytics for CEBAF Operations

We report on the progress achieved during a 2-year Laboratory Directed Research and Development (LDRD) project titled “Graph Analytics for CEBAF Operations”. The objective of this project is to leverage deep learning on graph representations of CEBAF’s injector beamline in order to create a tool for improving the efficiency of beam tuning tasks. Specifically, we use graphs to represent the injector beamline at any arbitrary date and time and invoke a graph neural network (GNN) to extract a low-dimensional, informative representation that can be visualized in two-dimensions. By analyzing years of operational data from the CEBAF archiver, good and bad regions of parameter space can be identified. The goal is to exercise this framework as a real-time tool to aid beam tuning, which represents the dominant source of machine downtime.

43 PARTICLE ACCELERATORS↗
DOE OSTI2021

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

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

30 DIRECT ENERGY CONVERSION↗