Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “agent based”

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 1,621 records · Page 90
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Composition‐Conditioning Agent for Doped Spiro‐OMeTAD to Realize Highly Efficient and Stable Perovskite Solar Cells

Abstract The doped Spiro‐OMeTAD hole transport layer (HTL) formed using the lithium bis(trifluoromethane) sulfonimide salt and 4‐tert‐butylpyridine with phenethylammonium iodide surface treatment on a perovskite film has continuously dominated the record power conversion efficiencies (PCEs) of perovskite solar cells (pero‐SCs). However, unstable HTL compositions and iodide salts can cause severe device degradation. In this study, an HTL composition‐conditioning agent (CCA), Spiro‐BD‐2OEG, is designed, which contains a Spiro‐OMeTAD‐like backbone, functional pyridine units, and oligo (ethylene glycol) chains. This finely designed CCA presents good miscibility with Spiro‐OMeTAD and its dopants and acts as a conditioning agent through weak bond interactions. As a result, the CCA‐regulated HTL shows a pinhole‐free and smooth morphology with enhanced Spiro‐OMeTAD ordering and improves dopant stability. In addition, the gradient‐distributed CCA in the HTL can narrow the energy level offset with the valence band of the perovskite. The resultant pero‐SCs exhibit an excellent PCE of 24.19% without any interface treatment and weak size dependence. A remarkable PCE of 22.63% is obtained even for a 1.004‐cm 2 device. Importantly, the strategy shows good universality and significantly promotes the long‐term stability of the pero‐SCs based on the classical doped Spiro‐OMeTAD.

Chemistry↗
DOE OSTI2022

Performance and Economic Analysis of Organosolv Softwood and Herbaceous Lignins to Activated Carbons as Electrode Materials in Supercapacitors

In this work, yellow pine (YP, softwood) and switchgrass (SG, grass) lignins were extracted as a coproduct of an organosolv γ-valerolactone (GVL) biorefinery that also produces biofuels and furfural. The extracted lignins were converted to carbon precursors for synthesizing porous activated carbon electrodes for high energy-density supercapacitors. This research details the impact of lignin composition on the derived porous structures and electrochemical properties of activated carbons. Lignin precursors with various syringyl (S) to guaiacyl (G) contents were characterized using 31 P nuclear magnetic resonance (NMR) and two-dimensional 1 H‒ 13 C NMR. A two-step activation process, using steam and carbon dioxide as the activating agents, enabled the formation of porous carbons structures with high surface area. The capacitive behavior of supercapacitors was systematically characterized by cyclic voltammetry, charge-discharge cycling, and electrochemical impedance spectroscopy. The specific capacitance of YP and SG capacitors reached 367 and 221 F g −1 , respectively. Both types of capacitors demonstrated remarkably stable capacitance (capacitance retention >90%) along with excellent Coulombic efficiency (>99%) over 10,000 cycles. Compared to SG electrode, the better electrochemical performance achieved with YP electrodes was mainly due to shorter diffusion path, improved ionic mobility, and increased active surface area. The inexpensive lignin-based porous electrodes synthesized in this work can be used for various electrochemical devices for improved performance, decreased cost, and enhanced durability. This work also demonstrates that the selection of feedstock and appropriate processing conditions can tailor the structure of carbon composites for targeted applications. Techno-economic analysis indicates that YP and SG activated carbons can be produced at a minimum selling price of $8,493 and $6,670 per ton, respectively, which is competitive with the commercially available supercapacitor-grade activated carbons.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1998

A Cognitive-System Model for En Route Air Traffic Management

NASA Ames Research Center has been engaged in the development of advanced air traffic management technologies whose basic form is cognitive aiding systems for air traffic controller and flight deck operations. In the design and evaluation of such systems the dynamic interaction between the airborne aiding system and the ground-based aiding systems forms a critical coupling for control. The human operator is an integral control element in the system and the optimal integration of human decision and performance parameters with those of the automation aiding systems offers a significant challenge to cognitive engineering. This paper presents a study in full mission simulation and the development of a predictive computational model of human performance. We have found that this combination of methodologies provide a powerful design-aiding process. We have extended the computational model Man Machine Integrated Design and Analysis System (N13DAS) to include representation of multiple cognitive agents (both human operators and intelligent aiding systems), operating aircraft airline operations centers and air traffic control centers in the evolving airspace. The demands of this application require the representation of many intelligent agents sharing world-models, and coordinating action/intention with cooperative scheduling of goals and actions in a potentially unpredictable world of operations. The operator's activity structures have been developed to include prioritization and interruption of multiple parallel activities among multiple operators, to provide for anticipation (knowledge of the intention and action of remote operators), and to respond to failures of the system and other operators in the system in situation-specific paradigms. We have exercised this model in a multi-air traffic sector scenario with potential conflict among aircraft at and across sector boundaries. We have modeled the control situation as a multiple closed loop system. The inner and outer loop alerting structure of air traffic management has many implications that need to be investigated to assure adequate design. First, there are control and stability factors implicit in the design. As the inner loop response time approaches that of the outer loop, system stability may be compromised in that controllers may be solving a problem the nature of which has already been changed by pilot action. Second, information exchange and information presentation for both air and ground must be designed to complement as opposed to compete with each other. Third, the level of individual and shared awareness in trajectory modification and flight conformance needs to be defined. Fourth, the level of required awareness and performance impact of mixed fleet operations and failed-mode recovery must be explored.

Corker, Kevin M.↗
NASA NTRS2002

The Intelligent Data Understanding Element of NASA's Intelligent Systems Program

Within the NASA Intelligent Systems Program, the Intelligent Data Understanding (IDU) element develops techniques for transforming data into scientific understanding. Automating such tools is critical for space science, space-based earth science, and planetary exploration with onboard scientific data analysis. Intelligent data understanding (IDU) is about extracting meaning from large, diverse science and engineering databases, via autonomous techniques that transform very large datasets into understanding. The earth science community in particular needs new tools for analyzing multi-formatted and geographically distributed datasets and for identifying cause-effect relationships in the complex data. Research within the IDU program element seeks to automate data analysis tasks so that humans can focus on creative hypothesis generation and knowledge synthesis. It may also enable NASA space missions in which autonomous agents must generate knowledge and take actions, and missions where limited bandwidth permits transmission of only the most interesting scientific observations, summaries, and conclusions. Twenty-seven research projects are-currently funded.

Coughlan, Joseph C.↗
DOE OSTI2023

Sensor Development for Liquid Water Detection in Dry Storage Casks: FY 2023 Status

Modeling efforts were undertaken in fiscal year (FY) 2023 to evaluate the feasibility and capability for sensing of liquid water inside canister-based dry cask storage systems (DCSSs). The focus was on the development of full-scale finite element models (FEMs) of ultrasound propagation through DCSS canister components. For these initial investigations, simulation of the baseplate component was targeted, envisioning water collection at the bottom of a vertically oriented canister. The effects of internal canister components (i.e., fuel basket, fuel assemblies) on ultrasound propagation in the baseplate component were evaluated in these efforts. The environment inside a DCSS confinement is intended to be inert and free of water to prevent potential corrosion of used fuel cladding or other internal hardware. Spent fuel assemblies are dried, after storage in water pools, to make sure water has been removed from assembly cavities. However, there is some uncertainty about the amount of residual water potentially left behind in a DCSS after drying processes, because water can become trapped in cavities or other small crevices in the surfaces formed by the fuel cladding, fuel assemblies, and other internal hardware components. Considering the complex space- and time-dependent temperature profiles in DCSSs, water may be in a liquid or gas phase depending on its location in the cask and how long the cask has been in storage. Evacuating most water and oxidizing agents contained within a canister is recommended by NUREG-1536 (NRC 2010), which covers DCSSs. As summarized by Salazar et al. (2020), existing guidance typically relies on achievable vacuum pumping pressures sustained over a hold period as a signal of dryness and water removal. However, time series data about water removal from full-scale commercial drying procedures are lacking (Hanson and Alsaed 2019). A review of drying specifications from several vendors led to the conclusion that if the specifications are followed correctly, the residual moisture in DCSSs should present an insignificant risk of cladding degradation (Knoll and Gilbert 1987). A more recent analysis concluded that much larger quantities of residual water could remain in DCSSs, but the amount would still not be expected to lead to significant corrosion of fuel cladding or other internal components (Jung et al. 2013). Industry drying procedures are mostly prescriptive in nature, and operational issues arising during the process could result in incomplete drying. A summary of operational issues and potential negative effects of residual water is provided by Salazar et al. (2019). Experimental efforts are ongoing to validate the extent of water removal in a DCSS based on drying procedures used at nuclear power plants through well-designed investigations of drying process efficacy and water retention (Durbin et al. 2021; Pulido et al. 2022a, Pulido et al. 2022b). This has been approached by simulating limited portions of a DCSS internal volume and fuel assemblies. So far, these efforts have focused on the effects of potential water trapping in the dashpot region of control rod guide tubes but are expected to be expanded to include the effects of other internal hardware features and failed fuel rod cladding. A method for detection and measurement of liquid water (Meyer et al. 2022; Meyer et al. 2021), in tandem with the drying information gained through experimental investigations, provides comprehensive bases for understanding the internal conditions of DCSSs and provides technical information to support licensing decisions.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
NASA NTRS1997

Business Plan: The Virginia Space Flight Center

The Virginia Commercial Space Flight Authority (VCSFA) was established on July 1, 1995 and codified at Sections 9-266.1 et seq., Code of Virginia. It is governed by an eleven person Board of Directors representing industry, state and local government and academia. VCSFA has designated the Center for Commercial Space Infrastructure as its Executive Directorate and Operating Agent. This Business Plan has been developed to provide information to prospective customers, prospective investors, state and federal government agencies, the VCSFA Board and other interested parties regarding development and operation of the Virginia Space Flight Center (VSFC) at Wallops Island. The VSFC is an initiative sponsored by VCSFA to achieve its stated objectives in the areas of economic development and education. Further, development of the VSFC is in keeping with the state's economic goals set forth in Opportunity Virginia, the strategic plan for jobs and prosperity, which are to: (1) Strengthen the rapidly growing aerospace industry in space based services including launch services, remote sensing, satellite manufacturing and telecommunications; and (2) Capitalize on intellectual and technical resources throughout the state and become a leader in the development of advanced technology businesses.

Reed, Billie M.↗
NASA NTRS2019

Incorporation of Functionalized Polyhedral Oligomeric Silsesquioxane Nanomaterials as Reinforcing Agents for Impact Ice Mitigating Coatings

In-flight, aircraft are exposed to a wide range of environments. One commonly exposed environment are clouds containing super-cooled water droplets. These water drop- lets exist in a metastable state below the freezing point of water, in the range of 0 to -20°C. As the vehicle impacts the droplets, latent heat is released and within milliseconds the droplets convert to ice. This process is referred to as impact icing or in-flight icing.1 Impact icing is a major concern for aircraft since it can lead to degraded aerodynamic performance and, if left un- treated, can lead to loss of the vehicle. Active approaches (i.e., pneumatic boots, heated air ducts) typically utilized in mitigating in-flight ice accretion significantly increases vehicle weight and cannot be applied to all aircraft.1-3 A passive approach based on coatings is desired, but durability issues are a concern, especially on the wing leading edge.3 Nanomaterials have been shown to afford significant improvement in coating and composite physical properties at low loading levels.4 In this study, Polyhedral Oligomeric Silsesquioxane (POSS) nanomaterials have been shown to increase coating durability. Also, with wide variety of functionalities present on the arm structure, POSS nanomaterials have been shown to readily alter coating surface chemistry to mitigate impact ice adhesion from -16 to -8°C in a simulated in-flight icing environment.

Beck, Devon M.↗
DOE OSTI2023

Efficient Reinforcement Learning for Real-Time Hardware-Based Energy System Experiments: Preprint

In the context of urgent climate challenges and the pressing need for rapid technology development, Reinforcement Learning (RL) stands as a compelling data-driven method for controlling real-world physical systems. However, RL implementation often entails time-consuming and computationally intensive data collection and training processes, rendering them inefficient for real-time applications that lack non-real-time models. To address these limitations, real-time emulation techniques have emerged as valuable tools for the lab-scale rapid prototyping of intricate energy systems. While emulated systems offer a bridge between simulation and reality, they too face constraints, hindering comprehensive characterization, testing, and development. In this research, we construct a surrogate model using limited data from simulated systems, enabling an efficient and effective training process for a Double Deep Q-Network (DDQN) agent for future deployment. Our approach is illustrated through a hydropower application, demonstrating the practical impact of our approach on climate-related technology development.

deep Q-learning↗
DOE OSTI2020

Domesticating a food spoilage yeast into an organic acid–tolerant metabolic engineering host: Lactic acid production by engineered Zygosaccharomyces bailii

Lactic acid represents an important class of commodity chemicals, which can be produced by microbial cell factories. However, due to the toxicity of lactic acid at lower pH, microbial production requires the usage of neutralizing agents to maintain neutral pH. Zygosaccharomyces bailii, a food spoilage yeast, can grow under the presence of organic acids used as food preservatives. This unique trait of the yeast might be useful for producing lactic acid. With the goal of domesticating the organic acid!tolerant yeast as a metabolic engineering host, seven Z. bailii strains were screened in a minimal medium with 10 g/L of acetic, or 60 g/L of lactic acid at pH 3. The Z. bailii NRRL Y7239 strain was selected as the most robust strain to be en- gineered for lactic acid production. By applying a PAN!ARS!based CRISPR!Cas9 system consisting of a transfer RNA promoter and NAT selection, we demonstrated the targeted deletion of ADE2 and site!specific integration of Rhizopus oryzae ldhA coding for lactate dehydrogenase into the PDC1 locus. The resulting pdc1::ldhA strain produced 35 g/L of lactic acid without ethanol production. Here, this study de- monstrates the feasibility of the CRISPR!Cas9 system in Z. bailii, which can be applied for a fundamental study of the species.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Conductive Organometallic Polymers from Soluble Superatom Ions

Superatomic crystals comprising ligand-capped, metal chalcogenide clusters and fullerenes are modular materials that exhibit enhanced electronic, magnetic, and thermal conductivity properties. We find that neutral, M 4 S 4 (M = Fe, Co) clusters stabilized with N -heterocyclic carbenes (NHCs) can transfer charge to C 60 fullerene to form binary superatomic crystals. Notably, these compounds are soluble in various organic solvents, allowing their properties to be investigated in solution, unlike traditional fullerene-based superatomic crystals. The ion pairs can be further assembled into organometallic polymers using Janus-bis(NHCs) to cross-link the oxidized M 4 S 4 units. We show that the superatomic polymers are more conductive than both the precursor superatomic crystals and the polymers containing only neutral M 4 S 4 clusters. Similar conductivity values can be obtained when neutral M 4 S 4 –NHC polymers are doped with solutions of C 60 fullerene. These findings demonstrate that next generation superatomic materials can be prepared via the combination of charge transfer and polymerization with appropriate cross-linking agents.

carbon nanomaterials↗
DOE OSTI2024

Heteroatom-Modified and Compacted Zeolite-Templated Carbons for Gas Storage

Methane, the primary component of natural gas, is an energy-rich fuel by weight but exhibits lower energy density by volume than diesel and gasoline; efforts to increase its “volumetric” energy density have been focused on the design and synthesis of a porous scaffold that can bind the methane favorably. Novel approaches typically invoke the highly designable class of materials known as metal-organic frameworks (MOFs) owing to the relative difficulty of controlling carbon structure and composition precisely, without the use of a metal-based node. In this project, we instead explore the concept of carbon templating to achieve a porous scaffold with atomistically thin walls separated at a distance of ~1.2 nm, ideal for the formation of precisely two layers of methane per pore. This bottom-up templating approach permits the use of any small molecular precursor to build up the carbon scaffold, begging the question: which other elements (besides carbon) can be used to tune the surface for optimal methane binding without compromising the structure of the overall material? Boron and nitrogen were both explored and nitrogen was identified as the ideal agent for increasing methane binding to the optimal strength for use at near room temperature. Densification of the resulting materials by mechanical compaction was then explored to produce free-standing pellets with the highest methane “delivery” capacities of any known material. Here we define the amount “delivered” as the amount stored minus the amount remaining in the tank at 73 psi, the lowest useful pressure for vehicular transportation applications. The key features of the resulting methane storage system are rapid discharging and refueling speeds, extremely long cycle life (infinite cycle life for high-purity methane), a near-constant heat of refueling (making the heat generated upon refueling easier to manage), and a low maximum pressure of operation (<1500 psi), enabling the use of all-metal Type I cylinders and significantly reducing the costs of conversion of existing vehicles to cleaner burning, abundantly available natural gas). The technical effectiveness of this approach was further improved by exploring several cost-reduction strategies and investigating the reliability of such storage systems in the presence of “real” natural gas mixtures which contain a mixture of gaseous species.

03 NATURAL GAS↗
NASA NTRS2018

SAM 3.0 Workshop Alliance SV Lab

Nissan is building a Seamless Autonomy Mobility (SAM) which is a cloud-based system of artificial intelligence that manages a fleet of autonomous vehicles. It pairs artificial intelligence with human intelligence in order to create a seamless solution that allows autonomous drive to be viable and marketable. They have created a Mobility Manager function which is a laborer who executed human-in-the-loop functions within SAM system. Nissan invited the HAT Lab to discuss NASA's lessons learned on remote operator support and to participate in a workshop discussion of one of their use cases. This presentation covers some of the HAT Lab's work on a human-autonomy teaming Agent and previous Reduced Crew Operations ground station development.

Human-Autonomy Teaming↗
NASA NTRSDate not supplied

Synthetic Data Generation for 3D Mesh Prediction and Spatial Reasoning During Multi-Agent Robotic Missions

In-space assembly operations require accurate reasoning over the pose, location, and structural organization of both the autonomous agents and assembly materials. In a full six-degree-of-freedom space, an accurate understanding of the full three-dimensional structure of the object of interest greatly enriches information for pose estimation and collision planning. Current methods of predicting pose estimation require a priori understanding of the shape of the object. Additionally, visual information in the space environment is impacted by variations in contrast and illumination. Using synthetic data allows us to rapidly generate large datasets with in varying environments and lighting conditions.This work details the generation of synthetic data used to explore the use of a region-based convolutional neural networks to detect objects of interest and predict a voxel-based three-dimensional mesh in order to understand their full three-dimensional shape. This mesh provides useful spatial information during in-space assembly operations without requiring either the complexity of maintaining models over the progress of building an object or observations from multiple angles. The generated meshes are then compared to that of ground truth in order to measure its performance.

synthetic data↗
DOE OSTI2021

Silica-modifying chemical admixtures for directed zeolitization of metakaolin-based alkali-activated materials

The effect of using trimethyladamantyl-ammonium hydroxide (TMAAOH) as a silica-modifying admixture to induce early-age mineralization during alkali-activation of metakaolin was investigated and reported herein. In all material formulations, the use of TMAAOH induced early-age mineralization, increased mixture stiffening in the fresh state, and lowered total heat of reaction. In activating solutions with silica moduli above unity (Ms >1.0), mineralogy results demonstrate that TMAAOH induces the nucleation and growth of metastable zeolitic phases, which correlated with increased permeability and increased plastic shrinkage of the paste. When TMAAOH was added to activating solutions with silica moduli near unity (Ms ~1.0), the controlled formation of crystalline, silica-rich faujasite was observed, which correlated with reduced permeability and lower plastic shrinkage. Together, these results demonstrate for the first time that silica-templating agents such as TMAAOH can be exploited in the design of new chemical admixtures that directly influence the dynamics of zeolitization in alkali-activated materials.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Linking Threat Agents to Targeted Organizations: A Pipeline for Enhanced Cybersecurity Risk Metrics

In this study, we present a methodology leveraging Large Language Models (LLMs) to transform Cybersecurity Threat Intelligence (CTI) narratives into actionable insights for individual organizations. Our approach automates the extraction of machine-readable adversary SKRAM (Skills, Knowledge, Resources, Authorities, and Motivation) attributes from open-source reports, extending LLM utility beyond typical interactions. This innovation enables precise, automated assessments of cybersecurity risks posed by various adversaries. Using a chain-of-thought and multi-shot prompting strategy, our methodology advances the automation of cybersecurity feature extraction for new machine-learning models that predict the risk of adversary targeting. This approach is refined using a substantial dataset of over 150 analyst-validated threat reports and synthetic organizational data from 900 companies. Here, by bootstrapping the training data with a rule-based heuristic over synthetic data, we have developed a high-accuracy machine-learning model that allows entities to dynamically prioritize threats and defensive actions.

Cyber Threat Intelligence↗
DOE OSTI2024

Design, Synthesis, and Evaluation of Noble Metal Nanoparticles and In Situ-Decorated Carbon-Supported Nanoparticle Electrocatalysts Using Hypergolic Reactions

Here, we report the first synthesis of metal nanoparticles and supported metal nanoparticles on carbon by using hypergolic reactions. Specifically, we report the synthesis of noble metal nanoparticles (Pt, Ag, and Au) using sodium hydride (NaH) as both an ignition trigger and a reducing agent for the corresponding metal salt precursors. In addition, we report the one-step, in situ synthesis of Pt nanoparticles supported on carbon by adding sucrose as the carbon source. The hypergolically synthesized nanoparticles display elliptical morphology and are more crystalline compared with those conventionally synthesized in solution using sodium borohydride (NaBH 4 ). When tested as electrocatalysts, the hypergolic Pt nanoparticles exhibit more than 2 times higher specific electrochemical active surface area (ECSA) and a higher half-wave potential (E 1/2 ) of 0.94 V vs the reversible hydrogen electrode (RHE) compared to the conventionally synthesized ones. In addition, the electrocatalyst based on the in situ synthesized carbon that was decorated with the Pt nanoparticles synthesized hypergolically outperforms an analogous, state of the art, commercial PtC system. For example, the former shows an attractive E 1/2 (0.94 V) compared with 0.9 V for the commercial PtC. Accelerated durability tests (ADT) in an alkaline environment add another advantage. After 10 000 cycles, the hypergolically synthesized system shows a smaller reduction of E 1/2 and less degradation compared to the commercial PtC (10 mV compared to ∼30 mV). The work described here represents the first reported synthesis using hypergolic reactions of metal nanoparticles as well as supported metal nanoparticles. The properties of the resulting electrocatalysts demonstrate the versatility and promise of the new approach in materials synthesis and open new avenues for further investigation as electrocatalysts.

Chalmpes, Nikolaos↗
DOE OSTI2022

Diffuse Radiation Forcing Constraints on Gross Primary Productivity and Global Terrestrial Evapotranspiration

Abstract The diffuse radiation fertilization effect—the increase in plant productivity in the presence of higher diffuse radiation ( K ↓,d )—is an important yet understudied aspect of atmosphere‐biosphere interactions and can modify the terrestrial carbon, energy, and water budgets. The K ↓,d fertilization effect links the carbon cycle with clouds and aerosols, all of which are large sources of uncertainties for our current understanding of the Earth system and for future climate projections. Here we establish to what extent observational and modeling uncertainty in sunlight's diffuse fraction ( k d ) affects simulated gross primary productivity (GPP) and terrestrial evapotranspiration ( λE ). We find only 48 eddy covariance sites with simultaneous sufficient measurements of K ↓,d with none in the tropical climate zone, making it difficult to constrain this mechanism globally using observations. Using a land modeling framework based on the latest version of the Community Land Model, we find that global GPP ranges from 114 Pg C year −1 when using k d forcing from the Modern‐Era Retrospective analysis for Research and Applications, version 2 reanalysis to a ∼7% higher value of 122 Pg C year −1 when using the Clouds and the Earth's Radiant Energy System satellite product, with especially strong differences apparent over the tropical region (mean increase ∼9%). The differences in λE , although smaller (−0.4%) due to competing changes in shaded and sunlit leaf transpiration, can be greater than regional impacts of individual forcing agents like aerosols. Our results demonstrate the importance of comprehensively and systematically validating the simulated k d by atmosphere modules as well as the response to differences in k d within land modules across Earth System Models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Learning Sequential Distribution System Restoration via Graph-Reinforcement Learning

We report a distribution service restoration algorithm as a fundamental resilient paradigm for system operators provides an optimally coordinated, resilient solution to enhance the restoration performance. The restoration problem is formulated to coordinate distribution generators and controllable switches optimally. A model-based control scheme is usually designed to solve this problem, relying on a precise model and resulting in low scalability. To tackle these limitations, this work proposes a graph-reinforcement learning framework for the restoration problem. We link the power system topology with a graph convolutional network, which captures the complex mechanism of network restoration in power networks and understands the mutual interactions among controllable devices. Latent features over graphical power networks produced by graph convolutional layers are exploited to learn the control policy for network restoration using deep reinforcement learning. The solution scalability is guaranteed by modeling distributed generators as agents in a multi-agent environment and a proper pre-training paradigm. Comparative studies on IEEE 123-node and 8500-node test systems demonstrate the performance of the proposed solution.

24 POWER TRANSMISSION AND DISTRIBUTION↗