Search NASA⌕ Search

SEARCH · Search NASA

Results for “agentic systems”

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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Reinforcement Learning for Volt- Var Control: A Novel Two-stage Progressive Training Strategy

This paper develops a reinforcement learning (RL) approach to solve a cooperative, multi-agent Volt-Var Control (VVC) problem for high solar penetration distribution systems. The ingenuity of our RL method lies in a novel two-stage progressive training strategy that can effectively improve training speed and convergence of the machine learning algorithm. In Stage 1 (individual training), while holding all the other agents inactive, we separately train each agent to obtain its own optimal VVC actions in the action space: fconsume, generate, do-nothingg. In Stage 2 (cooperative training), all agents are trained again coordinatively to share VVC responsibility. Rewards and costs in our RL scheme include (i) a system-level reward (for taking an action), (ii) an agent-level reward (for doing-nothing), and (iii) an agent-level action cost function. This new framework allows rewards to be dynamically allocated to each agent based on their contribution while accounting for the trade-off between control effectiveness and action cost. The proposed methodology is tested and validated in a modified IEEE 123-bus system using realistic PV and load profiles. Simulation results confirm that the proposed approach is robust and computationally efficient; and it achieves desirable volt-var control performance under a wide range of operation conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Towards philosophical reasoning with agentic LLMs: Socratic method for scientific assistance

As large language models (LLMs) become central tools in science, improving their reasoning capabilities is critical for meaningful and trustworthy applications. We introduce a Socratic agent for scientific reasoning, implemented through a structured system prompt that guides LLMs via classical principles of inquiry. Unlike typical prompt engineering or retrieval-based methods, our approach leverages definition, analogy, hypothesis elimination, and other Socratic techniques to generate more coherent, critical, and domain-aware responses. We evaluate the agent across diverse scientific domains and benchmark it on the abstraction and reasoning corpus challenge dataset, achieving 97.15% under a fixed prompting protocol and without fine-tuning or external tools. Expert evaluation shows improved reasoning depth, clarity, and adaptability over conventional LLM outputs, suggesting that structured prompting rooted in philosophical reasoning can improve the scientific utility of language models.

LLM reasoning↗
DOE OSTI2025

Reinforcement Learning‐Based Adaptation of Grid Following Inverter's Internal Controller to Networked Microgrids' Strengths

The varying topological configurations, generator commitments and dispatches, and dynamic load demand lead to changing system's strengths during the operations of networked microgrids. When the system's strengths significantly change, the fixed control gains at large devices may result in unsatisfactory system performance; this necessitates the tuning of the control gains at large devices to adapt to the changing system's strengths. In this paper, observer-based reinforcement learning (RL) is utilised to automatically tune the proportional-integral (PI) gains of phase lock loop (PLL) controller of grid-following (GFL) inverters to adapt to the changing strengths of microgrids and networked microgrids. The RL agent in this framework augments an observer predicting system's strengths, from which the RL control policy will adjust accordingly to tune the PLL controller's gains towards the system's strengths. Also, to enhance the control performance, the recently introduced Barrier function-based RL framework is leveraged for the design of reward function to prevent the high frequency nadir. An operational 26 kV electric distribution system, which is modelled as networked microgrids, is used to illustrate the need and effectiveness of the proposed RL-tuned control.

frequency response↗
DOE OSTI2025

Reinforcement Learning for In-Spill Optimization of the Mu2e Resonant Extraction: Compensating Non-Stationarity

We present design considerations and challenges for the fast machine learning component of a third-order resonant beam extraction regulation system being commissioned to deliver steady beam rates to the mu2e experiment at Fermilab. Dedicated quadrupoles drive the tune toward the 29/3 resonance each spill, extracting beam at kV multiwire septa. The overall Spill Regulation System consists of (1) a “slow” process using ~100-spill averages to adjust the base quad ramp infrequently, (2) a feedforward harmonic content compensator, and (3) the “fast” ML agent reacting during each ongoing spill with on-the-fly additive corrections to the sum of (1) and (2). We have demonstrated improved beam-rate steadying for a fast ML agent compared to a PID controller using a quasi-physical spill simulation, and demonstrated distillation of that simulation into a predictive surrogate model. Current work includes a data-and-training pipeline to generate data-aware surrogates with real-world dynamics, even as the dynamics shift unpredictably. The surrogates are to act as RL environments against which to train our fast ML control agents before deploying them on FPGA in the live system. Further current efforts focus on modeling and controlling beam loss around the storage ring, understanding additional available hardware inputs to the model, and the interplay of these with beam-steadying performance.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Purification and Fractionation of Lignin via ALPHA: Liquid–Liquid Equilibrium for the Lignin–Acetic Acid–Water System

In order to effectively practice the Aqueous Lignin Purification with Hot Agents (ALPHA) process for lignin purification and fractionation, the temperatures and feed compositions where regions of liquid–liquid equilibrium (LLE) exist must be identified. To this end, pseudo-ternary phase diagrams for the lignin–acetic acid–water system were mapped out at 45-95 °C and various solvent: feed lignin mass ratios (S:F). For a given temperature, the accompanying SL (solid–liquid), SLL (solid–liquid–liquid), and one-phase regions were also located. For the first time, ALPHA using acetic acid (AcOH)–water solutions was applied to a lignin recovered via the commercial LignoBoost process. Additionally to determining tie-line compositions for the two regions of LLE that were discovered, the distribution of lignin and key impurities (the latter can negatively impact lignin performance for materials applications) between the two liquid phases was also measured. As a representative example, lignin isolated in the lignin-rich phase was reduced 7x in metals and 4x in polysaccharides by using ALPHA with a feed solvent composition of 50-55% AcOH and an S:F of 6:1, with said lignin being obtained at a yield of 50-70% of the feed lignin and having a molecular weight triple that of the feed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Next Generation System Analysis Model: Recently Added Features and Future Plans

To better enable informed decision-making regarding the back-end of the nuclear fuel cycle, the Integrated Waste Management System (IWMS) program within the U.S. Department of Energy, Office of Nuclear Energy (DOE-NE) has been sponsoring the development and application of system analysis tools capable of analyzing various system options for the management of spent nuclear fuel (SNF) and high-level radioactive waste (HLW).With these tools, IWMS architecture analyses are being conducted to support the future deployment of a comprehensive nuclear waste management system that considers all major back-end aspects of the nuclear fuel cycle (i.e., transportation, storage, and disposal). The Next Generation System Analysis Model (NGSAM) is an agent-based simulation software tool expressly designed to be capable of modeling features within various IWMS architectures. NGSAM imports data from Oak Ridge National Laboratory (ORNL)’s unified database (e.g., historic assembly information, thermal profiles for assembly heat, and at-reactor dry storage loadings) to ensure that each simulation initializes with a realistic representation of the state of commercial SNF in the U.S. Recent major enhancements implemented into NGSAM in the period since NGSAM was last presented at the WM2019 conference include: • Tracking of railroad escort car acquisition and buffer car acquisition • The addition of heavy haul truck (HHT) and barge routes for some sites, as well as support for user-defined inter-modal routes • Updates to the logic that checks the transportation cask thermal limit maps prior to package transport • An allocation method that predicts when reactor sites would pack assemblies from their spent fuel pools for dry storage, and prioritizes shipments directly from the pools of those reactor sites in the preceding periods (before the predicted loadings to dry storage), thus reducing the number of casks loaded into dry storage at reactor sites • The addition of “reactor site family” operational limits to restrict the number of loads of SNF taken from the pool and from dry storage at a given reactor site each year • Added support for multiple canister loading map options and packages with multiple compatible transportation overpacks • Updates to the handling of non-commercial SNF, including a new database that contains data to support the updates • Updates to allow analysis of hypothetical scenarios which include repackaging at reactor sites, e.g., for possible comparative analysis with other scenarios • Implementation of additional output reports, or modification of existing ones • The ability to generate and implement user edits via the NGSAM website • The ability to model loading SNF from pool storage at an interim storage facility (ISF) into dry storage at the ISF • The ability to model consolidating SNF from different existing storage containers at a DOE site into the same DOE standard canister • The ability to model transferring SNF casks from one transportation mode to another, e.g., from HHT to rail, referred to as transloading. These new features have improved NGSAM capabilities and users’ experience with the model. Preliminary NGSAM requirements for modeling advanced reactor fuels, reprocessing, treatment, and conditioning were considered, and this paper describes them at a high level. Other nuclear fuel cycle system analysis tools developed under sponsorship of DOE-NE, like the VISION code developed at Idaho National Laboratory, might be better suited for initial high-level analysis of those technologies and advanced fuel cycles. As technologies are developed and system concepts evolve, NGSAM could provide value by providing more detailed modeling of transport, storage, and disposal of spent fuel and wastes from advanced reactors and advanced fuel cycles at the fuel element and waste container level.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2020

Influence of the Turing instability on the motion of domain boundaries

Turing's theory of pattern formation has provided crucial insights into the behavior of various biological, geographical, and chemical systems over the last few decades. Existing studies have focused on moving-boundary Turing systems for which the motion of the boundary is prescribed by an external agent. In this paper, we present an extension of this theory to a class of systems in which the front motion is governed by the physical processes that occur within the domain. Biological systems exhibiting apically dominant growth and corrosion of metals and alloys highlight some of the noteworthy examples of such systems. In this study, we characterize the nature of interaction between the moving front and the Turing-instability for both an activator-inhibitor and an activator-substrate model. Behavioral regimes of periodic, as well as nonperiodic (nonconstant), growth rates are obtained. Furthermore, the trends in the first show striking similarities with the cyclic-boundary-kinetics observed in experimental systems. In general, a stationary, periodic structure is also left behind the moving front. If the periodicity of the boundary kinetics agrees with the allowed range of the stable-periodic solutions, the pattern formed tends to persist. Lastly, it evolves to a nearby energy-minimum either by peak-splitting, peak-decay, or by settling down to a spatially homogeneous state.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2023

The onset of rare earth metallosis begins with renal gadolinium-rich nanoparticles from magnetic resonance imaging contrast agent exposure

The leitmotifs of magnetic resonance imaging (MRI) contrast agent-induced complications range from acute kidney injury, symptoms associated with gadolinium exposure (SAGE)/gadolinium deposition disease, potentially fatal gadolinium encephalopathy, and irreversible systemic fibrosis. Gadolinium is the active ingredient of these contrast agents, a non-physiologic lanthanide metal. The mechanisms of MRI contrast agent-induced diseases are unknown. Mice were treated with a MRI contrast agent. Human kidney tissues from contrast-naïve and MRI contrast agent-treated patients were obtained and analyzed. Kidneys (human and mouse) were assessed with transmission electron microscopy and scanning transmission electron microscopy with X-ray energy-dispersive spectroscopy. MRI contrast agent treatment resulted in unilamellar vesicles and mitochondriopathy in renal epithelium. Electron-dense intracellular precipitates and the outer rim of lipid droplets were rich in gadolinium and phosphorus. We conclude that MRI contrast agents are not physiologically inert. The long-term safety of these synthetic metal–ligand complexes, especially with repeated use, should be studied further.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Computational and Systems Biology Advances to Enable Bioagent Agnostic Signatures

Enumerated threat agent lists have long driven biodefense priorities. The global SARS-CoV-2 pandemic demonstrated the limitations of searching for known threat agents as compared to a more agnostic approach. Recent technological advances are enabling agent-agnostic biodefense, especially through the integration of multi-modal observations of host-pathogen interactions directed by a human immunological model. Although well-developed technical assays exist for many aspects of human-pathogen interaction, the analytic methods and pipelines to combine and holistically interpret the results of such assays are immature and require further investments to exploit new technologies. In this manuscript, we discuss potential immunologically based bioagent-agnostic approaches and the computational tool gaps the community should prioritize filling.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Adversarial Binaries: AI-guided Instrumentation Methods for Malware Detection Evasion

Adversarial binaries are executable files that have been altered without loss of function by an AI agent in order to deceive malware detection systems. Progress in this emergent vein of research has been constrained by the complex and rigid structure of executable files. Although prior work has demonstrated that these binaries deceive a variety of malware classification models which rely on disparate feature sets, a consensus as to the best approach has not been reached, either in terms of the optimization algorithms or the instrumentation methods. Furthermore, although inconsistencies in the data sets, target classifiers, and functionality verification methods make head-to-head comparisons difficult, here we extract lessons learned and make recommendations for future research.

malware obfuscation↗
DOE OSTI2026

REDI – Readiness Engine for Data Integration

The Readiness Engine for Data Integration (REDI) is an open-source framework for automating, standardizing, and assessing the process of preparing scientific data for AI training. REDI implements a five-stage pipeline (ingest, preprocess, transform, structure, output) with per-stage provenance instrumentation via Flowcept, domain-aware transformation logic (PII anonymization, regridding, graph encoding, and more), and built-in readiness assessment and validation modes. REDI has been evaluated across climate, proteomics, materials science, and nuclear fusion datasets, demonstrating near-ideal parallel scaling to 100 nodes on OLCF's Frontier system. REDI is deployable as an agent-callable skill in coding environments such as Claude Code and OpenAI Codex, and is complemented by SetGo for FAIR compliance and catalog publication.

Brewer, Wesley [Oak Ridge National Laboratory (ORN↗
DOE OSTI2022

System Analysis Modeling and Intermodal Transportation for Commercial Spent Nuclear Fuel

The United States Department of Energy (DOE) has long term goals to develop solutions for managing the nation’s spent nuclear fuel (SNF) and high-level waste (HLW) inventory. The Integrated Waste Management (IWM) program under the DOE office of Nuclear Energy (DOE-NE) is employing system-level engineering and analysis principles to inform potential future waste management system architectures. Managing the spent nuclear waste requires the use of system-level analysis software that takes various aspects of the fuel cycle into account like waste generation, on-site/centralized storage, transportation infrastructure, and long-term disposal. The Next Generation System Analysis Model (NGSAM) is an agent-based model that was developed to simulate the transportation and storage of SNF and HLW. As an agent-based model, NGSAM has the capability to detail the interaction and movement of individual components and groups, such as rail cars and casks. The SNF inventory from commercial nuclear reactors is currently in temporary storage at multiple locations spread across the United States. Shipping of SNF from these locations relies on one of three transportation modes: rail, heavy-haul truck, or barge. Out of the three modes identified, rail is generally the most preferred due to the size of the canisters and casks the SNF would be shipped in. However, under some scenarios, a direct rail route might not be readily available to a reactor site or improving the rail infrastructure at shutdown sites might be too cost-prohibitive for utilities to opt for a direct rail transfer. Under such scenarios, using a barge or heavy haul truck to de-inventory the site and transfer the SNF to a nearby intermodal transfer site with adequate rail infrastructure where the payload could be transferred to a rail car might prove to be an attractive option. This work initially presents the various intermodal transportation options that could be used to transfer SNF from reactor sites to rail cars. This is followed by exploring the operational steps in each of these modes to move the SNF from a reactor site and transfer it to a rail car. This work also presents the procedure of implementing the intermodal transfer methodology in NGSAM using various Java methods. Finally, the process times for accomplishing each of the individual steps are furnished. The implementation ideology, assumptions, and future steps are presented in this work.

Gadey, Harish Reddy↗
DOE OSTI2021

Modeling the Fundamental Radiation Chemistry of the Organic Diluent, and the Effect of Metal Ion Complexation on the Radiochemical Behavior of Active Compounds

All used nuclear fuel (UNF) reprocessing technologies must operate efficiently in the presence of an intense, multi-component (predominantly alpha, beta, and gamma) radiation field. Consequently, radiation-induced degradation of reprocessing systems is of concern, as it negatively impacts process performance over time due to the destruction of both active compounds (ligands, phase modifiers, holdback agents, etc.) and the formation of degradation products. Reprocessing solvent system radiolysis has been linked to changes in separation efficiency and physical properties of solvent mixtures, solvent-recycle longevity, crud formation, and other unexpected outcomes that impact the efficient recovery of valuable materials (e.g., the actinides) and the volume of hazardous radioactive waste for final disposal, i.e., in a geological repository. Consequently, a fundamental understanding of radiolytic processes and their effects on reprocessing solvent system performance is critical for: (i) the cost-effective development and innovation of separation technologies; (ii) the design and implementation of predictive radiation chemical models for process monitoring and lifetimes; and (iii) potentially the ability to exploit radiolytic phenomenon to our benefit, e.g., strategic radiolysis of active molecules to liberate specific degradation products that aid subsequent process stages. Despite extensive investigation into the radiolytic behavior of active solvent system compounds, little attention has been given to understanding (i) the radiation chemical behavior and modification of the organic diluent and (ii) the effect of metal ion complexation on the radiochemical behavior of active compounds.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗
DOE OSTI2022

Aggregation in bottom-up vulnerability assessments and equity implications: The case of Jordanian households’ water supply

Bottom-up methods for water resources modeling rely on acceptability thresholds to find, through a response surface, which deeply uncertain futures lead to system failure. They commonly treat water users as aggregate actors, which may preclude analysis of the equity impacts of interventions. This paper explores how aggregation choices for large groups of water users lead to different policy recommendations in response surface assessments. Herein, two aggregation methods with varying parameters are considered: percentile satisfaction targets and generalized mean. A 2-dimensional stress-test assessment across groundwater availability and population is applied to household water supply in Jordan. The study compares six different policies covering supply enhancement and rebalancing, using a country-wide multi-agent model that characterizes households across socioeconomic strata. For different aggregation levels, policies are ordered by their associated robustness index. Results show that aggregation choices may modify response surfaces as much as policy changes and strongly determine policy preference. Modifying allocation rules can substantially reduce the disparity in household vulnerability. Preferences defined by aggregation intervals provide a finer understanding of trade-offs among water users and may improve deliberation over equity under deep uncertainty.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Robust Resilient Signal Reconstruction under Adversarial Attacks

We consider the problem of signal reconstruction for a system under sparse signal corruption by a malicious agent. The reconstruction problem follows the standard error coding problem that has been studied extensively in the literature. We include a new challenge of robust estimation of the attack support. The problem is then cast as a constrained optimization problem merging promising techniques in the area of deep learning and estimation theory. A pruning algorithm is developed to reduce the "false positive" uncertainty of data-driven attack localization results, thereby improving the probability of correct signal reconstruction. Sufficient conditions for the correct reconstruction and the associated reconstruction error bounds are obtained for both exact and inexact attack support estimation. Moreover, a simulation of a water distribution system is presented to validate the proposed techniques.

Robust, Signal reconstruction, Resilient estimator↗
DOE OSTI2022

Bright, Modular, and Switchable Near-Infrared II Emission from Compact Tetrathiafulvalene-Based Diradicaloid Complexes

Near-infrared (NIR)-emitting molecules are promising candidates for biological sensing and imaging applications; however, many NIR dyes are large conjugated systems which frequently have issues with stability, solubility, and tunability. Here, we report a novel class of compact and tunable fluorescent diradicaloid complexes which are air-, water-, light-, and temperature-stable. These properties arise from a compressed pi manifold which promotes an intense ligand-centered π-π transition in the NIR II (1000-1700 nm) region and which subsequently emits at similar to ~1200 nm. This emission is among the brightest known for monomolecular lumiphores with deep NIR II (>1100 nm) emission, nearly an order of magnitude brighter than the commercially available NIR II dye IR 26. Additionally, this fluorescence is electrochemically sensitive, with efficient switching upon addition of redox agents. The brightness, stability, and modularity of this system distinguish it as a promising candidate for the development of new technologies built around NIR emission.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

ADEPT: A Pedagogical Framework for Integrating Agentic AI with Deterministic Scientific Workflows

The integration of Large Language Models (LLMs) into scientific research promises to accelerate discovery, yet a significant gap remains between the dynamic reasoning of Artificial Intelligence (AI) agents and the static, deterministic nature of canonical scientific workflows. This paper introduces ADEPT (Agentic Discovery and Exploration Platform for Tools), a reference architecture and pedagogical framework explicitly designed to bridge this gap. ADEPT's primary mission is to provide a transparent, "glass-box" environment where researchers and engineers can learn to effectively wrap established scientific software (e.g., BLAST, Nextflow pipelines) and compose it into reliable, agent-driven workflows. We describe its modular, multi-server architecture, which leverages the Model Context Protocol (MCP) for tool serving, LangGraph for robust agentic orchestration, and a secure nsjail-based sandbox for safe code execution. By prioritizing architectural clarity, safety, and modularity, ADEPT serves as an extensible blueprint for building trustworthy AI-augmented systems and fosters the collaborative development necessary to responsibly employ agentic AI for science. We provide practical examples of how to adapt and extend this framework, highlighting its utility in workforce development and AI-readiness capabilities across research and development projects.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Learning Distributed Geometric Koopman Operator for Sparse Networked Dynamical Systems

Koopman operator theory provides an alternative to study nonlinear networked dynamical systems by mapping the state space to an abstract higher dimensional space where the system evolution is linear. Recent works show the application of graph neural networks (GNNs) to learn state to object-centric embeddings and achieve centralized block-wise computation of Koopman operator (KO) under additional assumptions on the underlying agents properties and constraints on the KO structure. However, the computational complexity of learning the Koopman increases exponentially for networked systems where the number of possible system states grows in a combinatorial fashion with the number of nodes. The learning challenge is further amplified for sparse networks by two factors: 1) sample sparsity for learning the Koopman operator in the non-linear space, and 2) the divergence in the dynamics of individual nodes or from one subgraph to another. Our work aims to address these challenge by formulating the representation learning of networked dynamical systems into a multi-agent paradigm and learning the Koopman operator in a distributive manner. The computational as well as performance advantages of distributed Koopman is predominant for sparse networks whereas for fully connected networks, it is shown to coincide with the centralized one. The empirical study on rope system, network of oscillators and a synthetic power system show comparable and superior performance along with computational benefits with the state-of-the-art methods.

Mukherjee, Sayak↗