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 847 records · Page 47
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

CHEMREASONER: Heuristic Search over a Large Language Model’s Knowledge Space using Quantum-Chemical Feedback

The discovery of new catalysts is essential for the design of new and more efficient chemical processes in order to transition to a sustainable future. We introduce an AI-guided computational screening framework unifying linguistic reasoning with quantum-chemistry based feedback from 3D atomistic representations. Our approach formulates catalyst discovery as an uncertain environment where an agent actively searches for highly effective catalysts via the iterative combination of large language model (LLM)-derived hypotheses and atomistic graph neural network (GNN)-derived feedback. Identified catalysts in intermediate search steps undergo structural evaluation based on spatial orientation, reaction pathways, and stability. Scoring functions based on adsorption energies and barriers steer the exploration in the LLM's knowledge space toward energetically favorable, high-efficiency catalysts. We introduce planning methods that automatically guide the exploration without human input, providing competitive performance against expert-enumerated chemical descriptor-based implementations. By integrating language-guided reasoning with computational chemistry feedback, our work pioneers AI-accelerated, trustworthy catalyst discovery.

artificial intelligence↗
NASA NTRS1992

ASRM Process Development in Aqueous Cleaning

Viewgraphs are included on process development in aqueous cleaning which is taking place at the Aerojet Advanced Solid Rocket Motor (ASRM) Division under a NASA Marshall Space and Flight Center contract for design, development, test, and evaluation of the ASRM including new production facilities. The ASRM will utilize aqueous cleaning in several manufacturing process steps to clean case segments, nozzle metal components, and igniter closures. ASRM manufacturing process development is underway, including agent selection, agent characterization, subscale process optimization, bonding verification, and scale-up validation. Process parameters are currently being tested for optimization utilizing a Taguci Matrix, including agent concentration, cleaning solution temperature, agitation and immersion time, rinse water amount and temperature, and use/non-use of drying air. Based on results of process development testing to date, several observations are offered: aqueous cleaning appears effective for steels and SermeTel-coated metals in ASRM processing; aqueous cleaning agents may stain and/or attack bare aluminum metals to various extents; aqueous cleaning appears unsuitable for thermal sprayed aluminum-coated steel; aqueous cleaning appears to adequately remove a wide range of contaminants from flat metal surfaces, but supplementary assistance may be needed to remove clumps of tenacious contaminants embedded in holes, etc.; and hot rinse water appears to be beneficial to aid in drying of bare steel and retarding oxidation rate.

Swisher, Bill↗
DOE OSTI2023

Microgrid energy scheduling under uncertain extreme weather: Adaptation from parallelized reinforcement learning agents

Microgrids are useful solutions for integrating renewable energy resources and providing seamless green electricity to minimize carbon footprint. In recent years, extreme weather events happened often worldwide and caused significant economic and societal losses. Such events bring uncertainties to the microgrid energy scheduling problems and increase the challenges of microgrid operation. Traditional optimization approaches suffer from the inaccuracy of the uncertain microgrid model and the unseen events. Existing reinforcement learning (RL) - based approaches are also hampered by the limited generalization and the increasing computational burden when stochastic formulations are required to accommodate the uncertainties. This paper proposes a new parallelized reinforcement learning (PRL) method based on the probabilistic events to handle the microgrid energy uncertainties. Specifically, several local learning agents are employed to interact with pertinent microgrid environments in a distributed manner and report outcomes to the global agent, which will optimize microgrid energy resources online during extreme events. The stochastic microgrid energy optimization problem is reformulated to include all possible scenarios with probabilities. The advantage estimate functions of learning agents are designed with a backward sweep to transfer the outcomes to the value function updating process. Two simulation studies, stochastic optimization and online testing, are performed to compare with several existing RL approaches. Results substantiate that the proposed PRL method can achieve up to 20% optimization performance improvement with 4 and 28 times less computation cost than Q-learning with experience replay and multi-agent Q-learning approaches, respectively.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS2005

From Here to Autonomicity: Self-Managing Agents and the Biological Metaphors that Inspire Them

We seek inspiration for self-managing systems from (obviously, pre-existing) biological mechanisms. Autonomic Computing (AC), a self-managing systems initiative based on the biological metaphor of the autonomic nervous system, is increasingly gaining momentum as the way forward for integrating and designing reliable systems, while agent technologies have been identified as a key enabler for engineering autonomicity in systems. This paper looks at other biological metaphors such as reflex and healing, heart- beat monitors, pulse monitors and apoptosis for assisting in the realization of autonomicity.

Sterritt, Roy↗
DOE OSTI2024

Modulating Ce 3+ Sites in Ce-Zr Oxide Nanocatalysts through Protamine Biomineralization for Organophosphate Dephosphorylation

Catalytic is a crucial reaction for environmental detoxication of pesticides and neutralization of various molecules classified as chemical warfare agents. Herein, we report on a series of tunable Ce-Zr-based metal oxides, (Zr a Ce 1–a O x ) prepared using a facile biomineralization technique, as catalysts for organophosphates dephosphorylation. Synchrotron scattering and spectroscopy methods showcase that Zr a Ce 1–a O x catalysts are highly defective and exhibit an abundance of Ce 3+ sites that promote oxygen vacancies needed for enhanced dephosphorylation reactions. The catalytic performance was assessed using a model para-nitrophenyl phosphate reaction and showcases a strong dependence on Zr dopant concentration and subsequent tuning of the Ce 3+ /Ce 4+ ratio. Analysis of synchrotron datasets allowed structure-performance correlations between the Ce 3+ concentration and associated oxygen vacancies, the dephosphorylation rate constant, and Zr concentration to be established, confirming that Ce 3+ as active sites is positively correlated with the rate constant. We envision that similar biomineralization approaches can be used to fabricate Ce 3+ -rich Ce-Zr oxide for environmental application in dephosphorylation and other hydrolysis reactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Distributed deep reinforcement learning for simulation control

Abstract Several applications in the scientific simulation of physical systems can be formulated as control/optimization problems. The computational models for such systems generally contain hyperparameters, which control solution fidelity and computational expense. The tuning of these parameters is non-trivial and the general approach is to manually ‘spot-check’ for good combinations. This is because optimal hyperparameter configuration search becomes intractable when the parameter space is large and when they may vary dynamically. To address this issue, we present a framework based on deep reinforcement learning (RL) to train a deep neural network agent that controls a model solve by varying parameters dynamically. First, we validate our RL framework for the problem of controlling chaos in chaotic systems by dynamically changing the parameters of the system. Subsequently, we illustrate the capabilities of our framework for accelerating the convergence of a steady-state computational fluid dynamics solver by automatically adjusting the relaxation factors of the discretized Navier–Stokes equations during run-time. The results indicate that the run-time control of the relaxation factors by the learned policy leads to a significant reduction in the number of iterations for convergence compared to the random selection of the relaxation factors. Our results point to potential benefits for learning adaptive hyperparameter learning strategies across different geometries and boundary conditions with implications for reduced computational campaign expenses 4 4 Data and codes available at https://github.com/Romit-Maulik/PAR-RL . .

42 ENGINEERING↗
DOE OSTI2025

Mic-hackathon 2024: hackathon on machine learning for electron and scanning probe microscopy

Microscopy is one of the primary sources of information on materials structure and functionality at the nanometer and atomic scales. The data generated through microscopy is often contained in well-structured datasets, enriched with extensive metadata and sample histories, although not always with the same level of detail or storage format. The broad incorporation of data management plans by major funding agencies ensures the preservation and accessibility of this data. However, deriving insights from these rich datasets remains challenging due to the lack of established code ecosystems, standardized benchmarks, and integration strategies. Correspondingly, the efficiency of data usage is very low, and time expenditures at the analysis stage are enormous. In addition to post-acquisition data analysis, the emergence of application programming interfaces by major microscope manufacturers now creates opportunities for real-time ML-based data analytics to enable automated decision making, and particularly ML-agent controlled real-time microscope operation. Despite these opportunities, there is a significant gap in integrating the ML community with the broader microscopy community, limiting the value that these methods bring to physics and materials discovery and materials optimization. Hackathons address these challenges by fostering collaboration between ML experts and microscopy professionals, encouraging the development of innovative solutions that leverage ML for microscopy and preparing the workforce of the future both for microscopy-intensive domains areas, instrument manufacturers, and ML scientists interested in real world applications for fundamental research, materials optimization, and manufacturing. The hackathon generated benchmark datasets and digital twins of microscopes that further contribute to the development of the field and establish data analysis ecosystems. All the codes can be found at GitHub(https://github.com/KalininGroup/Mic-hackathon-2024-codes-publication/tree/1.0.0.1) and Zenodo (https://zenodo.org/records/15579940).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Deep reinforcement learning with online data augmentation to improve sample efficiency for intelligent HVAC control

Deep Reinforcement Learning (DRL) has started showing success in real-world applications such as building energy optimization. Much of the research in this space utilized simulated environments to train RL-agent in an offline mode. Very few research have used DRL-based control in real-world systems due to two main reasons: 1) sample efficiency challenge---DRL approaches need to perform a lot of interactions with the environment to collect sufficient experiences to learn from, which is difficult in real systems, and 2) comfort or safety related constraints---user's comfort must never or at least rarely be violated. In this work, we propose a novel deep Reinforcement Learning framework with online Data Augmentation (RLDA) to address the sample efficiency challenge of real-world RL. We used a time series Generative Adversarial Network (TimeGAN) architecture as a data generator. We further evaluated the proposed RLDA framework using a case study of an intelligent HVAC control. With a ≈28% improvement in the sample efficiency, RLDA framework lays the way towards increased adoption of DRL-based intelligent control in real-world building energy management systems.

Kurte, Kuldeep↗
DOE OSTI2020

Optimal Renewable Resource Allocation and Load Scheduling of Resilient Communities

This paper presents a methodology for enhancing community resilience through optimal renewable resource allocation and load scheduling in order to minimize unserved load and thermal discomfort. The proposed control architecture distributes the computational effort and is easier to be scaled up than traditional centralized control. The decentralized control architecture consists of two layers: The community operator layer (COL) allocates the limited amount of renewable energy resource according to the power flexibility of each building. The building agent layer (BAL) addresses the optimal load scheduling problem for each building with the allowable load determined by the COL. Both layers are formulated as a model predictive control (MPC) based optimization. Simulation scenarios are designed to compare different combinations of building weighting methods and objective functions to provide guidance for real-world deployment by community and microgrid operators. The results indicate that the impact of power flexibility is more prominent than the weighting factor to the resource allocation process. Allocation based purely on occupancy status could lead to an increase of PV curtailment. Further, it is necessary for the building agent to have multi-objective optimization to minimize unserved load ratio and maximize comfort simultaneously.

model predictive control↗
NASA NTRS1974

Vapor growth of GeTe single crystals in micro-gravity

The positive effects of micro-gravity on crystal growth and fundamental properties of the vapor transport reaction were established by analyzing the results of GeSe and GeTe vapor transport experiments performed on board Skylab. The analysis was based on a direct comparison of GeSe and GeTe crystals and of mass transport rate data obtained on earth and in space. For this purpose, a total of six transport experiments employing different concentrations of transport agent (GeI4) and two temperature gradients were performed during the Skylab 3 and 4 missions. Extensive ground-based studies demonstrated that the crystal morphology and the mass transport rates of the above systems are affected by the transport conditions, in particular by gravity-driven convection. The results demonstrate unambiguously a considerable improvement of the space crystals in terms of surface perfection, crystalline homogeneity and defect density. The observation of greater mass transport rates than expected in micro-gravity environment is of basic scientific and technological significance. This indicates that conventional transport models are incomplete and demonstrates that crystals of improved quality can be grown at reasonable rates by this technique in space. Results are of practical importance for the modification of crystal growth techniques on earth.

Wiedemeier, H.↗
NASA NTRS1981

Abradable dual-density ceramic turbine seal system

A plasma sprayed dual density ceramic abradable seal system for direct application to the HPT seal shroud of small gas turbine engines. The system concept is based on the thermal barrier coating and depends upon an additional layer of modified density ceramic material adjacent to the gas flow path to provide the desired abradability. This is achieved by codeposition of inert fillers with yttria stabilized zirconia (YSZ) to interrupt the continuity of the zirconia struture. The investigation of a variety of candidate fillers, with hardness values as low as 2 on Moh's scale, led to the conclusion that solid filler materials in combination with a YSZ matrix, regardless of their hardness values, have a propensity for compacting rather than shearing as originally expected. The observed compaction is accompanied by high energy dissipation in the rub interaction, usually resulting in the adhesive transfer of blade material to the stationary seal member. Two YSZ based coating systems which incorported hollow alumino silicate spheres as density reducing agents were surveyed over the entire range of compositions from 100 percent filler to 100 percent YSZ. Abradability and erosion characteristics were determined, hardness and permeability characterized, and engine experience acquired with several system configurations.

Clingman, D. L.↗
NASA NTRS2011

Intelligent Autonomy for Unmanned Surface and Underwater Vehicles

As the Autonomous Underwater Vehicle (AUV) and Autonomous Surface Vehicle (ASV) platforms mature in endurance and reliability, a natural evolution will occur towards longer, more remote autonomous missions. This evolution will require the development of key capabilities that allow these robotic systems to perform a high level of on-board decisionmaking, which would otherwise be performed by humanoperators. With more decision making capabilities, less a priori knowledge of the area of operations would be required, as these systems would be able to sense and adapt to changing environmental conditions, such as unknown topography, currents, obstructions, bays, harbors, islands, and river channels. Existing vehicle sensors would be dual-use; that is they would be utilized for the primary mission, which may be mapping or hydrographic reconnaissance; as well as for autonomous hazard avoidance, route planning, and bathymetric-based navigation. This paper describes a tightly integrated instantiation of an autonomous agent called CARACaS (Control Architecture for Robotic Agent Command and Sensing) developed at JPL (Jet Propulsion Laboratory) that was designed to address many of the issues for survivable ASV/AUV control and to provide adaptive mission capabilities. The results of some on-water tests with US Navy technology test platforms are also presented.

environmental sensing↗
NASA NTRS2018

Effect of Additives: Gamma-Ray Sensors Based on Ionizing Organic Nanocomposites

We have developed ionizing organic-based composites which have demonstrated great promise for radiation sensing. Nickel oxide has been proven as an active material for detecting high energy radiation. The oxidation state of unusual oxides such as nickel oxide in nanocomposites of ionizing organics changes much faster than thin film or bulk, and hence increases the sensitivity for radiation sensing. The resistivity of the oxide composite increases following sequential irradiation processes because of the decrease in holes' concentration. In this paper, we will present the effect of additional oxidizing agent on the morphology, processing and sensing of gamma-rays by oxides - urea based nanocomposites. It was observed that addition of MnO2 decreases the sensitivity. However, it recovers to original properties after irradiation much faster than undoped composites.

Vishall Dayal↗
DOE OSTI2022

Epoxy as Filler or Matrix for Polymer Composites

Epoxy is a widely used polymer because of its ease of processing, high adhesiveness, and high chemical resistance. Epoxy-based composites are commonly used in aerospace, automotive, and marine applications. The epoxy type, function, curing agent, and curing process are discussed in this chapter. Epoxy is used as either a filler or polymer matrix in composite applications. As a filler, the epoxy modification on the fiber is discussed. As a polymer matrix, the epoxy is reinforced by natural and synthetic fibers. The manufacturing process and the fabricated epoxy-based composites’ performance (e.g., mechanical and thermal properties) are investigated. The advantages and disadvantages of epoxy’s function are discussed and summarized. Epoxy modification is an effective approach to improve the composites’ performance.

Zhao, Xianhui↗
NASA NTRS2021

Inertial Transfer Concept for Autonomous In-Space Assembly

Rocket payload and fairing size have placed strict mass and volume limitations on single launch in-space structures. These limitations are what motivates In-Space Assembly (ISA)and where Inertial Transfer has potential to improve or augment ISA capabilities by utilizing an efficient multi-agent autonomous system to transport untethered payloads. Adoption of an inertial transfer approach significantly reduces system complexity, energy and time needed for assembly compared to conventional ISA capabilities, increasing the likelihood of early adoption for assembly. The Inertial Transfer concept is presented including autonomy capabilities, potential risks, and three system configurations. These configurations are based on the autonomous system’s available sensor coverage that tracks the payload’s state. Participating agents have well-defined roles, expectations, and assumptions in their physical organization and coordination of actions depending the mode of operation. The sensor information available to participating agents is considered as it affects the uncertainty in the estimated state and theactions required to capture the payload.

Multi-Agent↗
DOE OSTI2026

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48% ($H_T$) and 28% (AD), with a cumulative gain of up to 2% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56% ($H_T$) and 28% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO_LHC (see repo for details).

Ding, Zixin [Chicago U.]↗
DOE OSTI2025

Explainable and Differentiable Reinforcement Learning for Multi-objective Optimization in Particle Accelerators

Operating particle accelerators involves optimizing multiple goals simultaneously, which can be challenging due to trade-offs among objectives. While evolutionary algorithms like the genetic algorithm (GA) have been used for various Multi-Objective Optimization (MOO) tasks, they are not inherently suited for complex control problems. This talk highlights two variations of Reinforcement Learning (RL) for concurrently optimizing heat load and trip rates at the Continuous Electron Beam Accelerator Facility (CEBAF). The problem involves strict constraints on individual states, actions, and overall energy requirements of the beam. First, this talk highlights how differentiability can be harnessed through a Deep Differentiable Reinforcement Learning (DDRL) approach to address MOO issues within particle accelerators. We examine the DDRL method alongside Model Free Reinforcement Learning (MFRL), GA, and Bayesian Optimization (BO). The performance of these methods is assessed by generating a Pareto-front for two objectives. Our findings indicate that DDRL excels in handling high-dimensional problems more effectively than MFRL, BO, and GA. Next, we will show integration of explainable physics-based constraints into RL algorithms to enhance trans- parency and trust in decision-making processes by enabling users to verify that agents adhere to established physical principles. This surrogate function can be modeled using neural networks or sparse dictionary mod- els. By examining the mathematical form of the learned constraint function, we are able to confirm the agent has learned to use the established physics of each environment provided but the surrogate model. In addi- tion, we find that the introduction of a mathematical functional dictionary based surrogate model enables our reinforcement learning algorithms to reliably converge for difficult high-dimensional accelerator controls environments.

Rajput, Kishansingh [Thomas Jefferson National Acc↗
DOE OSTI2023

Direct synthesis of ordered mesoporous materials from thermoplastic elastomers

Abstract The ability to manufacture ordered mesoporous materials using low-cost precursors and scalable processes is essential for unlocking their enormous potential to enable advancement in nanotechnology. While templating-based methods play a central role in the development of mesoporous materials, several limitations exist in conventional system design, including cost, volatile solvent consumption, and attainable pore sizes from commercial templating agents. This work pioneers a new manufacturing platform for producing ordered mesoporous materials through direct pyrolysis of crosslinked thermoplastic elastomer-based block copolymers. Specifically, olefinic majority phases are selectively crosslinked through sulfonation reactions and subsequently converted to carbon, while the minority block can be decomposed to form ordered mesopores. We demonstrate that this process can be extended to different polymer precursors for synthesizing mesoporous polymer, carbon, and silica. Furthermore, the obtained carbons possess large mesopores, sulfur-doped carbon framework, with tailorable pore textures upon varying the precursor identities.

36 MATERIALS SCIENCE↗