Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Agents”

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 1,117 records · Page 62
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Molecular Interaction Regulates the Performance and Longevity of Defect Passivation for Metal Halide Perovskite Solar Cells

Defect passivation constitutes one of the most commonly used strategies to fabricate highly efficient perovskite solar cells (PSCs). However, the durability of the passivation effects under harsh operational conditions has not been extensively studied regardless of the weak and vulnerable secondary bonding between the molecular passivation agents and perovskite crystals. Here, we incorporated strategically designed passivating agents to investigate the effect of their interaction energies on the perovskite crystals and correlated these with the performance and longevity of the passivation effects. We unraveled that the passivation agents with a stronger interaction energy are advantageous not only for effective defect passivation but also to suppress defect migration. The prototypical PSCs treated with the optimal passivation agent exhibited superior performance and operational stability, retaining 81.9 and 85.3% of their initial performance under continuous illumination or nitrogen at 85 °C after 1008 h, respectively, while the reference device completely degraded during that time. Furthermore, this work provides important insights into designing operationally durable defect passivation agents for perovskite optoelectronic devices.

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

A Reinforcement Learning Approach to Augment Conventional PID Control in Nuclear Power Plant Transient Operation

The ability of nuclear reactors to operate their power conversion cycles more flexibly will enhance their value to energy grids with variable pricing. Current nuclear control systems are typically classical controllers that are often based on proportional-integral-derivative (PID) control. This paper presents a method of augmenting the existing PID control for difficult transient operations in nuclear power plants using a reinforcement learning–derived feedforward signal applied in real time. The agents, which are trained on a test thermal load-following problem, are designed to improve steam generator outlet temperature control for a range of fast load-following scenarios covering ramp rates from 9%/min to 15%/min. Several reinforcement learning algorithms were initially investigated for the training of the feedforward agents with deep Q-learning (DQN) and proximal policy optimization (PPO) networks, which were found to be the most promising. The DQN controllers utilize discrete actions, giving them a better disturbance rejection at steady state but inconsistent response to initial temperature deviations. In contrast, PPO-trained agents, which take continuous actions except for a dead zone around zero, were shown to have the best combination of high disturbance rejection at steady state and good tracking of the desired temperature value. The ability of the PPO agent was also examined, with the average time of decision making found to be on the order of 1 ms. The fault properties of the controller under the loss of the reinforcement learning agent feedforward signal were also examined. The controller showed strong performance in situations of “no-signal” faults. but was less good at handling “stuck-at” faults, where the feedforward signal remains at a set value. In both cases, however, the PID was able to successfully maintain stability, eventually returning the system to a steady state. It is hoped that this work will allow for the proposed control architecture to be examined for more difficult control problems such that it may eventually be used to adapt existing nuclear plants for more aggressive load-following on grids of the future.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

Monotonicity in the averaging process

In this work, we investigate an averaging process that describes how interacting agents approach consensus through binary interactions. In each elementary step, two agents are selected at random and they reach compromise by adopting their opinion average. We show that the fraction of agents with a monotonically decreasing opinion decays as e - αt , and that the exponent is selected as the extremum from a continuous spectrum of possible values. The opinion distribution of monotonic agents is asymmetric, and it becomes self-similar at large times. Furthermore, the tails of the opinion distribution are algebraic, and they are characterized by two distinct and nontrivial exponents. We also explore statistical properties of agents with an opinion strictly above average.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Characterization of Classical Sheep Scrapie in White-tailed Deer after Experimental Oronasal Exposure

Abstract Background Classic scrapie is a prion disease of sheep and goats that is associated with accumulation of abnormal prion protein (PrPSc) in the central nervous and lymphoid tissues. Chronic wasting disease (CWD) is the prion disease of cervids. This study was conducted to determine the susceptibility of white-tailed deer (WTD) to the classic scrapie agent. Methods We inoculated WTD (n = 5) by means of a concurrent oral/intranasal exposure with the classic scrapie agent from sheep or oronasally with the classic scrapie agent from goats (n = 6). Results All deer exposed to the agent of classic scrapie from sheep accumulated PrPSc. PrPSc was detected in lymphoid tissues at preclinical time points, and necropsies in deer 28 months after inoculation showed clinical signs, spongiform lesions, and widespread PrPSc in neural and lymphoid tissues. Western blots on samples from the brainstem, cerebellum, and lymph nodes of scrapie-infected WTD have a molecular profile similar to CWD and distinct from samples from the cerebral cortex, retina, or the original classic scrapie inoculum. There was no evidence of PrPSc in any of the WTD inoculated with classic scrapie prions from goats. Conclusions WTD are susceptible to the agent of classic scrapie from sheep, and differentiation from CWD may be difficult.

Immunology↗
DOE OSTI2021

Emergency Voltage Regulation in Power Systems via Ripple-Type Control

With increasing penetrations of volatile renewable generation and cyber-physical disruptions, ensuring the safe operation of bulk power systems has become unprecedentedly challenging. Because communication and computational costs restrict centralized system dispatch to being called upon every few minutes, and because purely local schemes are shown to be insufficient, distributed controls have been advocated for handling unanticipated system conditions in real time. The applicability of distributed control schemes, however, is fundamentally limited by their need for widespread communication and model cognizance. In this context, we put forth a hybrid, low-communication, saturation-driven protocol for the coordination of control agents that are distributed over a physical system and are allowed to communicate with peers over a "hotline" communication network. Under this protocol, when agents observe a constraint violation based on local measurements, they respond locally until their control resources saturate, in which case they send a beacon for assistance to peer agents. The scheme ensures that minor violations are efficiently mitigated via fast local controls, whereas severe violations can be handled by collaboration among a relatively small set of agents. We evaluate the performance of this scheme via numerical tests on the IEEE 14-bus test feeder, where agents act upon noisy measurements under diverse scenarios of load variations and severe low-/high-voltage events.

collaboration↗
DOE OSTI2020

A New Approach to Distributed Hypothesis Testing and Non-Bayesian Learning: Improved Learning Rate and Byzantine-Resilience

Here, we study a setting where a group of agents, each receiving partially informative private signals, seek to collaboratively learn the true underlying state of the world (from a finite set of hypotheses) that generates their joint observation profiles. To solve this problem, we propose a distributed learning rule that differs fundamentally from existing approaches, in that it does not employ any form of “belief-averaging”. Instead, agents update their beliefs based on a min-rule. Under standard assumptions on the observation model and the network structure, we establish that each agent learns the truth asymptotically almost surely. As our main contribution, we prove that with probability 1, each false hypothesis is ruled out by every agent exponentially fast, at a network-independent rate that is strictly larger than existing rates. We then develop a computationally-efficient variant of our learning rule that is provably resilient to agents who do not behave as expected (as represented by a Byzantine adversary model) and deliberately try to spread misinformation.

42 ENGINEERING↗
DOE OSTI2021

Decentralized Failure-Tolerant Optimization of Electric Vehicle Charging

We present a decentralized failure-tolerant algorithm for optimizing electric vehicle (EV) charging, using charging stations as computing agents. The algorithm is based on the alternating direction method of multipliers (ADMM) and it has the following features: (i) It handles capacity, peak demand, and ancillary services coupling constraints. (ii) It does not require a central agent collecting information and performing coordination (e.g. an aggregator), instead all agents exchange information and computations are carried out in a fully decentralized fashion. (iii) It can withstand the failure of any number of computing agents, as long as the remaining computing agents are in a connected communications network. We construct this algorithm by reformulating the optimal EV charging problem in a decomposable form, amenable to ADMM, and then developing efficient decentralized solution methods for the subproblems dealing with coupling constraints. We conduct numerical experiments on industry-scale synthetic EV charging datasets, with up to 1,152 charging stations, using a high performance computing cluster. The experiments demonstrate that the proposed algorithm can solve the optimal EV charging problem fast enough to permit the integration of EV charging with real-time electricity markets, even in the presence of failures.

42 ENGINEERING↗
DOE OSTI2021

Deep Reinforcement Learning Enabled Physical-Model-Free Two-Timescale Voltage Control Method for Active Distribution Systems

Active distribution networks are being challenged by frequent and rapid voltage violations due to renewable energy integration. Conventional model-based voltage control methods rely on accurate parameters of the distribution networks, which are difficult to achieve in practice. This paper proposes a novel physical-model-free two-timescale voltage control framework for active distribution systems. To achieve fast control of PV inverters, the whole network is first partitioned into several subnetworks using voltage-reactive power sensitivity. Then, the scheduling of PV inverters in the multiple sub-networks is formulated as Markov games and solved by a multi-agent soft actor-critic (MASAC) algorithm, where each subnetwork is modeled as an intelligent agent. All agents are trained in a centralized manner to learn a coordinated strategy while being executed based on only local information for fast response. For the slower time-scale control, OLTCs and switched capacitors are coordinated by a single agent-based SAC algorithm using the global information with considering control behaviors of the inverters. Particularly, the two-level agents are trained concurrently with information exchange according to the reward signal calculated from the data-driven surrogate model. Comparative tests with different benchmark methods on IEEE 33-and 123-bus systems and 342-node low voltage distribution system demonstrate that the proposed method can effectively mitigate the fast voltage violations and achieve systematical coordination of different voltage regulation assets without the knowledge of accurate system model.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Cyberwheel

Cyberwheel is a high fidelity training environment for autonomous cyber defense agents that includes both simulation and emulation capabilities. Cyberwheel simplifies customization of the training network and easily allows redefining the agent’s reward function, observation space, and action space to support rapid experimentation of novel approaches to agent design. It also provides visibility into agent behaviors necessary for agent evaluation and sufficient documentation and examples to lower the barrier to entry.

Oesch, TimothySean [Oak Ridge National Laboratory ↗
DOE OSTI2025

EVI-EnSitePy (Electric Vehicle Infrastructure – Energy Estimation and Site Optimization Tool in Python) [EVI-X Modeling Suite] [SWR-25-07]

EVI-EnSitePy is a comprehensive agent-based tool designed for the analysis and design of high-power charging sites, encompassing a wide array of site agents including Electric Vehicles (EVs), chargers, energy storage units (ESS), renewable energy resources (DER), and loads. This versatile tool offers diverse functionalities and a modular modeling approach, allowing detailed configuration of agents based on power ratings, port numbers, energy capacities, demand requirements, charger interfaces, and flexibility to customize the tool for project specific goals. By simulating agent interactions and employing various metrics, EVI-EnSitePy enables the assessment of site performance, exploration of energy management systems (EMS), and implementation of innovative EV charging policies. Utilizing EV charge schedules and arrival states, the tool performs thorough charging site simulations, with outputs consisting of agent and site-level power profiles and statistical metrics. Employing a tree graph structure, EVI-EnSitePy supports nested site structures and power distribution modeling. The tool's ability to generate charging schedules deterministically or via stochastic analysis further enhances its versatility. Through its features and capabilities, EVI-EnSitePy offers a powerful platform for informed decision-making in the realm of high-power charging site design and operation.

Jackson, Derek [National Renewable Energy Laborato↗
DOE OSTI2026

ORCHID: Orchestrated Retrieval-Augmented Classification of High-Risk Property with Intelligent Decision-Making

High-Risk Property (HRP) classification is critical at U.S. Department of Energy (DOE) sites, where inventories include sensitive and often dual-use equipment. Compliance must track evolving rules designated by various export control policies to make transparent and auditable decisions. Traditional expert-only workflows are time-consuming, backlog-prone, and struggle to keep pace with shifting regulatory boundaries. We propose ORCHID, a modular agentic framework for HRP classification that pairs retrieval-augmented generation (RAG) with human oversight to produce policy based outputs that can be audited. Small cooperating agents—retrieval, description refiner, classifier, validator, and feedback logger—coordinate via agent-to-agent messaging and invoke tools through the Model Context Protocol (MCP) for model-agnostic on-premise operation. The interface follows an "Item to Evidence to Decision" loop with step-by-step reasoning, on-policy citations, and append-only audit bundles (run-cards, prompts, evidence). In preliminary tests on real HRP cases, ORCHID improves accuracy and traceability over a non-agentic baseline while deferring uncertain items to Subject Matter Experts (SMEs). The demonstration shows single item submission, grounded citations, SME feedback capture, and exportable audit artifacts—illustrating a practical path to trustworthy LLM assistance in sensitive DOE compliance workflows.

Das, Sanjay [ORNL] (ORCID:0009000542591915)↗
DOE OSTI2021

Adapting Secure MultiParty Computation to Support Machine Learning in Radio Frequency Sensor Networks

In this project we developed and validated algorithms for privacy-preserving linear regression using a new variant of Secure Multiparty Computation (MPC) we call "Hybrid MPC" (hMPC). Our variant is intended to support low-power, unreliable networks of sensors with low-communication, fault-tolerant algorithms. In hMPC we do not share training data, even via secret sharing. Thus, agents are responsible for protecting their own local data. Only the machine learning (ML) model is protected with information-theoretic security guarantees against honest-but-curious agents. There are three primary advantages to this approach: (1) after setup, hMPC supports a communication-efficient matrix multiplication primitive, (2) organizations prevented by policy or technology from sharing any of their data can participate as agents in hMPC, and (3) large numbers of low-power agents can participate in hMPC. We have also created an open-source software library named "Cicada" to support hMPC applications with fault-tolerance. The fault-tolerance is important in our applications because the agents are vulnerable to failure or capture. We have demonstrated this capability at Sandia's Autonomy New Mexico laboratory through a simple machine-learning exercise with Raspberry Pi devices capturing and classifying images while flying on four drones.

42 ENGINEERING↗
DOE OSTI2021

Ripple-Type Control for Enhancing Resilience of Networked Physical Systems

Distributed control agents have been advocated as an effective means for improving the resiliency of our physical infrastructures under unexpected events. Purely local control has been shown to be insufficient, centralized optimal resource allocation approaches can be slow. In this context, we put forth a hybrid low-communication saturation-driven protocol for the coordination of control agents that are distributed over a physical system and are allowed to communicate with peers over a “hotline” communication network. According to this protocol, agents act on local readings unless their control resources have been depleted, in which case they send a beacon for assistance to peer agents. Our ripple-type scheme triggers communication locally only for the agents with saturated resources and it is proved to converge. Moreover, under a monotonicity assumption on the underlying physical law coupling control outputs to inputs, the devised control is proved to converge to a configuration satisfying safe operational constraints. The assumption is shown to hold for voltage control in electric power systems and pressure control in water distribution networks. Numerical tests corroborate the efficacy of the novel scheme.

distributed control↗
NASA NTRS1976

Feasibility study of launch vehicle ground cloud neutralization

The distribution of hydrogen chloride in the cloud was analyzed as a function of launch pad geometry and rate of rise of the vehicle during the first 24 sec of burn in order to define neutralization requirements. Delivery systems of various types were developed in order to bring the proposed chemical agents in close contact with the hydrogen chloride. Approximately one-third of the total neutralizing agent required can be delivered from a ground installed system at the launch pad; concentrated sodium carbonate solution is the preferred choice of agent for this launch pad system. Two-thirds of the neutralization requirement appears to need delivery by aircraft. Only one chemical agent (ammonia) may be reasonably considered for delivery by aircraft, because weight and bulk of all other agents are too large.

Vanderarend, P. C.↗
NASA NTRS1976

Bioprocessing development: Immune/cellular applications: Anti-Ig autoantibody and complement-mediated destruction of neoplastic cells

This space bioprocessing contract effort was comprised of four general objectives. These were: (1) the evaluation of current separation processes, (2) the identification of problems relevant to the separation of important biologicals, (3) the identification of ground-based assay methods needed for pre- and postflight analysis of space bioprocessing separation technology; and (4) the establishment of methods to determine the efficiency of space bioprocessing separation procedures. Immunology was deemed advantageous to study the diversity of cells and cell products involved and the extensive interest being given to their separation. Upon recognition of a cellular or molecular agent as foreign to the body, the immune system becomes activated to produce cells whose function is to destroy that agent and cell products whose function is to inactivate the agent and assist in its destruction. Long after the agent is removed from the body, some cells remain in a state of readiness to continue these destructive actions specifically against that agent should further exposure to it occur. This is the basis of acquired immunity to disease.

Twomey, J. J.↗
NASA NTRS1990

Constraint-based evaluation of sequential procedures

Constraining the operation of an agent requires knowledge of the restrictions to physical and temporal capabilities of that agent, as well as an inherent understanding of the desires being processed by that agent. Usually a set of constraints are available that must be adhered to in order to foster safe operations. In the worst case, violation of a constraint may be cause to terminate operation. If the agent is carrying out a plan, then a method for predicting the agent's desires, and therefore possible constraint violations, is required. The conceptualization of constraint-based reasoning used herein assumes that a system knows how to select a constraint for application as well as how to apply that constraint once it is selected. The application of constraint-based reasoning for evaluating certain kinds of plans known as sequential procedures is discussed. By decomposing these plans, it is possible to apply context dependent constraints in production system fashion without incorporating knowledge of the original planning process.

Barry, Matthew R.↗
NASA NTRS1991

A hierarchical distributed control model for coordinating intelligent systems

A hierarchical distributed control (HDC) model for coordinating cooperative problem-solving among intelligent systems is described. The model was implemented using SOCIAL, an innovative object-oriented tool for integrating heterogeneous, distributed software systems. SOCIAL embeds applications in 'wrapper' objects called Agents, which supply predefined capabilities for distributed communication, control, data specification, and translation. The HDC model is realized in SOCIAL as a 'Manager'Agent that coordinates interactions among application Agents. The HDC Manager: indexes the capabilities of application Agents; routes request messages to suitable server Agents; and stores results in a commonly accessible 'Bulletin-Board'. This centralized control model is illustrated in a fault diagnosis application for launch operations support of the Space Shuttle fleet at NASA, Kennedy Space Center.

Adler, Richard M.↗
NASA NTRS1992

Dimensions of complexity in learning from interactive instruction

Robot systems deployed in space must exhibit flexibility. In particular, an intelligent robotic agent should not have to be reprogrammed for each of the various tasks it may face during the course of its lifetime. However, pre-programming knowledge for all of the possible tasks that may be needed is extremely difficult. Therefore, a powerful notion is that of an instructible agent, one which is able to receive task-level instructions and advice from a human advisor. An agent must do more than simply memorize the instructions it is given (this would amount to programming). Rather, after mapping instructions into task constructs that it can reason with, it must determine each instruction's proper scope of applicability. In this paper, we will examine the characteristics of instruction, and the characteristics of agents, that affect learning from instruction. We find that in addition to a myriad of linguistic concerns, both the situatedness of the instructions (their placement within the ongoing execution of tasks) and the prior domain knowledge of the agent have an impact on what can be learned.

Huffman, Scott B.↗