Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed reinforcement learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13

Decentralized Voltage Control with Peer-to-peer Energy Trading in a Distribution Network

Utilizing distributed renewable and energy storage resources via peer-to-peer (P2P) energy trading has long been touted as a solution to improve energy system’s resilience and sustainability. Consumers and prosumers (those who have energy generation resources), however, do not have expertise to engage in repeated P2P trading, and the zero-marginal costs of renewables present challenges in determining fair market prices. To address these issues, we propose a multi-agent reinforcement learning (MARL) framework to help automate consumers’ bidding and management of their solar PV and energy storage resources, under a specific P2P clearing mechanism that utilizes the so-called supply-demand ratio. In addition, we show how the MARL framework can integrate physical network constraints to realize decentralized voltage control, hence ensuring physical feasibility of the P2P energy trading and paving ways for real-world implementations.

Feng, Chen↗

PowerGridworld: A Framework for Multi-Agent Reinforcement Learning in Power Systems [SWR-22-07]

NREL's PowerGridworld provides a modular simulation environment for training heterogenous, grid-aware, multi-agent reinforcement learning (RL) policies at scale. The package enables the user to create component gym environments that can be composed into more complex agents. For example, a grid interactive building environment can be created by composing together component environments each encapsulating the building, PV, and battery physics. These multi-component environments can then be combined into multi-agent simulation where each agent's power consumption/injection becomes an input for solving the optimal power flow on a distribution feeder modeled in OpenDSS. Information from OpenDSS, such as bus voltages and line flows, can be included in the agents' observation spaces to enable grid-aware rewards. The default API for the PowerGridworld simulator conforms to RLLib's MultiAgent API and thus enables distributed training using HPC and cloud resources.

Biagioni, David↗

Demonstration of reconstruction-free static magnetic control of DIII-D plasma with deep reinforcement learning

This paper presents the development and experimental validation of a reinforcement learning (RL)-based magnetic controller on the DIII-D tokamak. The controller directly maps raw magnetic diagnostic signals to actuator commands, replacing the traditional isoflux control algorithm based on equilibrium reconstruction. Four RL controllers are trained using the Soft Actor–Critic algorithm with an asymmetric Actor–Critic architecture in the NSFsim simulator. All controllers are deployed in the DIII-D Plasma Control System and operated with a 4 kHz feedback loop. Two randomization strategies are evaluated during training: evolving kinetic profiles and fixed kinetic profiles within each episode. The latter approach is found to better capture experimental deviations in the current density profile and to provide overall improved control performance. Robust operation is demonstrated across heating power scans in both L- and H-mode plasmas, as well as during transient events such as L–H transitions and pellet injections. Control errors in plasma shape and radial position remained within 1.5–2.0 cm and 1 cm, respectively. A notable discrepancy was observed in the vertical X-point position, with errors of up to approximately 4 cm, attributed to the current density distribution mismatches between simulations and experiments.

DIII-D↗

AIMES Final Technical Report

Many important advances in science and engineering are due to large-scale distributed computing. Notwithstanding this reliance, we are still learning how to design and deploy large-scale production Distributed Computing Infrastructures (DCI). The AIMES project was conceived against this backdrop, following on the heels of a comprehensive survey of scienti c distributed applications [1]. The survey established, arguably for the rst time, the relationship between infrastructure and scienti c distributed applications. It examined well known contributors to the complexity associated with infrastructure, such as inconsistent internal and external interfaces, and demonstrated the correlation with application brittleness. It discussed how infrastructure complexity reinforces the challenges inherent in developing distributed applications.

97 MATHEMATICS AND COMPUTING↗

Safe Deep Reinforcement Learning for Robust Frequency and Voltage-Constrained Networked Microgrid Restoration

Here, this paper proposes a safe soft actor-critic reinforcement learning (RL) algorithm–based controller for networked microgrid restoration. It formulates the post black-start start as a finite-horizon constrained Markov decision process. The RL agent co-optimizes real and reactive power set-points for both grid-forming and grid-following inverters under explicit voltage and frequency constraints, while enforcing proper power sharing via the Mean Active Power Sharing Index (MPSI) and Mean Reactive Power Sharing Index (MQSI). Numerical results obtained on the IEEE 123-bus distribution system show that the proposed method achieves a mean voltage build-up time of 0.01 s without breaching the 5% sharing-violation budget under various load scenarios, considering MPSI and MQSI indices. These findings demonstrate that the proposed method yields fast and safe black-start schedules without resorting to heuristic penalties.

Selim, Alaa [Dartmouth College, Hanover, NH (Unite↗

Designing Agent Utilities for Coordinated, Scalable and Robust Multi-Agent Systems

Coordinating the behavior of a large number of agents to achieve a system level goal poses unique design challenges. In particular, problems of scaling (number of agents in the thousands to tens of thousands), observability (agents have limited sensing capabilities), and robustness (the agents are unreliable) make it impossible to simply apply methods developed for small multi-agent systems composed of reliable agents. To address these problems, we present an approach based on deriving agent goals that are aligned with the overall system goal, and can be computed using information readily available to the agents. Then, each agent uses a simple reinforcement learning algorithm to pursue its own goals. Because of the way in which those goals are derived, there is no need to use difficult to scale external mechanisms to force collaboration or coordination among the agents, or to ensure that agents actively attempt to appropriate the tasks of agents that suffered failures. To present these results in a concrete setting, we focus on the problem of finding the sub-set of a set of imperfect devices that results in the best aggregate device. This is a large distributed agent coordination problem where each agent (e.g., device) needs to determine whether to be part of the aggregate device. Our results show that the approach proposed in this work provides improvements of over an order of magnitude over both traditional search methods and traditional multi-agent methods. Furthermore, the results show that even in extreme cases of agent failures (i.e., half the agents failed midway through the simulation) the system's performance degrades gracefully and still outperforms a failure-free and centralized search algorithm. The results also show that the gains increase as the size of the system (e.g., number of agents) increases. This latter result is particularly encouraging and suggests that this method is ideally suited for domains where the number of agents is currently in the thousands and will reach tens or hundreds of thousands in the near future.

Tumer, Kagan↗

Replace Human Intelligence with Fast and Smart Geometric Reasoning and Graph Neural Network to Accelerate Next Gen ModSim Workflows

We present an agent-guided approach to CAD geometry decomposition that automates hex/hybrid meshing with graph neural networks (GNNs) to accelerate next-generation ModSim workflows. Our end-to-end pipeline (i) reduces 3D boundary-representation (B-Rep) models to a 2D chordal axis skeleton (CAT) and then to a 1D bipartite graph of surface and curve nodes, (ii) assigns per node labels as Cubit® WebCut actions, (iii) trains a multi-action GNN under supervised learning, and (iv) predicts five surface-node and three curve-node actions on out-of-distribution test geometries. Each graph node carries geometric, topological, and meshing attributes drawn from the B-Rep “skin” and CAT “skeleton,” with two-way mappings across 3D↔2D↔1D representations to maintain traceability back to 3D CAD. The supervised learning model exhibits stable convergence of the binary cross-entropy loss and achieves 98.7% accuracy on unseen lattice models. To operationalize decision-making, we rank predicted commands by geometric significance and prototyped the agent-guided workflow through the Cubit® Meshing PowerTool GUI. As a stretch goal, we explore reinforcement learning (RL) to reduce or remove label requirements and to learn policies for action sequences that maximize total reward (e.g., size of hex-meshable regions and resulting hex mesh quality). When all-hex meshing is not feasible, the agent assists in producing hybrid meshes—prioritizing hex in critical regions and transitioning to tetrahedral elements (tets) elsewhere—maintaining fidelity while ensuring robustness. The overarching objective is to replace manual, heuristics-based decomposition with data-driven, reproducible automation, cutting meshing turnaround time by orders of magnitude. We anticipate direct impact on simulation workflows through intelligent, scalable decomposition of complex CAD models into hex-meshable subdomains.

97 MATHEMATICS AND COMPUTING↗

Closing the Loop between In Situ Stress Complexity and EGS Fracture Complexity

We present an agent-guided approach to CAD geometry decomposition that automates hex/hybrid meshing with graph neural networks (GNNs) to accelerate next-generation ModSim workflows. Our end-to-end pipeline (i) reduces 3D boundary-representation (B-Rep) models to a 2D chordal axis skeleton (CAT) and then to a 1D bipartite graph of surface and curve nodes, (ii) assigns per node labels as Cubit® WebCut actions, (iii) trains a multi-action GNN under supervised learning, and (iv) predicts five surface-node and three curve-node actions on out-of-distribution test geometries. Each graph node carries geometric, topological, and meshing attributes drawn from the B-Rep “skin” and CAT “skeleton,” with two-way mappings across 3D↔2D↔1D representations to maintain traceability back to 3D CAD. The supervised learning model exhibits stable convergence of the binary cross-entropy loss and achieves 98.7% accuracy on unseen lattice models. To operationalize decision-making, we rank predicted commands by geometric significance and prototyped the agent-guided workflow through the Cubit® Meshing PowerTool GUI. As a stretch goal, we explore reinforcement learning (RL) to reduce or remove label requirements and to learn policies for action sequences that maximize total reward (e.g., size of hex-meshable regions and resulting hex mesh quality). When all-hex meshing is not feasible, the agent assists in producing hybrid meshes—prioritizing hex in critical regions and transitioning to tetrahedral elements (tets) elsewhere—maintaining fidelity while ensuring robustness. The overarching objective is to replace manual, heuristics-based decomposition with data-driven, reproducible automation, cutting meshing turnaround time by orders of magnitude. We anticipate direct impact on simulation workflows through intelligent, scalable decomposition of complex CAD models into hex-meshable subdomains.

42 ENGINEERING↗

Crew Health Care System (CHeCS) Design Research, Documentations, and Evaluations

The Crew Health Care System (CHeCS) is a group within the Space Life Science Directorate (SLSD) that focuses on the overall health of astronauts by reinforcing the three divisions - the Environmental Maintenance System (EMS), the Countermeasures System (CMS), and the Health Maintenance System (HMS). This internship provided opportunity to gain knowledge, experience, and skills in CHeCS engineering and operations tasks. Various and differing tasks allowed for occasions to work independently, network to get things done, and show leadership abilities. Specific exercises included reviewing hardware certification, operations, and documentation within the ongoing Med Kit Redesign (MKR) project, and learning, writing, and working various common pieces of paperwork used in the engineering and design process. Another project focused on the distribution of various pieces of hardware to off-site research facilities with an interest in space flight health care. The main focus of this internship, though, was on a broad and encompassing understanding of the engineering process as time was spent looking at each individual step in a variety of settings and tasks.

CLement, Bethany M.↗

Ensuring Workforce Readiness for the Energy Industry Through 3D Simulations

In February 2009, during his joint address to Congress, President Obama challenged every American to commit to at least one year of college education and set a goal that the United States should once again have the highest proportion of college graduates in the world by the year 2020. This push by the Obama administration reflected the understanding that while a college degree was becoming a prerequisite to obtain many of the jobs in the 21. century economy; the United States still lagged behind other countries with respect to higher education attainment. Because of this, the U.S. is facing a skills gap between the credentials required for the jobs of today and tomorrow, and the number of skilled workers eligible to fill those jobs. Within the energy industry, the skills gap is compounded by two factors: an expanding domestic industry in energy generation and transmission, and an aging workforce set to retire in large numbers. In fact, a 2013 survey administered by Center for Energy Workforce Development (CEWD) estimates that 36% of the workforce may need to be replaced in the next five years, and nearly 50% will need to be replaced by the year 2023. As a 2014 report by CEWD notes, 'The message for workforce development is clear: qualified and skilled utility technicians and engineers will continue to be in high demand in all types of energy generation, transmission, and distribution jobs'. Simulations provide students the ability to practice proper procedure in a safe environment, test out multiple scenarios to determine an optimal outcome, and repeat practice to solidify learning. Industry currently applies simulation (i.e., 3D Electrical Substation Simulation) as it affords an opportunity for technicians to train in a safe, virtual environment, thus eliminating risks associated with arc flash hazards. The intent of 3D simulation is to enhance the students' understanding of the concepts learned in the course and better prepare students for field work in a related technical discipline. The 3D simulation will reinforce the concepts learned during a specific course and are intended to add real world context to the curriculum. The 3D simulation will be designed such that a student will perform specific job activities incorporating concepts learned during the course. This method of instruction ensures students have an understanding of the concepts and are able to apply the concepts in a real world context. Different scenarios will be developed to ensure a thorough understanding of the concepts with real time feedback when there is a misunderstanding of the concepts. (authors)

97 MATHEMATICS AND COMPUTING↗

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↗

Predicting Mechanical Properties from Microstructure Images in Fiber-Reinforced Polymers Using Convolutional Neural Networks

Evaluating the mechanical response of fiber-reinforced composites can be extremely time-consuming and expensive. Machine learning (ML) techniques offer a means for faster predictions via models trained on existing input–output pairs and have exhibited success in composite research. This paper explores a fully convolutional neural network modified from StressNet, which was originally used for linear elastic materials, and extended here for a non-linear finite element (FE) simulation to predict the stress field in 2D slices of segmented tomography images of a fiber-reinforced polymer specimen. The network was trained and evaluated on data generated from the FE simulations of the exact microstructure. The testing results show that the trained network accurately captures the characteristics of the stress distribution, especially on fibers, solely from the segmented microstructure images. The trained model can make predictions within seconds in a single forward pass on an ordinary laptop, given the input microstructure, compared to 92.5 h to run the full FE simulation on a high-performance computing cluster. These results show promise in using ML techniques to conduct fast structural analysis for fiber-reinforced composites and suggest a corollary that the trained model can be used to identify the location of potential damage sites in fiber-reinforced polymers.

Sun, Yixuan (ORCID:0000000311093380)↗

Distributed Adaptive Control: Beyond Single-Instant, Discrete Variables

In extensive form noncooperative game theory, at each instant t, each agent i sets its state x, independently of the other agents, by sampling an associated distribution, q(sub i)(x(sub i)). The coupling between the agents arises in the joint evolution of those distributions. Distributed control problems can be cast the same way. In those problems the system designer sets aspects of the joint evolution of the distributions to try to optimize the goal for the overall system. Now information theory tells us what the separate q(sub i) of the agents are most likely to be if the system were to have a particular expected value of the objective function G(x(sub 1),x(sub 2), ...). So one can view the job of the system designer as speeding an iterative process. Each step of that process starts with a specified value of E(G), and the convergence of the q(sub i) to the most likely set of distributions consistent with that value. After this the target value for E(sub q)(G) is lowered, and then the process repeats. Previous work has elaborated many schemes for implementing this process when the underlying variables x(sub i) all have a finite number of possible values and G does not extend to multiple instants in time. That work also is based on a fixed mapping from agents to control devices, so that the the statistical independence of the agents' moves means independence of the device states. This paper also extends that work to relax all of these restrictions. This extends the applicability of that work to include continuous spaces and Reinforcement Learning. This paper also elaborates how some of that earlier work can be viewed as a first-principles justification of evolution-based search algorithms.

Wolpert, David H.↗

Design and optimization of a modular hydrogen-based integrated energy system to maximize revenue via nuclear-renewable sources

Here, this paper demonstrates a novel modular distributed framework that uses optimal energy-dispatching strategies to enable greater flexibility and profitability in nuclear-renewable integrated energy systems (NR-IES). Hydrogen is used as a commodity in this framework since its production can improve grid stability and system operational flexibility, decarbonize heavy industry, and create an additional revenue stream for electricity generators, particularly nuclear power plants with high operational expenses. The proposed solution addresses the challenges associated with merging multiple software and services from various domains by using functional mock-up units (FMU) to co-simulate diverse subsystems designed in various platforms. The tightly coupled integrated energy system (IES) is optimized to maximize revenue by utilizing the deep reinforcement learning (DRL) technique to make smart dispatching decisions based on variable electricity prices and the availability of renewable energy. Proximal policy optimization (PPO) algorithm is used in training and testing the DRL agent. Over a period of 120 days, the proposed hydrogen-based IES framework showed about 10% revenue boost compared to a non-hydrogen generating baseline IES while also providing an easily-adoptable framework which can help to improve the flexibility of future generation nuclear power plants.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Extreme Risk Mitigation in Reinforcement Learning using Extreme Value Theory

Risk-sensitive reinforcement learning (RL) has garnered significant attention in recent years due to the growing interest in deploying RL agents in real-world scenarios. A critical aspect of risk awareness involves modelling highly rare risk events (rewards) that could potentially lead to catastrophic outcomes. These infrequent occurrences present a formidable challenge for data-driven methods aiming to capture such risky events accurately. While risk-aware RL techniques do exist, they suffer from high variance estimation due to the inherent data scarcity. Our work proposes to enhance the resilience of RL agents when faced with very rare and risky events by focusing on refining the predictions of the extreme values predicted by the state-action value distribution. To achieve this, we formulate the extreme values of the state-action value function distribution as parameterized distributions, drawing inspiration from the principles of extreme value theory (EVT). We propose an extreme value theory based actor-critic approach, namely, Extreme Valued Actor-Critic (EVAC) which effectively addresses the issue of infrequent occurrence by leveraging EVT-based parameterization. Importantly, we theoretically demonstrate the advantages of employing these parameterized distributions in contrast to other risk-averse algorithms. Our evaluations show that the proposed method outperforms other risk averse RL algorithms on a diverse range of benchmark tasks, each encompassing distinct risk scenarios.

Wang, Yu↗

Numerical Modeling & Optimization of the iProTech Pitching Inertial Pump (PIP) Wave Energy Converter (WEC) (CRADA Final Report)

This project represents a continuation of the collaboration between iProTech and NLR to simulate, optimize and design the iProTech Pitching Inertial Pump (PIP) device. The objectives of this TEAMER project are twofold: 1. Refining the physical characteristics of the existing iProTech PIP WEC-Sim model to enhance the model’s fidelity and include controllable components. Key model enhancements target the inclusion of Coulomb friction, the introduction of a controllable bypass valve, and the replacement of traditional check valves with advanced motorized ones. 2. Exploring traditional and advanced control algorithms. From traditional methods like latching control to cutting-edge reinforcement learning (RL) algorithms, the goal is to ensure the PIP device's adaptability and optimal performance across a range of ocean conditions. NLR is tasked with augmenting the WEC-Sim model and implementing the control algorithms, culminating in performance comparison analyses. iProTech will update their existing 3D models, advise on model improvements, and determine crucial system metrics. WEC-Sim, developed in MATLAB/SIMULINK with Simscape Multibody, is the main piece of software that will be used in this project. Coupled with the MATLAB RL Toolbox, it offers a robust platform for in-depth simulation and optimization of the iProTech PIP device. Building on previous work to explore the PIP design space and optimize its geometry, mass distribution, center of gravity and other key parameters, this project aims to refine iProTech’s existing numerical models and develop effective control algorithms that can seamlessly integrate into their future hardware testing campaigns.

16 TIDAL AND WAVE POWER↗

xcalib : a focal spot calibrator for intense X-ray free-electron laser pulses based on the charge state distributions of light atoms

The xcalib toolkit has been developed to calibrate the beam profile of an X-ray free-electron laser (XFEL) at the focal spot based on the experimental charge state distributions (CSDs) of light atoms. Characterization of the fluence distribution at the focal spot is essential to perform the volume integrations of physical quantities for a quantitative comparison between theoretical and experimental results, especially for fluence-dependent quantities. The use of the CSDs of light atoms is advantageous because CSDs directly reflect experimental conditions at the focal spot, and the properties of light atoms have been well established in both theory and experiment. Theoretical CSDs are obtained using xatom , a toolkit to calculate atomic electronic structure and to simulate ionization dynamics of atoms exposed to intense XFEL pulses, which involves highly excited multiple core-hole states. Employing a simple function with a few parameters, the spatial profile of an XFEL beam is determined by minimizing the difference between theoretical and experimental results. The optimization procedure employing the reinforcement learning technique can automatize and organize calibration procedures which, before, had been performed manually. xcalib has high flexibility, simultaneously combining different optimization methods, sets of charge states, and a wide range of parameter space. Hence, in combination with xatom , xcalib serves as a comprehensive tool to calibrate the fluence profile of a tightly focused XFEL beam in the interaction region.

Instruments & Instrumentation↗

Biopolymer-Templated Titania Film Formation for Nanostructured Coatings Revealed by Machine Learning-Supported Time-Resolved Analysis

This study presents a machine learning approach to derive the film formation of biopolymer-templated titania nanostructures during spray deposition, in combination with in situ grazing-incidence small-angle X-ray scattering (GISAXS). A neural network trained on synthetic GISAXS data directly predicts domain-size distributions from experimental two-dimensional scattering patterns, capturing the full kinetics of nanostructure evolution with high temporal resolution. The predictions reveal hierarchical size distributions and periodic growth features, consistent with layer-by-layer spray deposition and validated by complementary scanning electron microscopy (SEM) imaging. Quantitative comparison with conventional parametric GISAXS fits shows good qualitative agreement, with systematic differences explained by domain-shape assumptions and resolved by applying a geometric scaling factor. Simulated SEM-like surfaces derived from neural network outputs reproduce the porous, foam-like nanoscale morphology observed experimentally, reinforcing the method’s credibility. This integrated approach enables real-time, nondestructive, statistically averaged monitoring of bulk nanostructure development in functional coatings, offering a scalable methodology to accelerate the characterization and process control of sustainably manufactured nanostructured titania films for energy-related applications such as photocatalysis and photovoltaics.

Heger, JulianEliah↗