SEARCH · Engineering Papers

Results for “dynamic behavior”

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.

SourceResearch connectionFrom yearThrough year

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

At least 73 records · Page 4

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Anomaly Detection, Localization and Classification using Drifting Synchrophasor Data Streams

With ongoing automation and digitization of the electric power system, several Phasor Measurement Units(PMUs) have been deployed for monitoring and control. PMU data can have multiple anomalies, and many of the researchers in the past have concentrated on training machine/deep learning algorithms offline for anomaly detection over PMU data (i.e., not in real time). These machine/deep learning algorithms, when trained offline on a sample rather than a population of the dataset, fail to consider the dynamic behavior of the power grid in real-time, resulting in low accuracy. Considering the dynamic behavior of the power grid (e.g., change in load, generation, distributed energy resources (DERs) switching, network, controls), the definition of data anomalies varies in time and requires online training. A fundamental challenge is to enable online (i.e., real-time) training of machine/deep learning algorithms for anomaly detection over streaming PMU data. While machine/deep learning is often desirable to manage data streams, training a deep learning algorithm over streaming PMU data is nontrivial due to changes in data statistics caused by dynamic streaming data. This paper proposes PMUNET: a novel device-level deep learning-based data-driven approach for anomaly detection, localization, and classification over streaming PMU data, using online learning and multivariate data-drift detection algorithm .Two variants of PMUNET, Dynamic data Change Driven Learning (DCDL) and Continuity Driven Learning (CDL), are proposed and compared. DCDL aims to train the deep learning algorithm whenever the definition of anomaly changes due to the power grid dynamics. On the other hand, CDL continuously trains the deep learning algorithm over the PMU data-stream. The experimental results verify that DCDL outperforms CDL and other efficient anomaly detection methods over multiple events such as faults and load/ generator/capacitor/DERs variations/switching for IEEE 14 and 39 Bus test system as well as real PMU industrial data. The result verifies that DCDL variant of PMUNET improves over existing approach with a gain of 2% - 10% in terms of accuracy, false-positive rate, and false-negative rate.

adversarial deep learning↗

DOE OSTI2024

Aggregate data‐driven dynamic modeling of active distribution networks with DERs for voltage stability studies

Abstract Electric distribution networks increasingly host distributed energy resources based on power electronic converter (PEC) toward active distribution networks (ADN). Despite advances in computational capabilities, electromagnetic transient models are limited in scalability because of their reliance on exact data about the distribution system and each of its components. Similarly, the use of the DER_A model, which is intended to examine the combined dynamic behavior of many DERs, is limited by the difficulty in parameterization. There is a need for improved dynamic models of DERs for use in large power system simulations for stability analysis. This paper proposes an aggregate model‐free, data‐driven approach for deriving a dynamic partitioned model (DPM) of ADNs. Detailed residential distribution feeders were first developed, including PEC‐based DERs and composite load models (CMLDs), from which the aggregated DPM was derived. The performance was evaluated through various case studies and validated against the detailed ADN model and state‐of‐the‐art DER_A model with CMLD. The data‐driven DPM achieved a of over 90%, accurately representing the aggregated dynamic behavior of ADNs. Furthermore, the DPM significantly accelerated the simulation process with a computational speedup of 68 times compared to the detailed ADN and a 3.5 times speedup compared to the DER_A CMLD model.

42 ENGINEERING↗

DOE OSTI2025

Backbone Dynamics of Bottlebrush Polymers Studied by Neutron Scattering

Bottlebrush polymers have a versatile architecture that is highly customizable due to the combination of a linear backbone and side chains. As a result of the chemical tethering of the side chains, both parts cannot be easily be separated. Huge effort is seen regarding the dynamical behavior of the side chains or the entire bottlebrush polymer, whereas few studies are available considering the backbone. Here, isotopic labeling in combination with quasi-elastic neutron scattering was used to compare the dynamical behavior of the bottlebrush’s backbone with the side chain dynamics. Keeping the side chains deuterated, (h-PNB)- g -(d-PPO), leads to the scattering signal dominated by backbone dynamics, while the fully protonated sample, (h-PNB)- g -(h-PPO), gives side chain dynamics. Both results reveal slower dynamics associated with the backbone with less heterogeneity, as seen for the side chains. Additionally, a plasticizer effect for the backbone dynamics is confirmed by extracting the glass transition temperature and comparing it with pure linear PNB.

neutron scattering↗

DOE OSTI2025

Probing the Self-Assembly dynamics of cellulose nanocrystals by X-ray photon correlation spectroscopy

Charge-stabilized colloidal cellulose nanocrystals (CNCs) can self-assemble into higher-ordered chiral nematic structures by varying the volume fraction. The assembly process exhibits distinct dynamics during the isotropic to liquid crystal phase transition, which can be elucidated using X-ray photon correlation spectroscopy (XPCS). Anionic CNCs were dispersed in propylene glycol (PG) and water spanning a range of volume fractions, encompassing several phase transitions. Coupled with traditional characterization techniques, XPCS was conducted to monitor the dynamic evolution of the different phases. Additionally, simulated XPCS results were obtained using colloidal rods and compared with the experimental data, offering additional insights into the dynamic behavior of the system. The results indicate that the particle dynamics of CNCs undergo a stepped decay in three stages during the self-assembly process in PG, coinciding with the observed phases. The phase transitions are associated with a total drop of Brownian diffusion rates by four orders of magnitude, a decrease of more than a thousand times slower than expected in an ideal system of repulsive Brownian rods. Given the similarity in the phase behaviors in CNCs dispersed in PG and in water, we hypothesize that these dynamic behaviors can be extrapolated to other polar solvent environments. Importantly, these findings represent the direct measurement of CNC dynamics using XPCS, underscoring the feasibility of directly assessing the dynamic behavior of other rod-like colloidal suspensions.

36 MATERIALS SCIENCE↗

DOE OSTI2024

Investigating Mountain Watershed Headwater‐To‐Groundwater Connections, Water Sources, and Storage Selection Behavior With Dynamic‐Flux Particle Tracking

Abstract Climate change will impact mountain watershed streamflow both directly—with changing precipitation amounts and variability—and indirectly—through temperature shifts altering snowpack, melt, and evapotranspiration. To understand how these complex processes will affect ecosystem functioning and water resources, we need tools to distinguish connections between water sources (rain/snowmelt), groundwater storage, and exit fluxes (streamflow/evapotranspiration), and to determine how these connections change seasonally and as climate shifts. Here, we develop novel watershed‐scale approaches to understand water source, storage, and exit flux connections using a dynamic‐flux particle tracking model (EcoSLIM) applied in California's Cosumnes Watershed, which connects the Sierra Nevada and Central Valley. This work develops new visualizations and applications to provide mechanistic understanding that underpins the interpretation of isotopic field data at watershed scales to distinguish sources, flow paths, residence times, and storage selection. In our simulations, streamflow comes primarily from snow‐derived water while evapotranspiration generally comes from rain. Most streamflow starts above 1,000 m while evapotranspiration is sourced relatively evenly across the watershed and is generally younger than streamflow. Modeled streamflow consists primarily of water sourced from precipitation in the previous 5 years but before the current water year, while ET consists primarily of water from precipitation in the current water year. ET, and to a lesser extent streamflow, are both younger than water in groundwater storage. However, snowmelt‐derived streamflow preferentially discharges older water from snow‐derived storage. Dynamic‐flux particle tracking and new approaches presented here enable novel model‐tracer comparisons in large‐scale watersheds to better understand watershed behavior in a changing climate.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

In-Situ Observation of FCC→HCP Transformation-Induced Plasticity Behavior During Dynamic Deformation of CoCrNi Multi-principal Element Alloys

CoCrNi multi-principal element alloys (MPEAs) have potential as high toughness, blast resistant materials, but limited studies of their high strain rate behavior have been performed. Their high toughness has been attributed to low stacking fault energies and the activation of toughness enhancing deformation mechanisms, such as twinning- and transformation-induced plasticity (TWIP and TRIP, respectively), that may be affected by increased strain rate. This work examines a series of CoCrNi MPEAs tested with a miniaturized pressure bar, while simultaneous synchrotron x-ray diffraction and imaging are performed, and compares these results to those collected during quasi-static deformation at elevated temperatures. Here, the collected diffraction indicates that TRIP was active at elevated strain rates for the most Co-rich alloys containing 50 or 55 at. pct. Co, while TRIP behavior was suppressed at elevated strain rates in a 40 at. pct. Co alloy anticipated to experience TRIP. The suppression of TRIP in the lower Co alloy at high strain rates is attributed to adiabatic heating, consistent with suppression of TRIP as a result of elevated temperatures.

36 MATERIALS SCIENCE↗

DOE OSTI2020

Distribution System Segmented Model Simplification with Independent Dynamically Changing End-Use Loads

With more emerging technologies in power distribution systems, to support quasi-static time-series (QSTS) simulation over short time frames, accurately modeling the dynamic behavior of end-use load under various conditions is im- portant. However, time and cost can be a burden for the QSTS simulation of thousands of houses representing dynamic load behaviors on large distribution feeders. There are two main contributions of this paper in order to reduce the QSTS simulation complexity. First, the paper presents a method to stochasti- cally model the dynamic behavior of end-use load for QSTS simulations. Second, using the stochastically modeled load behavior as an input to a distribution sys- tem, and building on our previous QSTS simulation simplification/acceleration method that relies on uniform load behavior, this paper presents a segment substitution approach for the simplification of distribution system models with independent dynamically changing end-use loads. Our numerical results show that compared to GridLAB-D, a software tool that simulates end-use load be- havior through physical models, the proposed stochastic modeling approach achieved an 89.3% simulation time reduction for end-use loads, while accurately capturing the temporally changing total load behavior. Moreover, the QSTS simulation of a large test distribution system through OpenDSS showed that the proposed simplification method decreased the feeder simulation time by a factor of three at less than 0.3% voltage error. Compared to our previous method with independent dynamic loads, the voltage error is reduced by a factor of four.

Power distribution system modeling, stochastic pro↗

DOE OSTI2025

Accurate and efficient predictions of keyhole dynamics in laser materials processing using machine learning-aided simulations

The keyhole phenomenon has been widely observed in laser materials processing, including laser welding, remelting, cladding, drilling, and additive manufacturing. Keyhole-induced defects, primarily pores, dramatically affect the performance of final products, impeding the broad use of these laser-based technologies. The formation of these pores is typically associated with the dynamic behavior of the keyhole. So far, the accurate characterization and prediction of keyhole features, particularly keyhole depth, as a function of time, has been a challenging task. In situ characterization of keyhole dynamic behavior using the synchrotron X-ray technique is informative but complicated and expensive. Current simulations are generally hindered by their poor accuracy and generalization abilities in predicting keyhole depths due to the lack of accurate laser absorptance data. In this study, we develop a machine learning-aided simulation method that accurately predicts keyhole dynamics, especially in keyhole depth fluctuations, over a wide range of processing parameters. In two case studies involving titanium and aluminum alloys, we achieve keyhole depth prediction with a mean absolute percentage error of 10 %, surpassing those simulated using the ray-tracing method with an error margin of 30 %, while also reducing computational time. This exceptional fidelity and efficiency empower our model to serve as a cost-effective alternative to synchrotron experiments. Our machine learning-aided simulation method is affordable and readily deployable for a large variety of materials, opening new doors to eliminate or reduce defects for a wide range of laser materials processing techniques.

Computational fluid dynamics↗

DOE OSTI2024

Gold–Thiolate Nanocluster Dynamics and Intercluster Reactions Enabled by a Machine Learned Interatomic Potential

Monolayer protected metal clusters comprise a rich class of molecular systems and are promising candidate materials for a variety of applications. While a growing number of protected nanoclusters have been synthesized and characterized in crystalline forms, their dynamical behavior in solution, including prenucleation cluster formation, is not well understood due to limitations both in characterization and first-principles modeling techniques. Recent advancements in machine-learned interatomic potentials are rapidly enabling the study of complex interactions such as dynamical behavior and reactivity on the nanoscale. Here, we develop an Au-S-C-H atomic cluster expansion (ACE) interatomic potential for efficient and accurate molecular dynamics simulations of thiolate-protected gold nanoclusters (Au n (SCH 3 ) m ). Trained on more than 30,000 density functional theory calculations of gold nanoclusters, the interatomic potential exhibits ab initio level accuracy in energies and forces and replicates nanocluster dynamics including thermal vibration and chiral inversion. Long dynamics simulations (up to 0.1 μs time scale) reveal a mechanism explaining the thermal instability of neutral Au 25 (SR) 18 clusters. Specifically, we observe multiple stages of isomerization of the Au 25 (SR) 18 cluster, including a chiral isomer. Additionally, we simulate coalescence of two Au 25 (SR) 18 clusters and observe series of clusters where the formation mechanisms are critically mediated by ligand exchange in the form of [Au-S] n rings.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

Effects of pore connectivity and tortuosity on the dynamics of fluids confined in sub-nanometer pores

Dynamical behavior of fluids under nano-pore confinement is studied extensively as it has important implications for several industrial as well as geological processes. Pore network in many porous materials exhibits a varied degree of inter connections. The extent of this pore connectivity may affect the structural and dynamical behavior of the confined fluid. However, studies of fluid confinement addressing these effects systematically are lacking. In this paper we report molecular dynamics simulation studies addressing the effects of pore connectivity on the dynamics of two representative fluids – CO 2 and ethane in silicalite by systematically varying the degree of pore connectivity through selectively blocking some pore space with immobile methane molecules. By selectively turning off the pore spaces in the shape of straight, or tortuous zigzag channels, we also probe the effects of pore tortuosity. In general, pore connectivity is found to facilitate both the translational as well as rotational dynamics of both fluids, while the intermolecular modes of vibration in both fluids remain largely unaffected. The effects of providing connections between a set of straight or zigzag channel-like pores are however more nuanced. Pore tortuosity facilitates the rotational motion, but suppresses the translational motion of CO 2 , while its effects on the rotational and translational motion of ethane are less pronounced. The intermolecular vibrational modes of both fluids shift to higher energies with an increase in the number of tortuous pores. The results reported here provide a detailed molecular level understanding of the effects of pore connectivity on the dynamics of fluids and thus have implications for applications like fluid separation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

Machine learning-based ethylene concentration estimation, real-time optimization and feedback control of an experimental electrochemical reactor

With the increase in electricity supply from clean energy sources, electrochemical reduction of carbon dioxide (CO 2 ) has received increasing attention as an alternative source of carbon-based fuels. As CO 2 reduction is becoming a stronger alternative for the clean production of chemicals, the need to model, optimize and control the electrochemical reduction of the CO 2 process becomes inevitable. However, on one hand, a first-principles model to represent the electrochemical CO 2 reduction has not been fully developed yet because of the complexity of its reaction mechanism, which makes it challenging to define a precise state-space model for the control system. On the other hand, the unavailability of efficient concentration measurement sensors continues to challenge our ability to develop feedback control systems. Gas chromatography (GC) is the most common equipment for monitoring the gas product composition, but it requires a period of time to analyze the sample, which means that GC can provide only delayed measurements during the operation. Moreover, the electrochemical CO 2 reduction process is catalyzed by a fast-deactivating copper catalyst and undergoes a selectivity shift from the product-of-interest at the later stages of experiments, which can pose a challenge for conventional control methods. To this end, machine learning (ML) techniques provide a potential approach to overcome those difficulties due to their demonstrated ability to capture the dynamic behavior of a chemical process from data. Motivated by the above considerations, we propose a machine learning-based modeling methodology that integrates support vector regression and first-principles modeling to capture the dynamic behavior of an experimental electrochemical reactor; this model, together with limited gas chromatography measurements, is employed to predict the evolution of gas-phase ethylene concentration. The model prediction is directly used in a proportional-integral (PI) controller that manipulates the applied potential to regulate the gas-phase ethylene concentration at energy-optimal set-point values computed by a real-time process optimizer (RTO). Specifically, the RTO calculates the operation set-point by solving an optimization problem to maximize the economic benefit of the reactor. Finally, suitable compensation methods are introduced to further account for the experimental uncertainties and handle catalyst deactivation. The proposed modeling, optimization, and control approaches are the first demonstration of active control for a CO 2 electrolyzer and contribute to the automation and scale-up efforts for electrified manufacturing of fuels and chemicals starting from CO 2 .

42 ENGINEERING↗

DOE OSTI2020

Nonlinear modeling and performance analysis of cracked beam microgyroscopes

Cracks constitute a common structural defect in microelectromechanical systems that may arise during manufacturing, mechanical fatigue, or shock loading. Areas weakened by cracks increase the flexibility of the microstructure. Furthermore, depending on the crack severity, the increased flexibility can cause dramatic changes in the static and dynamic behaviors of electrically actuated MEMS. In this study, a numerical investigation of the depth and location of multiple edge cracks on the performance of beam microgyroscopes is conducted. Applying the Griffith strain energy release theorem for an edge crack, the vibrational characteristics of microgyroscopes with multiple cracks including the static deflection, natural frequencies, and mode shapes are studied analytically. Then, the differential quadrature method is employed to model the nonlinear dynamic behaviors of the cracked gyroscope system. Cracks initiated along the driving or sensing directions of the microgyroscope are found to have differing effects on its nonlinear dynamic response. This numerical study reveals that the onset of severe single cracks or a series of multiple cracks can significantly affect the sensitivity of the microgyroscope to base rotations and therefore leads to a degradation in the performance of the damaged MEMS sensor. This study also shows that a potential design approach for broadband microgyroscope systems manufactured with cracks can be proposed.

42 ENGINEERING↗

DOE OSTI2023

Supramolecular polymers: Dynamic assemblies of “dancing” monomers

The objective of this perspective on supramolecular polymers is to highlight how their dynamic nature based on noncovalent bonding among monomers can have a profound impact on their functions. We focus here on peptide amphiphile supramolecular polymers developed in our laboratory and use recent results on their dynamic behavior to reflect on the exciting functions that might emerge in these systems. We are greatly motivated by recent results that demonstrate unprecedented bioactivity in supramolecular polymers to address the enormous scientific challenge of finding strategies to reverse paralysis caused by traumatic injuries or disease. We suggest that future opportunities exist for novel functions in supramolecular polymers by tuning dynamic behavior through the chemical design of monomers, and also by gaining an understanding of the precise spatiotemporal mechanisms linked to supramolecular motion in these emerging polymers.

36 MATERIALS SCIENCE↗

DOE OSTI2025

DESI 2024: Constraints on physics-focused aspects of dark energy using DESI DR1 BAO data

Baryon acoustic oscillation data from the first year of the Dark Energy Spectroscopic Instrument (DESI) provide near percent-level precision of cosmic distances in seven bins over the redshift range z=0.1–4.2. Here, this paper is the follow-up to the original DESI BAO cosmology paper [A. G. Adame et al. (DESI Collaboration), arXiv:2404.03002], which considered the conventional w 0 w a cold dark matter (CDM) model. We use the novel DESI data, together with other cosmic probes, to constrain the background expansion history using some well-motivated physical classes of dark energy. In particular, we explore three physics-focused behaviors of dark energy from the equation of state and energy density perspectives: the thawing class (matching many simple quintessence potentials), emergent class (where dark energy comes into being recently, as in phase transition models), and mirage class [where phenomenologically the distance to cosmic microwave background (CMB) last scattering is close to that from a cosmological constant Λ despite dark energy dynamics]. All three classes fit the data at least as well as Λ ⁢CDM, and indeed can improve on it by Δ⁢χ 2 ≈ –5 to –17 for the combination of DESI BAO with CMB and supernova data while having one more parameter. The mirage class does essentially as well as w 0 ⁢w a CDM and exhibits moderate to strong Bayesian evidence preference with respect to Λ⁢ CDM. These classes of dynamical behaviors highlight worthwhile avenues for further exploration into the nature of dark energy.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2023

Analysis of Dynamical Peculiarities in Nanoalloys at Subsystems Level: Dynamical Degrees of Freedom, Temperature Differences, and the Chameleon Effect

Abstract A novel analysis of the dynamical behavior of nanoalloy systems, as represented by model Ni/Al 13‐atom clusters, over a broad range of energies that cover the stage‐wise transition of the systems from their solid‐like to liquid‐like state is presented. Conceptually, the analysis is rooted in partitioning the systems into judiciously chosen subsystems and characterizing the latter in terms of subsystem‐specific dynamical descriptors that include dynamical degrees of freedom , root‐mean‐square bond‐length fluctuation , and element‐specific subsystem temperature . The analysis reveals a host of intriguing new peculiarities in the dynamical behavior of the Ni/Al 13‐mers, among which are what we call the chameleon effect and the difference in the temperatures of the Ni and Al subsystems at high energies, a difference that strongly depends on the cluster composition and also changes with energy. These do not have an analog in pure Ni 13 and Al 13 and are explained in terms of the coupled effects of the difference between the masses of the Ni and Al atoms (the mass effect) and of the difference in the anharmonicity of the overall interaction potential as experienced by the Ni and Al subsystems of the clusters (the potential effect).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2020

Self-Secure Inverters Against Malicious Setpoints

The next generation of grid-interactive inverters brings a communication feature that allows data sharing from utility supervisory controllers and smart devices that are connected to the same network. This feature enhances the control capabilities of grid-interactive inverters to provide services beyond active power injection. However, communication networks entail more vulnerable surfaces to malicious attacks that may result in modifying active and reactive power setpoints and causing weak-grid conditions or abnormal inverter operation. In this paper, steady-state and the dynamic behavior of the inverter for the incoming setpoints are analyzed to detect false data injection attacks and provide device-level security. The steady-state behavior of the inverter in the operating region is determined from the grid parameters such as the grid voltage and the grid impedance. These estimations are accomplished by the proposed self-security technique through a low-frequency signal injection-based approach combined with the recursive least square method. Moreover, a reduced fourth-order inverter model is used as the dynamic reference model, and grid parameters as well as the incoming setpoints are implemented to the reference model to verify whether the dynamic behavior of the inverter is inside the permissible region of operation. The validity and performance of the proposed method are verified experimentally through Allen-Bradley Powerflex 755 three-phase inverter and a 12 kW NHR 9410 regenerative power grid emulator. The results show that the self-secure smart-inverter is able to accept or reject the incoming commands and thus is protected from malicious cyber-physical attacks.

Hossen, Tareq↗

DOE OSTI2020

Energy-Storage Fed Smart Inverters for Mitigation of Voltage Fluctuations in Islanded Microgrids

The continuous integration of intermittent low-carbon energy resources makes islanded microgrids vulnerable to voltage fluctuations. Besides, different dynamic response of synchronous-based and inverter-based distributed generation (DG) units can result in an instantaneous power imbalance between supply and demand during transients. As a result, the ac-bus voltage of microgrid starts oscillating which might have severe consequences such as blackouts. This paper modifies the conventional control scheme of battery energy storage systems (BESSs) to participate in improving the dynamic behavior of islanded microgrids by mitigating the voltage fluctuations. A piecewise linear-elliptic (PLE) droop is proposed and employed in BESS to achieve an enhanced voltage profile by injecting/absorbing reactive power during transients. In this way, the conventional inverter implemented in BESS turns into a smart inverter to cope with fast transients. Using the proposed approach in this paper, any linear droop curve with a specified coefficient can be replaced by a PLE droop curve. Compared with linear droop, an enhanced dynamic response is achieved by utilizing the proposed PLE droop. Case study results are presented using PSCAD/EMTDC to demonstrate the superiority of the proposed approach in improving the dynamic behavior of islanded microgrids.

Pilehvar, Mohsen S.↗

DOE OSTI2025

Model-based Hierarchical Reinforcement Learning for Improved Physical Security Design: A Prototype

Prior work in FY24 developed an adversarial AI agent aid in path analysis of physical protection systems. This agent, trained using a model-based reinforcement learning algorithm, was able to successfully learn the most vulnerable path in facilities. It was able to extend the current state of practice for physical protection design by exhibiting dynamic behavior based on current environmental conditions. Whereas PathTrace largely performs a static, graph-based analysis, the AI agent was able to make decisions based on relative position in the facility, current conditions (was the adversarial agnet discovered?), and proximity to secondary targets. The agent demonstrated some novel capabilities, but had limitations that need to be resolved before it can be used for production purposes. For example, the adversarial agent generalizes poorly and takes a relatively long time to train. Nonetheless, there is still considerable promise for developing the adversarial agent further in order to explore even richer, more dynamic behaviors (e.g., adversary motivations, environmental debris, and more). This work considers a complementary idea; development of a planning agent. The planning agent is envisioned as an auto-complete-like tool that can help accelerate security system design by human experts. The agent would respect existing barriers and sensors placed by a human expert while offering cost-effective suggestions (i.e., implicitly balancing effectiveness with cost) to improve the design. The goal is for this agent to be part of an expert’s toolbox, not to totally upend the current state-of-practice, or to displace human experts. The ultimate goal would be concurrent training of both the adversarial and planning agent together, to learn entirely through self-play. This would represent an entirely new way of performing system deign. We selected a hierarchical, model-based reinforcement learning algorithm to serve as the planning agent. This is an extension of concepts used in the prior FY24 adversarial agent work. There, we had a single agent acting an environment. Here, we have two different sub-agents (policies), working together, to form a complete agent. There is a manager policy, which can select abstract goals on slower time scales, and a worker, which performs primitive actions to reach goals selected by the manager. It is worth noting that this class of algorithm is challenging to work with. From our understanding, our work is one of the first successful uses of model-based reinforcement learning (MBRL) in nuclear energy1 , and likely the first hierarchical model-based reinforcement learning application in nuclear energy. Further, this work is one of the first known attempts to apply AI to perform a design tasks in nuclear energy. Consequently, there were significant implementation challenges and the bulk of the work was focused on successful implementation and algorithm design. The results presented here are very low technology readiness level as a consequence of the lack of related literature, but still represent a significant step forward in the pursuit of applied AI for design.

42 ENGINEERING↗