Search NASA⌕ Search

SEARCH · Search NASA

Results for “learning dynamics”

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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Virtual to Physical: Reinforcement Learning to Optimize SNS Particle Accelerator Controls

Complex accelerators must have control systems that can handle dynamic nonlinear environments. This makes traditional control methods unsuitable as they can struggle to adapt to these uncertainties. This provides an ideal environment for reinforcement learning algorithms as they are adaptable and generalizable. We present a reinforcement learning pipeline that can effectively handle the dynamics of a complex accelerator. We test and prove our pipelines capabilities on multiple environments including the Spallation Neutron Source (SNS) and the Beam Test Facility (BTF) at Oakridge National Lab (ORNL). Due to the limited time available to train an online algorithm like reinforcement learning on a real accelerator, we utilize a virtual twin accelerator (VIRAC) developed by ORNL to pretrain the policy and show its ability to converge in the virtual environment. We then test the adaptability of the pretrained RL model by applying it on the real accelerator and comparing the results. Utilizing our Scientific Optimization and Controls Toolkit (SOCT) and open-source standards such as Gymnasium we create and solve for a MEBT orbit correction problem in the SNS and an emittance maximization problem in the BTF. We show how Twin Delayed Deep Deterministic Policy Gradient (TD3) can solve this optimization environment in the virtual accelerator and transfer this policy onto the real accelerator for inference and model retraining. We show how reinforcement learning can be utilized as a control system for complex accelerators and provide a model pipeline for how an implementation performs and can be adapted to new accelerator control problems.

Kasparian, Armen [Thomas Jefferson National Accele↗
DOE OSTI2025

Structural origin of disorder-induced ion conduction in NaFePO 4 cathode materials

Diffusion in NaFePO 4 can be enhanced through amorphization. Based on computations using DFT and machine learning potentials, we ascribe this phenomenon to the formation of less constrained Na-ion environments upon disordering. Most modern battery technologies depend on solid-state crystalline cathode materials. However, some of these materials are constrained by the low ionic conductivity of their most stable phases. An example of this is maricite (NaFePO 4 ). Interestingly, experiments have shown that maricite can improve its rate capability through disordering (amorphization). However, experimental characterization of amorphous cathode materials remains a major challenge, hindering a clear understanding of the structural origin of the disorder-induced improvement in sodium-ion mobility. To address this, we here employ molecular dynamics simulations by first training a machine learning potential for NaFePO 4 based on the atomic cluster expansion approach and a batch active learning potential parameterization scheme. This potential is then applied to explore the structural and dynamical properties of NaFePO 4 glasses as cathode materials. Specifically, we investigate the effect of glass structure on sodium-ion diffusion, revealing the relative influences of short-range and medium-range order features. We find significant heterogeneity in sodium-ion diffusivity in the glass, with fast-conducting ions residing in less constrained atomic environments with fewer P and Fe neighbors. These more mobile ions are also surrounded by larger ring-type structures. Overall, the results and developed approach present promising avenues for developing high-performance glassy cathodes for next-generation batteries.

Christensen, Rasmus↗
NASA NTRS2022

Exploring Applications of Machine Learning for Wildfire Monitoring and Detection using Unmanned Aerial Vehicles

Wildfires are increasing in frequency and severity around the world, including the United States. The losses caused by wildfires could be mitigated if high-risk areas, hotspots, and flare-ups could be monitored continuously, such as through the use of Unmanned Aerial Vehicles (UAVs). This paper documents exploratory efforts using machine learning to determine efficient flight paths for UAVs and to detect wildfires using image classification. On path planning, three machine learning techniques—Genetic Algorithm, Simulated Annealing, and Dynamic Programming—were explored. Genetic Algorithm was found to be an effective approach for path planning for wildfire monitoring and surveillance by UAVs. For a scenario of 25 locations in a circular arrangement, the algorithm was able to return the optimal path. The accuracy and execution time was found to be sensitive to the algorithm hyperparameters selected, which was especially evident in scenarios with hundreds or thousands of locations. Simulated Annealing was also found to be an effective approach for UAV path planning, with a major benefit of avoiding getting trapped in local minima and being straightforward to implement. Like Genetic Algorithm, the performance of Simulated Annealing was also found to be sensitive to the algorithm hyperparameters selected. By comparison, Dynamic Programming guarantees optimality for any number of locations, but it was found to be less practical in terms of execution time for scenarios with more than about a couple dozen locations. On wildfire detection, image classification using deep learning with a convolutional neural network was explored. Transfer learning was found to be a useful technique to efficiently train deep learning models. Also, it was determined that GPU processing can increase training speed by an order of magnitude, which enables significantly faster development. For a validation test set of 500 images, there were only two false negatives and zero false positives. These results demonstrate that detecting wildfires in static cameras using machine learning is feasible and establish a baseline for using images captured by UAVs in flight for wildfire detection.

Wildfire management↗
DOE OSTI2024

Hydrologic connectivity and dynamics of solute transport in a mountain stream: Insights from a long-term tracer test and multiscale transport modeling informed by machine learning

The movement of solutes in a watershed is a complex process with multiple interactions and feedbacks across spatial and temporal scales. Modeling the dynamics of solute transport along diverse hydrologic pathways within watersheds – from hillslopes to stream channels and in and out of the hyporheic zones – is challenging but critically important, as these processes integrate and contribute to the biogeochemical functioning of the river corridor up to the river network scale. Here we use results from a long-term network-scale tracer test at the H.J. Andrews experimental forest in western Cascade Mountains, Oregon, USA to inform a multiscale framework for transport in stream corridors. The framework uses a Lagrangian-based subgrid model to represent the effects of hyporheic exchange flow and advective transport at stream network scales. The spatially and temporally resolved stream discharge needed for the transport model is imputed across the river system by an entity-aware long short-term memory network. Modeled concentrations show good agreements with the observations and exhibit power scaling laws indicative of a very wide range of timescales over which hyporheic exchange flow occurs. Our results demonstrate a data-informed modeling framework that links dynamical processes occurring at small scales to a network context to help understand how changes at reach scale cascade into network-scale effects, providing a useful tool for sustainable river basin management.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Efficient machine learning interatomic potentials robust for liquid and multiple solid polymorphs of NaF and KF

Achieving atomic-level understanding of crystallization of molten salts is of importance to a wide range of technological applications. Recent work [Fan et al., Proc. Natl. Acad. Sci. USA 122, e2425702122 (2025)] revealed that crystal nucleation in molten LiF salt is a multistage process according to the molecular-dynamics (MD) simulations based on an atomic cluster expansion (ACE) machine-learning interatomic potential (MLIP). In order to understand the influence of increasing cation size on nucleation pathways and nucleation rates of molten fluoride salts, here we develop two new ACE MLIPs for NaF and KF. The two ACE MLIPs feature DFT-SCAN-level accuracy for liquid and multiple solid polymorphs over a wide temperature (0–2000 K) and pressure (0–100 GPa) range, and also reproduce well a number of experimental data for solid and liquid equilibrium properties. The efficiency of the two ACE MLIPs enable million-atom-scale or microsecond-scale MD simulations. The two general-purpose ACE MLIPs are expected to be useful for atomistic simulations for different purposes, in addition to studying crystallization of molten NaF and KF salts.

Crystal melting↗
DOE OSTI2026

Phase instability-coupled fracture behavior in garnet LLZO solid electrolytes: a machine learning-enabled atomistic study

Fracture in the garnet-type solid electrolyte Li 7 La 3 Zr 2 O 12 (LLZO) poses a critical threat to both the performance and safety of solid-state batteries. To unravel the coupled chemomechanical processes that govern fracture in LLZO under external loading, we carry out large-scale molecular-dynamics simulations with a validated machine-learning force field. Our results demonstrate that triaxial stresses at crack flanks trigger a localized cubic-to-tetragonal phase transformation, which is accompanied by Li-ion rearrangement. The emergent tetragonal domains feature lattice contraction normal to the fracture plane, imposing coherent misfit strains that provide an additional driving force for further crack propagation. Crucially, introducing Li deficiencies stabilizes the cubic phase, postponing the phase transition and thereby delaying fracture initiation. These findings highlight the role of intrinsic phase instability in dictating LLZO's fracture resistance and its critical connection to local Li concentration. This chemomechanical coupling points toward targeted strategies to enhance the mechanical robustness of garnet electrolytes, including tuning Li content, ensuring dopant homogeneity, and refining processing protocols.

Materials science↗
DOE OSTI2026

ECLEIRS: Exact conservation law embedded identification of reduced states for parameterized nonlinear conservation laws from sparse and noisy data

Multi-query applications such as parameter estimation, uncertainty quantification and design optimization for parameterized partial differential equation (PDE) systems are expensive. While reduced/latent state dynamics approaches for parameterized PDEs offer a viable alternative, these approaches rely on high-quality data and struggle with highly sparse spatiotemporal noisy measurements typically obtained from experiments. Furthermore, there is no guarantee that these models satisfy governing physical conservation laws. In this article, we propose a reduced state dynamics approach, referred to as ECLEIRS, that embeds exact conservation in the solution and flux representation by utilizing a space-time divergence-free neural network formulation. We compare ECLEIRS with other reduced state dynamics approaches, those that do not enforce any physical constraints and those with physics-informed loss functions, for three shock-propagation problems: 1-D advection, 1-D Burgers and 2-D Euler equations. In conclusion, the numerical experiments conducted in this study demonstrate that ECLEIRS provides the most accurate prediction of dynamics for unseen parameters even in the presence of highly sparse and noisy data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Direct Ab Initio Simulation of the Synthesis of BaZrO 3 and the Microstructure Impacts on Proton Transport

Controlling and predicting the processing-structure-performance relationship in functional materials is a grand challenge in materials science, with important implications for a wide range of emerging applications; a high fidelity understanding of the performance impact of microstructures formed under synthesis conditions is required to develop advanced materials, such as solid-state fuel cells and electrolyzers. Using the ceramic BaZrO 3 as a case study, we directly simulate the synthesis and investigate how proton transport is dictated by microstructures. We develop a framework that couples density functional theory (DFT), machine-learning interatomic potential (MLIP) driven molecular dynamics, and grand canonical Monte Carlo to perform large-scale, microstructure-resolved, atomistic simulations of proton transport in experimentally representative polycrystalline structures. Our fully ab initio approach, using a MLIP as a proxy for DFT, allows us to quantify the competition between two distinct diffusion mechanisms: one associated with grain-boundary regions and another within grains. When the impacts of grain boundaries are taken into account, proton transport exhibits substantial deviation from the bulk oxide limit. This addresses long-standing discrepancies between theory and experiments. Our integrated approach provides atomistic insight into microstructure-dependent proton pathways in BaZrO 3 and establishes a general protocol for predicting processing-structure-performance relationships.

organic↗
DOE OSTI2024

Neuromorphic intermediate representation: A unified instruction set for interoperable brain-inspired computing

Abstract Spiking neural networks and neuromorphic hardware platforms that simulate neuronal dynamics are getting wide attention and are being applied to many relevant problems using Machine Learning. Despite a well-established mathematical foundation for neural dynamics, there exists numerous software and hardware solutions and stacks whose variability makes it difficult to reproduce findings. Here, we establish a common reference frame for computations in digital neuromorphic systems, titled Neuromorphic Intermediate Representation (NIR). NIR defines a set of computational and composable model primitives as hybrid systems combining continuous-time dynamics and discrete events. By abstracting away assumptions around discretization and hardware constraints, NIR faithfully captures the computational model, while bridging differences between the evaluated implementation and the underlying mathematical formalism. NIR supports an unprecedented number of neuromorphic systems, which we demonstrate by reproducing three spiking neural network models of different complexity across 7 neuromorphic simulators and 4 digital hardware platforms. NIR decouples the development of neuromorphic hardware and software, enabling interoperability between platforms and improving accessibility to multiple neuromorphic technologies. We believe that NIR is a key next step in brain-inspired hardware-software co-evolution, enabling research towards the implementation of energy efficient computational principles of nervous systems. NIR is available atneuroir.org

Science & Technology - Other Topics↗
DOE OSTI2025

Nanoscale wetting controls reactive Pd ensembles in synthesis of dilute PdAu alloy catalysts

The performance of bimetallic dilute alloy catalysts is largely determined by the size of minority metal ensembles on the nanoparticle surface. By analyzing the synthesis of catalysts comprising Pd 8 Au 92 nanoparticles supported on silica using surface-sensitive techniques, we report that whether Pd overgrowth occurs before or after Au nanoparticle deposition onto the support controls the surface Pd ensemble size and abundance. These differences in Pd ensembles influence catalytic reactivity in H 2 –D 2 isotope exchange and benzaldehyde hydrogenation, which, in correlation with theoretical calculations, is used to elucidate the active site(s) in each reaction. To clarify how the synthetic sequence controls the formation of Pd ensembles, we combine numerical wetting calculations and molecular dynamics simulations (with a machine-learned force field) to visualize Pd deposition and migration on the nanoparticle surface, respectively. Our results suggest that the nanoparticle–support interface restricts nanoparticle accessibility to Pd deposition, which consequently controls the Pd ensemble size, illustrating the critical role of nanoscale wetting phenomena during bimetallic catalyst preparation.

36 MATERIALS SCIENCE↗
DOE OSTI2025

MultiTaskDeltaNet: change detection-based image segmentation for operando ETEM with application to carbon gasification kinetics

Transforming in situ transmission electron microscopy (TEM) imaging into a tool for spatially-resolved operando characterization of solid-state reactions requires automated, high-precision semantic segmentation of dynamically evolving features. However, traditional deep learning methods for semantic segmentation often face limitations due to the scarcity of labeled data, visually ambiguous features of interest, and scenarios involving small objects. To tackle these challenges, we introduce MultiTaskDeltaNet (MTDN), a novel deep learning architecture that creatively reconceptualizes the segmentation task as a change detection problem. By implementing a unique Siamese network with a U-Net backbone and using paired images to capture feature changes, MTDN effectively leverages minimal data to produce high-quality segmentations. Furthermore, MTDN utilizes a multi-task learning strategy to exploit correlations between physical features of interest. In an evaluation using data from in situ environmental TEM (ETEM) videos of filamentous carbon gasification, MTDN demonstrated a significant advantage over conventional segmentation models, particularly in accurately delineating fine structural features. Notably, MTDN achieved a 10.22% performance improvement over conventional segmentation models in predicting small and visually ambiguous physical features. This work bridges key gaps between deep learning and practical TEM image analysis, advancing automated characterization of nanomaterials in complex experimental settings.

08 HYDROGEN↗
DOE OSTI2026

2D nanoconfinement distorts the solvation structure of hydroxide but not of hydronium

Understanding ion-specific behavior in nanoconfined water is essential for controlling charge transport and selectivity in two-dimensional membranes. Motivated by recent experiments revealing anomalous dielectric and transport behavior of water confined between hBN sheets, we use machine-learning-accelerated first-principles molecular dynamics to investigate the interfacial propensities of hydronium and hydroxide ions under similar confinement. We find that hydronium remains interfacial across all confinement regimes, whereas hydroxide shifts toward the interior as the environment becomes more bulk-like. This contrasting behavior reflects the combined influence of hydrogen bonding, interfacial water layering, and the polarization of hBN, which collectively stabilize hydronium at the surface while making hydroxide slightly more favorable within the structured interior. These findings expose an asymmetry in ion-surface coupling and establish a microscopic origin for hydronium’s enhanced interfacial affinity. The results provide mechanistic insight into ion partitioning in two-dimensional channels and highlight the collective structuring of confined water as a key determinant of interfacial ion thermodynamics.

organic↗
DOE OSTI2024

Generalized representative structures for atomistic systems

A new method is presented to generate atomic structures that reproduce the essential characteristics of arbitrary material systems, phases, or ensembles. Previous methods allow one to reproduce the essential characteristics (e.g. the chemical disorder) of a large random alloy within a small crystal structure. The ability to generate small representations of random alloys, along with the restriction to crystal systems, results from using the fixed-lattice cluster correlations to describe structural characteristics. A more general description of the structural characteristics of atomic systems is obtained using complete sets of atomic environment descriptors. These are used within for generating representative atomic structures without restriction to fixed lattices. A general data-driven approach is provided here utilizing the atomic cluster expansion (ACE) basis. The N-body ACE descriptors are a complete set of atomic environment descriptors that span both chemical and spatial degrees of freedom and are used within for describing atomic structures. The generalized representative structure (GRS) method presented within generates small atomic structures that reproduce ACE descriptor distributions corresponding to arbitrary structural and chemical complexity. It is shown that systematically improvable representations of crystalline systems on fixed parent lattices, amorphous materials, liquids, and ensembles of atomic structures may be produced efficiently through optimization algorithms. With the GRS method, we highlight reduced representations of atomistic machine-learning training datasets that contain similar amounts of information and small 40–72 atom representations of liquid phases. The ability to use GRS methodology as a driver for informed novel structure generation is also demonstrated. The advantages over other data-driven methods and state-of-the-art methods restricted to high-symmetry systems are highlighted.

atomic cluster expansion↗
DOE OSTI2024

Characterization of the astrophysical diffuse neutrino flux using starting track events in IceCube

In this article, a measurement of the diffuse astrophysical neutrino spectrum is presented using IceCube data collected from 2011-2022 (10.3 years). We developed novel detection techniques to search for events with a contained vertex and exiting track induced by muon neutrinos undergoing a charged-current interaction. Searching for these starting track events allows us to not only more effectively reject atmospheric muons but also atmospheric neutrino backgrounds in the southern sky, opening a new window to the sub-100 TeV astrophysical neutrino sky. The event selection is constructed using a dynamic starting track veto and machine learning algorithms. We use this data to measure the astrophysical diffuse flux as a single power law flux (SPL) with a best-fit spectral index of γ=2.58$_{-0.09}^{+0.10}$ and per-flavor normalization of $\phi$$_{per-flavor}^{Astro}$=1.68$_{-0.22}^{+0.19}$×10 -18 ×GeV -1 cm -2 s -1 sr -1 (at 100 TeV). The sensitive energy range for this dataset is 3-550 TeV under the SPL assumption. This data was also used to measure the flux under a broken power law, however we did not find any evidence of a low energy cutoff.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Application of Modified Meshgraphnets for Subsurface Prediction during CO2 Sequestration

In the face of the increasingly dire consequences of anthropogenic climate change, capturing and storing carbon dioxide is paramount. However, several impediments exist to the safe and effective subsurface storage of CO2, such as cost of transport, identification of suitable sites for subsurface storage, and assessment of long-term risk from storage in subsurface aquifers. Accurate subsurface modeling is necessary to ensure that CO2 storage is both safe and effective. Still, such modeling has traditionally required either substantial time and computational power (numerical simulation) or a substantial amount of pre-existing data for training (machine learning models). Additionally, these models lack flexibility in dealing with both changes in discretization of the input data and generalizability beyond the data on which they are trained. In order to address these issues, this research applies graph neural networks (GNNs) to predict subsurface saturation and pressure during CO₂ injection in a model of the Illinois Basin-Decatur Project (IBDP). GNNs provide a flexible, intuitive method for representing and manipulating complex unstructured data, which is often found in many practical domain problems such as fluid flow and subsurface characterization. These unstructured grids are easily represented in GNNs by representing spatially-localized features such as permeability, porosity, saturation, and pressure as nodes in a graph and relationships between these properties as edges connecting these nodes. This research applies a specific GNN model called MeshGraphNets (MGN) to model the change in CO2 saturation and pressure over a 50-month time period (36 months of injection, 14 months post-injection). The MGN model leverages a message passing process that allows the network to learn both the spatial and temporal dynamics of this system simultaneously. Additionally, training on a limited dataset (64 realizations, 20 time points each) resulted in a high degree of accuracy in saturation prediction both within the same timeframe as the training (20 months, 0.039 average RMSE) and when projecting out to the end of injection (36 months, 0.053 average RMSE). Temporal predictions such as those generated by MGNs and other similar models are prone to accumulated error over time; in order to address this, a multi-step rollout (MSR) training process was applied to calculate training loss. This method mimics the forward prediction during inference by “rolling out” multiple time points in a single training step using the previous prediction as input to the MGN model. By calculating the loss several time steps forward from the current prediction, the model is forced to find a more stable state over time. Application of MSR to the MGN model resulted in an average 15% reduction in inference error over time during forward prediction. This study showcases the immense potential of GNNs as a game-changing methodology for predicting pressure and saturation evolution in CCS projects, ultimately paving the way for more sustainable and effective carbon storage solutions. Presentation prepared for the 2024 AiChE Annual Meeting, October 27 to November 1 2024, San Diego, CA.

Holcomb, Paul↗
DOE OSTI2024

Rational Design of High-Performing Electrodes in Energy Storage Devices

Rational design of interfaces with proper physico-chemical properties is necessary to optimize the performance of electrochemical devices, which requires fundamental understanding of the interfaces. In this project, we leveraged quantum mechanical and molecular dynamics simulations as well as machine learning (ML) technique to correlate the relationships between structure and properties including ion transport and electronic structures of cathode host materials in lithium-sulfur (Li-S) and carbon anode in sodium (Na) batteries for improved electrochemical performance.

25 ENERGY STORAGE↗
DOE OSTI2026

Digital Twin User Guide for Chelan County Public Utility District

This user manual offers a comprehensive guide for developing a Digital twin (DT) of a Kaplan turbine at Chelan County Public Utility District (Chelan PUD) using neural networks. As variable renewable generation expands, hydropower units must operate with optimal efficiency and stability. For Kaplan machines, this flexibility is achieved through coordinated control of guide vane (wicket gates) opening and runner blade pitch, which amplifies the plant’s inherent nonlinear behavior and challenges traditional physics-only modeling. The efficiency of the Kaplan turbine varies with different combinations of the guide vans (wicket gate) opening and the blade angle. Each guide van opening and blade angle has a corresponding highest efficiency point, forming a cam relationship that represents the optimal combination.The discharge of a hydraulic turbine is controlled by the opening angle of the guide vans. Therefore, for each value of head, there is a certain guide van opening and blade angle that corresponds to the highest efficiency. For a given head, different combinations of the guide van opening and blade angle have different efficiencies. Therefore, coordinate cam curves are used to describe the relationship between the wicket gate opening and blade angle with different water head. To address these challenges, the manual details a data-driven modeling and learning workflow centered on structured neural networks. The approach is designed to forecast critical operational variables—discharge flow, net head, penstock (or scroll-case) pressure, and generator electrical outputs—by leveraging real-time inputs such as the generator power control setpoint, exciter field current and field voltage, together with hydromechanical commands (e.g., gate position and, when available, runner blade-pitch angle). The neural models are trained and validated on operational data from a Kaplan unit operated by Chelan PUD, demonstrating that the structured NN architecture can learn the coupled gate–blade–electrical dynamics. The result is a robust DT that improves situational awareness and supports data-informed decision-making for Chelan PUD’s Kaplan turbine operations.

24 POWER TRANSMISSION AND DISTRIBUTION↗