Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Network Models”

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

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

At least 1,351 records · Page 75
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Learning viscoelasticity models from indirect data using deep neural networks

In this study, we propose a novel approach to model viscoelasticity materials, where rate-dependent and non-linear constitutive relationships are approximated with deep neural networks. We assume that inputs and outputs of the neural networks are not directly observable, and therefore common training techniques with input–output pairs for the neural networks are inapplicable. To that end, we develop a novel computational approach to both calibrate parametric and learn neural-network-based constitutive relations of viscoelasticity materials from indirect displacement data in the context of multiple-physics systems. We show that limited displacement data holds sufficient information to quantify the viscoelasticity behavior. We formulate the inverse computation – modeling viscoelasticity properties from observed displacement data – as a PDE-constrained optimization problem and minimize the error functional using a gradient-based optimization method. The gradients are computed by a combination of automatic differentiation and implicit function differentiation rules. The effectiveness of our method is demonstrated through numerous benchmark problems in geomechanics and porous media transport.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Improved Gas Plume Identification Using Nearest Neighbor Methods for Background Estimation

Longwave infrared (LWIR) hyperspectral imaging (HSI) can be used for many tasks in remote sensing, including detecting and identifying effluent gases by LWIR sensors on airborne platforms. Identification is used after detection to increase confidence in weakly detected plumes, reduce false positives from detection, and distinguish between similar and confounding material signatures. Background estimation is an important step used to reveal the unique spectral characteristics of the detected gas, allowing the identification model to determine what the gas is specifically. The importance of proper background estimation increases when dealing with weak signals, large libraries of gases of interest, and uncommon or heterogeneous backgrounds. In this article, we propose two methods for background estimation: a novel k-nearest segments (KNS) algorithm and the standard k-nearest neighbors (KNN) algorithm. We test our methods and three existing background estimation methods for comparison against global background estimation to determine which performs best at estimating the true background radiance under a plume and for increasing identification confidence using a neural network classification model. We compare the different methods using 640 simulated weak plumes in an urban environment. For identification, our KNS algorithm improves median neural network identification confidence by 53.2%. For background radiance estimation, the KNN algorithm provides a median of 49 times less RMSE than global background estimation. Furthermore, KNN is the easiest method to tune for different plumes, making it an excellent “out of the box” background estimator.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

Dynamics of particle network in composite battery cathodes

We report that improving composite battery electrodes requires a delicate control of active materials and electrode formulation. The electrochemically active particles fulfill their role as energy exchange reservoirs through interacting with the surrounding conductive network. We formulate a network evolution model to interpret the regulation and equilibration between electrochemical activity and mechanical damage of these particles. Through statistical analysis of thousands of particles using x-ray phase contrast holotomography in a LiNi 0.8 Mn 0.1 Co 0.1 O 2 -based cathode, we found that the local network heterogeneity results in asynchronous activities in the early cycles, and subsequently the particle assemblies move toward a synchronous behavior. Our study pinpoints the chemomechanical behavior of individual particles and enables better designs of the conductive network to optimize the utility of all the particles during operation.

25 ENERGY STORAGE↗
NASA NTRS1982

Orbital performance of a one-year lifetime superfluid helium dewar based on ground testing and computer modelling

The ground-test results of the 550-liter superfluid-He dewar built for IRAS are presented and discussed, and a computer projection of its orbital performance life is developed. The ground-testing program included flow-swap, cold-flow, power-off, hot/cold-viewport, and strap-thermal-efficiency tests. After one set of trials, some design changes were made, and the thermal-network computer model of the system was modified as well. The driving boundary conditions of dewar orbital operation used in this model are vacuum-shell temperature = 170 K, aperture heating = 8.0 mW, and focal-plane power dissipation = 16.2 mW. An operational lifetime of 319 days and an He loss rate of 2.54 mg/sec are predicted.

Hopkins, R. A.↗
DOE OSTI2023

Towards POI-based large-scale land use modeling: spatial scale, semantic granularity, and geographic context

The combination of spatial distribution, semantic characteristics, and sometimes temporal dynamics of POIs inside a geographic region can capture its unique land use characteristics. Most previous studies on POI-based land use modeling research focused on one geographic region and select one spatial scale and semantic granularity for land use characterization. There is a lack of understanding on the impact of spatial scale, semantic granularity, and geographic context on POI-based land use modeling, particularly large-scale land use modeling. In this study, we developed a scalable POI-based land use modeling framework and examined the impact of these three factors on POI-based land use characterization using data from three geographic regions. We developed a unified semantic representation framework for POI semantics that can help fuse heterogeneous POI data sources. Then, by combining POIs with a neural network language model, we developed a spatially explicit approach to learn the embedding representation of POIs and AOIs. We trained multiple supervised classifiers using AOI embeddings as input features to predict AOI land use at different semantic granularities. The classification performance of different land use classes was analyzed and compared across three geographic regions to identify the semantic representativeness of POI-based AOI embedding and the impact of geographic context.

58 GEOSCIENCES↗
DOE OSTI2023

Lorentz group equivariant autoencoders

Abstract There has been significant work recently in developing machine learning (ML) models in high energy physics (HEP) for tasks such as classification, simulation, and anomaly detection. Often these models are adapted from those designed for datasets in computer vision or natural language processing, which lack inductive biases suited to HEP data, such as equivariance to its inherent symmetries. Such biases have been shown to make models more performant and interpretable, and reduce the amount of training data needed. To that end, we develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant with respect to the proper, orthochronous Lorentz group $$\textrm{SO}^+(3,1)$$ SO + ( 3 , 1 ) , with a latent space living in the representations of the group. We present our architecture and several experimental results on jets at the LHC and find it outperforms graph and convolutional neural network baseline models on several compression, reconstruction, and anomaly detection metrics. We also demonstrate the advantage of such an equivariant model in analyzing the latent space of the autoencoder, which can improve the explainability of potential anomalies discovered by such ML models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Ecosystems and Networks Integrated with Genes and Molecular Assemblies (ENIGMA) (Final Scientific/Technical Report, Subcontract Award No. 6953691)

The overarching objective the Baliga lab is to develop a mechanistic understanding of the field relevant metabolic processes underlying community partitioning of respiration pathways and physiological state shifts. In order to dissect how environmental perturbations influence microbial regulatory networks, we have constructed data-driven models that capture the dynamic interplay between abiotic and biotic factors during laboratory simulations. Specifically, we have developed multiple computational models (i.e., cMonkey2, EGRIN 2.0, and multiple genome-scale metabolic models) connecting environmental influence to genome-encoded regulatory programs. In addition, we have generated new tools such as Live Anaerobic Cell Sorting (LAnCS) and Fluidized Bed Reactors (FBRs) to investigate the dynamics of community partitioning of metabolic processes supporting sulfate or nitrate respiration, which are two critical activities observed from the field site at Oak Ridge National Labs (ORNL). This framework of connecting field phenomena with laboratory simulations and computational modeling has created the Environmental Simulations and Modelling (EnvSim) campaign within ENIGMA. The campaign, which spans multiple laboratories across ENIGMA and is led by the Baliga group, has established synthetic communities (SynComs) to discover, characterize, and dissect key microbial processes relevant to field observations, such as emissions of the greenhouse gas nitrous oxide (N 2 O). For example, the EnvSim campaign so far has deduced four potential mechanisms that may account for the N 2 O emissions at the FRC and are currently being investigated by labs across ENIGMA. Denitrification may be driven by complete denitrifiers, however, their NosZ enzymes, which catalyze the final step in denitrification by converting N 2 O to N 2 , may be sensitive at a lower pH. Additionally, the denitrification process could be partitioned among organisms and some may have pH-sensitive NosZ genes. Another factor contributing to variable N 2 O emissions relates to the metal co-factors involved in the denitrification pathway. For instance, excess Cu, Al, Mn, U, Ni, Co, Cu, and/or Cd may have inhibitory effects on multiple enzymatic steps during denitrification, while the enzymatic production of nitrite, the precursor of N 2 O, can be limited by the essential metal Mo. Lastly, abiotic production of N 2 O may occur as a result of chemodenitrification, in which metals like Fe, Mn, and some organic compounds can drive redox reactions that convert nitrogen cycle intermediates to N 2 O under the right conditions. Current investigations and analyses have generated multiple transcriptomic profiles aiding the refinement of new metabolic and gene regulatory network model for our denitrifying SynCom. In sum, the previous funding cycle has generated 33 peer-reviewed publications spanning predictive network biology to co-operativity of mutualistic interspecies interactions of evolved communities. Our integration of multiple omics datasets and the development of new network modelling tools has revealed important insights that can explain field observed phenomenon and generated new hypotheses to be investigated in the field.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Semi-supervised Learning of Dynamical Systems with Neural Ordinary Differential Equations: A Teacher-Student Model Approach

Modeling dynamical systems is crucial for a wide range of tasks, but it remains challenging due to complex nonlinear dynamics, limited observations, or lack of prior knowledge. Recently, data-driven approaches such as Neural Ordinary Differential Equations (NODE) have shown promising results by leveraging the expressive power of neural networks to model unknown dynamics. However, these approaches often suffer from limited labeled training data, leading to poor generalization and suboptimal predictions. On the other hand, semi-supervised algorithms can utilize abundant unlabeled data and have demonstrated good performance in classification and regression tasks. We propose TS-NODE, the first semi-supervised approach to modeling dynamical systems with NODE. TS-NODE explores cheaply generated synthetic pseudo rollouts to broaden exploration in the state space and to tackle the challenges brought by lack of ground-truth system data under a teacher-student model. TS-NODE employs an unified optimization framework that corrects the teacher model based on the student's feedback while mitigating the potential false system dynamics present in pseudo rollouts. TS-NODE demonstrates significant performance improvements over a baseline Neural ODE model on multiple dynamical system modeling tasks.

Wang, Yu↗
NASA NTRSDate not supplied

Early Information Parameter-Set Analysis for Satellite Close Approaches using Machine Learning

Understanding orbital mechanics is essential in space flight and navigation applications, and leveraging modern force models for flight path projection remains an important aspect in space mission design and operation. However, force models do not capture all the dynamics or perturbations in the space environment and thus are subject to errors in predicting the state vectors. The further out the predicted miss distance between spacecraft is from the time of closest approach (TCA), the larger the propagated errors in the predicted miss distance at TCA is. The dependency on these force models for spacecraft flight state prediction calls for a more reliable method that can quantify, or even reduce, these propagated errors. With recent advances in the field artificial intelligence, specifically in machine and deep learning algorithms, a model that implements these approaches can improve on the modern force model approach. The goal for this work is to provide an early-information decision-making threshold, in order to prioritize risk assessment implementation, given the ongoing increase of space objects. In analyzing the relationship of several parameters from conjunction data messages(CDMs) and solar information, early information becomes viable in miss distance prediction with unsupervised learning techniques, which learn the parameters that are linked together with miss distance and probability of collision (Pc) variables. Another approach implemented for identifying relationships within CDMs is supervised learning, in which a shallow neural network binary classifier learns to distinguish events with Pc values¡108. These parameters detected in the unsupervised process are then applied to a regression neural network, which predicts the miss distance at TCA for a specific event within a given uncertainty bound. For the regression neural network, a Long Short Term Memory (LSTM) neural network is implemented, which yields memory about each time step in an event. Using an LSTM network, the model learns to predict miss distance within 0.2km of the value measured at TCA. Although there is a limited amount of "close miss" data to train a network, the network learns to associate parameters, like large energy dissipation rates with the secondary object, with an elevated Pc

Brianna I. Robertson↗
DOE OSTI2023

Machine Learning Predictions of Simulated Self-Diffusion Coefficients for Bulk and Confined Pure Liquids

Diffusion properties of bulk fluids have been predicted using empirical expressions and machine learning (ML) models, suggesting that predictions of diffusion also should be possible for fluids in confined environments. The ability to quickly and accurately predict diffusion in porous materials would enable new discoveries and spur development in relevant technologies such as separations, catalysis, batteries, and subsurface applications. Here in this work, we apply artificial neural network (ANN) models to predict the simulated self-diffusion coefficients of real liquids in both bulk and pore environments. The training data sets were generated from molecular dynamics (MD) simulations of Lennard-Jones particles representing a diverse set of 14 molecules ranging from ammonia to dodecane over a range of liquid pressures and temperatures. Planar, cylindrical, and hexagonal pore models consisted of walls composed of carbon atoms. Our simple model for these liquids was primarily used to generate ANN training data, but the simulated self-diffusion coefficients of bulk liquids show excellent agreement with experimental diffusion coefficients. ANN models based on simple descriptors accurately reproduced the MD diffusion data for both bulk and confined liquids, including the trend of increased mobility in large pores relative to the corresponding bulk liquid.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1982

Photochemical processes in the inner coma

Models for the inner-coma chemistry of comets are reviewed. The physics relevant to the coma's chemistry is summarized, and the interaction of solar radiation with the coma is described, along with photolytic and chemical processes. The formation and destruction of several observed species are traced through a chemical reaction network, and model results are compared with observations. The species considered include CN, C2, C3, NH2, CH, CO, CO(+), OH, H2O, HCO, C2H4, and C2H3. The models most consistent with observations are shown to indicate that only trace amounts (2% in all) of molecules bearing CN, C2, C3, and NH2 can be present in the nucleus of a comet.

Huebner, W. F.↗
DOE OSTI2025

Understanding latent timescales in neural ordinary differential equation models of advection-dominated dynamical systems

The neural ordinary differential equation (ODE) framework has shown considerable promise in recent years in developing highly accelerated surrogate models for complex physical systems characterized by partial differential equations (PDEs). For PDE-based systems, state-of-the-art neural ODE strategies leverage a two-step procedure to achieve this acceleration: a nonlinear dimensionality reduction step provided by an autoencoder, and a time integration step provided by a neural-network based model for the resultant latent space dynamics (the neural ODE). This work explores the applicability of such autoencoder-based neural ODE strategies for PDEs in which advection terms play a critical role. More specifically, alongside predictive demonstrations, physical insight into the sources of model acceleration (i.e., how the neural ODE achieves its acceleration) is the scope of the current study. Such investigations are performed by quantifying the effects of both autoencoder and neural ODE components on latent system time-scales using eigenvalue analysis of dynamical system Jacobians. To this end, the sensitivity of various critical training parameters – de-coupled versus end-to-end training, latent space dimensionality, and the role of training trajectory length, for example – to both model accuracy and the discovered latent system timescales is quantified. Furthermore, this work specifically uncovers the key role played by the training trajectory length (the number of rollout steps in the loss function during training) on the latent system timescales: larger trajectory lengths correlate with an increase in limiting neural ODE time-scales, and optimal neural ODEs are found to recover the largest time-scales of the full-order (ground-truth) system. Demonstrations are performed across fundamentally different unsteady fluid dynamics configurations influenced by advection: (1) the Kuramoto–Sivashinsky equations (2) Hydrogen-Air channel detonations (the compressible reacting Navier–Stokes equations with detailed chemistry), and (3) 2D Atmospheric flow.

Advection-dominated dynamical systems↗
DOE OSTI2026

ML-based Micro-CT SOFC Microstructure Models (from Kent 2026 Microstructural Augmentation paper)

Overview -------------------------- This repository contains datasets from the manuscript **"Enhanced Generalizability to Deep-Learning Quantification of 3D Microstructural Characteristics through Microstructurally Aware Augmentation of Scarce Data"** (*William F. Kent, Rochan Bajpai, Rachel C. Kurchin, William K. Epting, Harry W. Abernathy, Paul A. Salvador. Submitted 2026*). The methods are also described in the dissertation **Data Intensive Analysis of Solid Oxide Cell Microstructures** (*Doctoral dissertation, Carnegie Mellon University, 2025*). The datasets here are trained convolutional neural network (CNN) models for predicting key microstructural properties of solid oxide cell (SOC) electrodes from low-res, 2-channel 3D images, as well as some helpful code. The parameters for input images are provided in the paper. Sample data is provided in the file `Combined_anode_aug_dual_1k_examples` - that particular data was used to train `anode_all_aug.pth` and will work most accurately with that model. Please familiarize yourself with all caveats on accuracy and applicability, as detailed in the associated paper. Usage -------------------------- The basic usage is as follows, assuming `model_fn` is the path to the .pth file, and `X` is 2-channel input image(s) of the proper dimensions (either one image of shape `[2,12,24,24]`, or a batch of N input images of shape `[N,2,12,24,24]`): from CNN_inferencer import load_model_for_inference model = load_model_for_inference(model_fn) y_predicted = model(X) The model object automatically handles input scaling and output de-scaling based on the way the models were trained - in other words, pass in a 2-channel micro-CT image, and it will output microstructural property values in real units. ## Other model object attributes Note that model has useful attributes other than its forward pass model(X). * `model.output_descaler` - returns the output descaler object. Model does the de-scaling when generating inferences, but you may want to re-use this de-scaler on other values to e.g. compare predictions to ground truth from already-scaled training data. * `model.prop_names` - Gives the property names of the predicted y values, in order. Only exists if there's an output scaler as part of the model object, which there will be in the models provided here. ## Usage with sample data Here is a short script to use with the included sample data. from CNN_inferencer import display_predictions, load_model_for_inference, calculate_mape, parity_plot import h5py import numpy as np model_fn = 'anode_all_aug.pth' data_fn = 'Combined_anode_aug_dual_1k_examples.h5' N_samples = 200 figure_outdir = '.' model = load_model_for_inference(model_fn) with h5py.File(data_fn,'r') as f: XX = f['X'] #These are the 2-channel 3D images yy = f['y'] #These are the ground-truth microstructural properties, but they have been scaled for training - need to de-scale below N = XX.shape[0] #How many images total in the input data file #Run inferences on N_samples random samples from XX. #Run in a batch, much more efficient than one at a time. ii = np.random.choice(N,N_samples,replace=False) ii.sort() y_pred = model(XX[ii]) #Get the original/true (but normalized/scaled) values from the training dataset... #Because they were normalized, they are not in real units yet. So let's also de-scale them using model.output_scaler. y_true = model.output_scaler.transform(yy[ii]) #Let's display actual values for just 5 random ones for i in np.random.choice(N_samples,5,replace=False): display_predictions(y_true[i], y_pred[i], model.prop_names) #Make parity plots for each property (ground truth vs predicted values) #Also label each plot with the mean abs. percent error (MAPE) of the predicted values for i,key in enumerate(model.prop_names): mape = calculate_mape(y_true[:,i], y_pred[:,i]) parity_plot(y_true[:,i], y_pred[:,i], figure_outdir, key, extra_title=f' ({mape:.2f}% MAPE)')

3D microstructure↗
DOE OSTI2021

Crosslinked Matrimid®-like polyimide membranes with unimodal network structure for enhanced stability and gas separation performance

Gas separation membranes have attracted academic and industrial attention, and crosslinking has been identified to be one of the most effective ways to enhance membrane stability. In this paper, a series of crosslinked Matrimid®-like films with unimodal network structures are prepared via thermally end-linking phenylethnyl-terminated BTDA-DAPI oligomers with well-controlled molecular weight (i.e., 3000–15,000 g/mol), wherein the crosslink density (the inter-crosslink chain length) of resulting unimodal networks is systematically varied by using oligomers with various molecular weight. Comprehensive characterizations of chemical structure, thermal properties, microstructures are performed. Pure-gas permeation measurements are performed focusing on H 2 /CH 4 and CO 2 /CH 4 separations as a function of crosslink density. In sharp contrast to the commonly observed permeability reduction in randomly crosslinked networks, all the crosslinked unimodal films, even when densely crosslinked, present markedly enhanced permeability and well-maintained ideal selectivity relative to the uncrosslinked linear counterpart, leading to almost horizontal movements towards upper bounds along with expectedly enhanced membrane stability. In conclusion, it is concluded that introducing bulky groups at the crosslink sites provide a practical means to counteract the densification effect induced by crosslinking and the construction of unimodal networks exemplifies a fundamentally new strategy to regulate the microstructure and property of crosslinked membranes for gas separations.

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

Freely scalable and reconfigurable optical hardware for deep learning

Abstract As deep neural network (DNN) models grow ever-larger, they can achieve higher accuracy and solve more complex problems. This trend has been enabled by an increase in available compute power; however, efforts to continue to scale electronic processors are impeded by the costs of communication, thermal management, power delivery and clocking. To improve scalability, we propose a digital optical neural network (DONN) with intralayer optical interconnects and reconfigurable input values. The path-length-independence of optical energy consumption enables information locality between a transmitter and a large number of arbitrarily arranged receivers, which allows greater flexibility in architecture design to circumvent scaling limitations. In a proof-of-concept experiment, we demonstrate optical multicast in the classification of 500 MNIST images with a 3-layer, fully-connected network. We also analyze the energy consumption of the DONN and find that digital optical data transfer is beneficial over electronics when the spacing of computational units is on the order of $$>10\,\upmu $$ > 10 μ m.

42 ENGINEERING↗
NASA NTRSDate not supplied

Flight and Direct to Earth/Space Relay Communication System Architecture for GSFC CubeSat Missions

The CubeSat platform is finding increasing use in space science applications due to its low cost and comparative ease of launch. It is becoming a key scientific discovery tool in low Earth orbit (LEO) and beyond, including geosynchronous equatorial orbit (GEO), the Lagrange Points, Lunar missions, and more. The increasing complexity of these missions and their scientific goals must be supported by equal advancements in communications technology. Higher data rates and greater reliability are required every year. However, the reduced Size, Weight, and Power (SWaP) constraints of CubeSat platforms introduce unique challenges in the area of satellite communications. There is currently a lack of communication equipment tailored specifically to the CubeSat platform. This lack of standardized, tested equipment extends development time and reduces mission confidence. Furthermore, missions utilizing the CubeSat platform are often subject to more difficult design constraints. Antenna placement, size, and pointing are often subordinate to the requirements of the payload instruments and mission goals. Traditional link margin estimation techniques are insufficient in these cases, as they emphasize worst case scenarios. In reality the actual link parameters may vary widely even during a single pass. This presents new challenges in predicting communications performance and scheduling ground station contacts, but also new opportunities for improving efficiency. This paper presents the integration, testing, and validation process for a new software defined radio (SDR) designed for the CubeSat platform in conjunction with Vulcan Wireless, Inc. The SDR is planned for use on 5 upcoming CubeSat missions at NASAs Goddard Space Flight Center (GSFC) including a Geosynchronous Transfer Orbit (GTO) mission and it may also serve as a standard and well-tested option for future missions by enabling a standardized, rapid and low cost CubeSat communication system network integration process. Detailed simulations have been developed to estimate the communication performance of these missions, taking the unique antenna placements and attitude behavior of each satellite into account. These simulations allow a much more accurate analysis of the expected link margin, which varies considerably during each pass for the NASA Space Relay (SR) and Direct to Earth (DTE) network. The modelling procedures are outlined, and the results are used to predict communications performance of the missions.

Space Networks↗
DOE OSTI2021

Machine Learning Self-Diffusion Prediction for Lennard-Jones Fluids in Pores

Predicting the diffusion coefficient of fluids under nanoconfinement is important for many applications including the extraction of shale gas from kerogen and product turnover in porous catalysts. Due to the large number of important variables, including pore shape and size, fluid temperature and density, and the fluid–wall interaction strength, simulating diffusion coefficients using molecular dynamics (MD) in a systematic study could prove to be prohibitively expensive. Here, we use machine learning models trained on a subset of MD data to predict the self-diffusion coefficients of Lennard-Jones fluids in pores. Our MD data set contains 2280 simulations of ideal slit pore, cylindrical pore, and hexagonal pore geometries. We use the forward feature selection method to determine the most useful features (i.e., descriptors) for developing an artificial neutral network (ANN) model with an emphasis on easily acquired features. Our model shows good predictive ability with a coefficient of determination (i.e., R 2 ) of ~0.99 and a mean squared error of ~2.9 × 10 –5 . Finally, we propose an alteration to our feature set that will allow the ANN model to be applied to nonideal pore geometries.

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

Machine learning force field model for kinetic Monte Carlo simulations of itinerant Ising magnets

Here, we present a scalable machine learning (ML) framework for large-scale kinetic Monte Carlo (kMC) simulations of itinerant electron Ising systems. As the effective interactions between Ising spins in such itinerant magnets are mediated by conducting electrons, the calculation of energy change due to a local spin update requires solving an electronic structure problem. Such repeated electronic structure calculations could be overwhelmingly prohibitive for large systems. Assuming the locality principle, a convolutional neural network (CNN) model is developed to directly predict the effective local field and the corresponding energy change associated with a given spin update based on Ising configuration in a finite neighborhood. As the kernel size of the CNN is fixed at a constant, the model can be directly scalable to kMC simulations of large lattices. Our approach is reminiscent of the ML force field models widely used in first-principles molecular dynamics simulations. Applying our ML framework to a square-lattice double-exchange Ising model, we uncover unusual coarsening of ferromagnetic domains at low temperatures. Our work highlights the potential of ML methods for large-scale modeling of similar itinerant systems with discrete dynamical variables.

machine learning↗