Deep learning based surrogate models for first-principles global simulations of fusion plasmas
Explore the source record for details and available documents.
Dong, G. (ORCID:0000000277343736)↗
SEARCH · Engineering Papers
Results for “supervised learning”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Self-supervised physics-informed generative networks for phase retrieval from a single X-ray hologram
X-ray phase contrast imaging significantly improves the visualization of structures with weak or uniform absorption, broadening its applications across a wide range of scientific disciplines. Propagation-based phase contrast is particularly suitable for time- or dose-critical in vivo/in situ/operando (tomography) experiments because it requires only a single intensity measurement. However, the phase information of the wave field is lost during the measurement and must be recovered. Conventional algebraic and iterative methods often rely on specific approximations or boundary conditions that may not be met by many samples or experimental setups. In addition, they require manual tuning of reconstruction parameters by experts, making them less adaptable for complex or variable conditions. Here we present a self-learning approach for solving the inverse problem of phase retrieval in the near-field regime of Fresnel theory using a single intensity measurement (hologram). A physics-informed generative adversarial network is employed to reconstruct both the phase and absorbance of the unpropagated wave field in the sample plane from a single hologram. Unlike most state-of-the-art deep learning approaches for phase retrieval, our approach does not require paired, unpaired, or simulated training data. This significantly broadens the applicability of our approach, as acquiring or generating suitable training data remains a major challenge due to the wide variability in sample types and experimental configurations. The algorithm demonstrates robust and consistent performance across diverse imaging conditions and sample types, delivering quantitative, high-quality reconstructions for both simulated data and experimental datasets acquired at beamline P05 at PETRA III (DESY, Hamburg), operated by Helmholtz-Zentrum Hereon. Furthermore, it enables the simultaneous retrieval of both phase and absorption information.
36 MATERIALS SCIENCE↗
Physics-driven learning of Wasserstein GAN for density reconstruction in dynamic tomography
Object density reconstruction from projections containing scattered radiation and noise is of critical importance in many applications. Existing scatter correction and density reconstruction methods may not provide the high accuracy needed in many applications and can break down in the presence of unmodeled or anomalous scatter and other experimental artifacts. Incorporating machine-learning models could prove beneficial for accurate density reconstruction, particularly in dynamic imaging, where the time evolution of the density fields could be captured by partial differential equations or by learning from hydrodynamics simulations. In this work, we demonstrate the ability of learned deep neural networks to perform artifact removal in noisy density reconstructions, where the noise is imperfectly characterized. Here, we use a Wasserstein generative adversarial network (WGAN), where the generator serves as a denoiser that removes artifacts in densities obtained from traditional reconstruction algorithms. We train the networks from large density time-series datasets, with noise simulated according to parametric random distributions that may mimic noise in experiments. The WGAN is trained with noisy density frames as generator inputs, to match the generator outputs to the distribution of clean densities (time series) from simulations. A supervised loss is also included in the training, which leads to an improved density restoration performance. In addition, we employ physics-based constraints such as mass conservation during the network training and application to further enable highly accurate density reconstructions. Our preliminary numerical results show that the models trained in our frameworks can remove significant portions of unknown noise in density time-series data.
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
Anomaly detection with flow-based fast calorimeter simulators
Recently, several normalizing flow-based deep generative models have been proposed to accelerate the simulation of calorimeter showers. Using caloflow as an example, we show that these models can simultaneously perform unsupervised anomaly detection with no additional training cost. As a demonstration, we consider electromagnetic showers initiated by one (background) or multiple (signal) photons. The caloflow model is designed to generate single-photon showers, but it also provides access to the shower likelihood. We use this likelihood as an anomaly score and study the showers tagged as being unlikely. As expected, the tagger struggles when the signal photons are nearly collinear but is otherwise effective. This approach is complementary to a supervised classifier trained on only specific signal models using the same low-level calorimeter inputs. While the supervised classifier is also highly effective at unseen signal models, the unsupervised method is more sensitive in certain regions, and, thus, we expect that the ultimate performance will require a combination of these approaches.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
femto-PIXAR: a self-supervised neural network method for reconstructing femtosecond X-ray free electron laser pulses
X-ray Free Electron Lasers (X-FELs) operate in a wide range of lasing configurations for a broad variety of scientific applications at ultrafast time-scales such as structural biology, materials science, and atomic and molecular physics. Shot-by-shot characterization of the X-FEL pulses is crucial for analysis of many experiments as well as tuning the X-FEL performance. However, for the weak pulses found in advanced configurations, e.g. those needed for coherent, two-pulse studies of quantum materials, there is no current method for reliably resolving pulse profiles. Here we show that a physics-based U-net model can reconstruct the individual pulse power profiles for sub-picosecond pulse separation without the need for simulations. Using experimental data from weak X-FEL pulse pairs, we demonstrate we can learn the pulse characteristics on a shot-by-shot basis when conventional methods fail.
43 PARTICLE ACCELERATORS↗
Avian Activity Classification Using Recurrent Networks to Fuse Videos with Metadata on Imbalanced Datasets
Activity classification plays a crucial role in various real-life scenarios involving both humans and animals. There is an increasing need for precise activity classification focused on avian-solar interactions, as the usage of solar energy facilities, such as photovoltaic array power stations, has been observed to impact bird species richness, behavior, and activity. However, there has been no work to develop an automated system to monitor and classify these avian-solar interactions. All current methods rely on human observers, which is time and human resources costly and subject to errors related to searcher efficiency. With the recent success of Deep Learning models in activity classification problems, this paper develops a recurrent neural network-based model to automatically classify six avian activities around solar energy facilities. Our proposed model integrates critical feature engineering metadata with video frame data, enabling improved learning and more accurate activity classification. Furthermore, we address the challenge of data imbalance during training and demonstrate the efficacy of our model in detecting and classifying different activities within video tracks. Additionally, we analyze the saliency/backpropagation map of the trained proposed model and validate its decision-making rationale.
Avian activity classification; bidirectional LSTM;↗
Monitoring leaf phenology in moist tropical forests by applying a superpixel-based deep learning method to time-series images of tree canopies
Tropical leaf phenology-particularly its variability at the tree-crown scale-dominates the seasonality of carbon and water fluxes. However, given enormous species diversity, accurate means of monitoring leaf phenology in tropical forests is still lacking. Time series of the Green Chromatic Coordinate (GCC) metric derived from tower-based red-green-blue (RGB) phenocams have been widely used to monitor leaf phenology in temperate forests, but its application in the tropics remains problematic. To improve monitoring of tropical phenology, we explored the use of a deep learning model (i.e. superpixel-based Residual Networks 50, SP-ResNet50) to automatically differentiate leaves from non-leaves in phenocam images and to derive leaf fraction at the tree-crown scale. To evaluate our model, we used a year of data from six phenocams in two contrasting forests in Panama. Here, we first built a comprehensive library of leaf and non-leaf pixels across various acquisition times, exposure conditions and specific phenocams. We then divided this library into training and testing components. We evaluated the model at three levels: 1) superpixel level with a testing set, 2) crown level by comparing the model-derived leaf fractions with those derived using image-specific supervised classification, and 3) temporally using all daily images to assess the diurnal stability of the model-derived leaf fraction. Finally, we compared the model-derived leaf fraction phenology with leaf phenology derived from GCC. Our results show that: 1) the SP-ResNet50 model accurately differentiates leaves from non-leaves (overall accuracy of 93%) and is robust across all three levels of evaluations; 2) the model accurately quantifies leaf fraction phenology across tree-crowns and forest ecosystems; and 3) the combined use of leaf fraction and GCC helps infer the timing of leaf emergence, maturation and senescence, critical information for modeling photosynthetic seasonality of tropical forests. Collectively, this study offers an improved means for automated tropical phenology monitoring using phenocams.
54 ENVIRONMENTAL SCIENCES↗
Physics and chemistry from parsimonious representations: image analysis via invariant variational autoencoders
Electron, optical, and scanning probe microscopy methods are generating ever increasing volume of image data containing information on atomic and mesoscale structures and functionalities. This necessitates the development of the machine learning methods for discovery of physical and chemical phenomena from the data, such as manifestations of symmetry breaking phenomena in electron and scanning tunneling microscopy images, or variability of the nanoparticles. Variational autoencoders (VAEs) are emerging as a powerful paradigm for the unsupervised data analysis, allowing to disentangle the factors of variability and discover optimal parsimonious representation. Here, we summarize recent developments in VAEs, covering the basic principles and intuition behind the VAEs. The invariant VAEs are introduced as an approach to accommodate scale and translation invariances present in imaging data and separate known factors of variations from the ones to be discovered. We further describe the opportunities enabled by the control over VAE architecture, including conditional, semi-supervised, and joint VAEs. Several case studies of VAE applications for toy models and experimental datasets in Scanning Transmission Electron Microscopy are discussed, emphasizing the deep connection between VAE and basic physical principles. Python codes and datasets discussed in this article are available at https://github.com/saimani5/VAE-tutorials and can be used by researchers as an application guide when applying these to their own datasets.
36 MATERIALS SCIENCE↗
Synopsis of NREL's Automated Mobility District (AMD) Research Program and Associated Publications
An automated mobility district (AMD) envisions a system of integrated mobility options that serves major activity centers such as campuses, central business districts, and large medical facilities. The National Renewable Energy Laboratory (NREL) has been investigating the implementation prospects for fully automated passenger transport systems that are deployed to operate within dense urban settings. This document provides a synopsis of findings revealed over the last three phases of work, which have yielded insights into the creation and management of AMDs anticipated to use automated vehicle (AV) technology over the next decade. Phase I and Phase II tracked the deployment and lessons learned from 10 early-stage demonstrations of automated shuttle deployments, and their associated insights into the challenges for automated driving systems to achieve safe operations within district-scale deployments. Phase III began in-depth investigations of critical subsystem components, as automation, electrification, and on-demand service continue to converge within initial AMD operations. These directed studies focus on elements of electrification, curbfront/station management, the role of infrastructure sensing, and overall integration of AMD safety management in central, simultaneous coordination of multiple AMD fleets. Future research in AMDs includes systems engineering methodology (more frequently referred to as "digital twins") for planning, design, testing, and ongoing operation of AMDs; location (or co-location) of management functions; and human supervision and passenger communications for safety and security in unattended vehicles. The synopsis references the foundational research products (papers and presentations) that have been published through conference proceedings, journal articles, and NREL reports.
33 ADVANCED PROPULSION SYSTEMS↗
Optimizing training trajectories in variational autoencoders via latent Bayesian optimization approach *
Unsupervised and semi-supervised ML methods such as variational autoencoders (VAE) have become widely adopted across multiple areas of physics, chemistry, and materials sciences due to their capability in disentangling representations and ability to find latent manifolds for classification and/or regression of complex experimental data. Like other ML problems, VAEs require hyperparameter tuning, e.g. balancing the Kullback–Leibler and reconstruction terms. However, the training process and resulting manifold topology and connectivity depend not only on hyperparameters, but also their evolution during training. Because of the inefficiency of exhaustive search in a high-dimensional hyperparameter space for the expensive-to-train models, here we have explored a latent Bayesian optimization (zBO) approach for the hyperparameter trajectory optimization for the unsupervised and semi-supervised ML and demonstrated for joint-VAE with rotational invariances. We have demonstrated an application of this method for finding joint discrete and continuous rotationally invariant representations for modified national institute of standards and technology database (MNIST) and experimental data of a plasmonic nanoparticles material system. The performance of the proposed approach has been discussed extensively, where it allows for any high dimensional hyperparameter trajectory optimization of other ML models.
42 ENGINEERING↗
Towards developing multiscale-multiphysics models and their surrogates for digital twins of metal additive manufacturing
Artificial intelligence (AI) embedded within digital models of manufacturing processes can be used to improve process productivity and product quality significantly. The application of such advanced capabilities particularly to highly digitalized processes such as metal additive manufacturing (AM) is likely to make those processes commercially more attractive. AI capabilities will reside within Digital Twins (DTs) which are living virtual replicas of the physical processes. DTs will be empowered to operate autonomously in a diagnostic control capacity to supervise processes and can be interrogated by the practitioner to inform the optimal processing route for any given product. The utility of the information gained from the DTs would depend on the quality of the digital models and, more importantly, their faster-solving surrogates which dwell within DTs for consultation during rapid decision-making. In this article, we point out the exceptional value of DTs in AM and focus on the need to create high-fidelity multiscale-multiphysics models for AM processes to feed the AI capabilities. We identify technical hurdles for their development, including those arising from the multiscale and multiphysics characteristics of the models, the difficulties in linking models of the subprocesses across scales and physics, and the scarcity of experimental data. We discuss the need for creating surrogate models using machine learning approaches for real-time problem-solving. We further identify non-technical barriers, such as the need for standardization and difficulties in collaborating across different types of institutions. We offer potential solutions for all these challenges, after reflecting on and researching discussions held at an international symposium on the subject in 2019. Here, we argue that a collaborative approach can not only help accelerate their development compared with disparate efforts, but also enhance the quality of the models by allowing modular development and linkages that account for interactions between the various sub-processes in AM. A high-level roadmap is suggested for starting such a collaboration.
36 MATERIALS SCIENCE↗
Forming a database to study reversed magnetic shear from the National Spherical Torus eXperiment using machine learning
Achieving a long-lived reversed magnetic shear (RMS) target plasma in the National Spherical Torus eXperiment Upgrade will require developing various sustainment scenarios. To help with the ongoing plasma control efforts, the development of a new analysis for the motional Stark effect (MSE) diagnostic using a machine learning algorithm, namely, MSE-ML, is described. MSE-ML will be used to identify patterns during RMS discharges, some of which suffer magnetohydrodynamic (MHD) events resulting in current redistribution and monotonic q-profiles. A database consisting of q and magnetic shear profiles is being constructed primarily based on the existing National Spherical Torus eXperiment data with equilibrium reconstructions constrained by the magnetic field pitch angle profile measured using the multi-channel MSE diagnostic. An unsupervised k-means clustering of the data is developed to study the RMS formation as a function of time. The initial clustering from the q-profiles shows significant differences in both amplitude and the duration of the RMS period. As a goal, the clustering results that detect and distinguish shots with substantial and sustained RMS are to be used as a preprocessing step in a supervised algorithm to identify the underlying conditions that lead to long-lasting improved confinement with RMS. Another aim of the MSE-ML study is to identify precursors of RMS-destroying MHD events in either derived data such as the q-profile or directly measured data such as the magnetic field pitch angle profile.
Uzun-Kaymak, I. U. (ORCID:0000000276251493)↗
Source-agnostic gravitational-wave detection with recurrent autoencoders
Abstract We present an application of anomaly detection techniques based on deep recurrent autoencoders (AEs) to the problem of detecting gravitational wave (GW) signals in laser interferometers. Trained on noise data, this class of algorithms could detect signals using an unsupervised strategy, i.e. without targeting a specific kind of source. We develop a custom architecture to analyze the data from two interferometers. We compare the obtained performance to that obtained with other AE architectures and with a convolutional classifier. The unsupervised nature of the proposed strategy comes with a cost in terms of accuracy, when compared to more traditional supervised techniques. On the other hand, there is a qualitative gain in generalizing the experimental sensitivity beyond the ensemble of pre-computed signal templates. The recurrent AE outperforms other AEs based on different architectures. The class of recurrent AEs presented in this paper could complement the search strategy employed for GW detection and extend the discovery reach of the ongoing detection campaigns.
47 OTHER INSTRUMENTATION↗
Neural network for 3D inertial confinement fusion shell reconstruction from single radiographs
In inertial confinement fusion (ICF), x-ray radiography is a critical diagnostic for measuring implosion dynamics, which contain rich three-dimensional (3D) information. Traditional methods for reconstructing 3D volumes from 2D radiographs, such as filtered backprojection, require radiographs from at least two different angles or lines of sight (LOS). In ICF experiments, the space for diagnostics is limited, and cameras that can operate on fast timescales are expensive to implement, limiting the number of projections that can be acquired. To improve the imaging quality as a result of this limitation, convolutional neural networks (CNNs) have recently been shown to be capable of producing 3D models from visible light images or medical x-ray images rendered by volumetric computed tomography. Here, we propose a CNN to reconstruct 3D ICF spherical shells from single radiographs. We also examine the sensitivity of the 3D reconstruction to different illumination models using preprocessing techniques such as pseudo-flatfielding. To resolve the issue of the lack of 3D supervision, we show that training the CNN utilizing synthetic radiographs produced by known simulation methods allows for reconstruction of experimental data as long as the experimental data are similar to the synthetic data. We also show that the CNN allows for 3D reconstruction of shells that possess low mode asymmetries. Further comparisons of the 3D reconstructions with direct multiple LOS measurements are justified.
3D Reconstruction↗
Multilevel Cybersecurity for Photovoltaic Systems
The motivation behind this project is to protect critical infrastructure in electric power generation pertaining to solar photovoltaic (PV) systems. This growing renewable energy resource is becoming a more vital part of the nation’s energy portfolio, particularly since it has achieved grid-parity to existing generation methods in terms of cost. It is thus vital that steps be taken to ensure the cybersecurity of these assets. The project goal was to devise a multilevel cybersecurity solution to address PV security gaps at the inverter and system levels, and field test the solution under the supervision and review of a US-based solar inverter manufacturer and PV installer/operator. A two-level cyberattack defense approach was formulated whereby the first level, the solar inverter level, hardens individual devices and achieves a deeply cyber-secure inverter. The inverter level security involves a multi-layer defense-in-depth approach for securing the inverter while also providing data for the system level algorithms. The second level, the system level, addresses intrusion detection and restoration involving an ensemble of inverters and relevant systems.
14 SOLAR ENERGY↗
A unified large language model–based framework for heterogeneous PV image diagnosis
With advances in imaging technologies, modern photovoltaic (PV) systems generate large volumes of heterogeneous image data, including visible, electroluminescence (EL), and infrared (IR) images. Existing PV image analysis models, particularly deep learning approaches, are typically task-specific and lack cross-modality generalization. To address this limitation, this paper proposes an open-source large language model (LLM)–based unified framework for heterogeneous PV image diagnostics. Through task-aware diagnostic prompting, the framework enables analysis of visible, EL, and IR images within a single pipeline, supporting both zero-shot and few-shot inference and binary and multiclass classification. It is compatible with state-of-the-art multimodal LLMs, including ChatGPT, Gemini, Claude, Qwen, and CLIP. The framework is evaluated on PV module condition classification (clean, soiling, snow, hail, and bird droppings) using visible images, cell crack detection using EL images, and hotspot detection using IR images. GPT-5.1 in few-shot mode achieves the best performance, with classification accuracy exceeding 97.3%. Open-source models such as Qwen and CLIP also deliver competitive results on visible images (around 90% accuracy), though their performance is more limited on EL and IR modalities. On the full ELPV dataset, the framework achieves 83.5% zero-shot accuracy, within 2.8% of the supervised CNN baseline, confirming scalability to larger benchmarks. Practical aspects such as reproducibility, response latency, and confidence estimation are systematically analyzed. The framework operates across PV image modalities without modality- or task-specific training, making it well suited as a rapid pre-screening tool to support downstream detailed diagnostics. A benchmark dataset of diverse labeled PV images is also released.
Li, Baojie↗
FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics
Large language models have revolutionized artificial intelligence by enabling large, generalizable models trained through self-supervision. This paradigm has inspired the development of scientific foundation models (FMs). However, applying this capability to experimental particle physics is challenging due to the sparse, spatially distributed nature of detector data, which differs dramatically from natural language. This work addresses if an FM for particle physics can scale and generalize across diverse tasks. We introduce a new dataset with more than 11 million particle collision events and a suite of downstream tasks and labeled data for evaluation. We propose a novel self-supervised training method for detector data and demonstrate its neural scalability with models that feature up to 188 million parameters. With frozen weights and task-specific adapters, this FM consistently outperforms baseline models across all downstream tasks. The performance also exhibits robust data-efficient adaptation. Further analysis reveals that the representations extracted by the FM are task-agnostic but can be specialized via a single linear mapping for different downstream tasks.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Snekmer: a scalable pipeline for protein sequence fingerprinting based on amino acid recoding
Abstract Motivation The vast expansion of sequence data generated from single organisms and microbiomes has precipitated the need for faster and more sensitive methods to assess evolutionary and functional relationships between proteins. Representing proteins as sets of short peptide sequences (kmers) has been used for rapid, accurate classification of proteins into functional categories; however, this approach employs an exact-match methodology and thus may be limited in terms of sensitivity and coverage. We have previously used similarity groupings, based on the chemical properties of amino acids, to form reduced character sets and recode proteins. This amino acid recoding (AAR) approach simplifies the construction of protein representations in the form of kmer vectors, which can link sequences with distant sequence similarity and provide accurate classification of problematic protein families. Results Here, we describe Snekmer, a software tool for recoding proteins into AAR kmer vectors and performing either (i) construction of supervised classification models trained on input protein families or (ii) clustering for de novo determination of protein families. We provide examples of the operation of the tool against a set of nitrogen cycling families originally collected using both standard hidden Markov models and a larger set of proteins from Uniprot and demonstrate that our method accurately differentiates these sequences in both operation modes. Availability and implementation Snekmer is written in Python using Snakemake. Code and data used in this article, along with tutorial notebooks, are available at http://github.com/PNNL-CompBio/Snekmer under an open-source BSD-3 license. Supplementary information Supplementary data are available at Bioinformatics Advances online.
59 BASIC BIOLOGICAL SCIENCES↗