Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep learning method”

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 685 records · Page 38

Fully Convolutional Spatio-Temporal Models for Representation Learning in Plasma Science

We have trained a fully convolutional spatio-temporal model for fast and accurate representation learning in the challenging exemplar application area of fusion energy plasma science. The onset of major disruptions is a critically important fusion energy science issue that must be resolved for advanced tokamak plasmas such as the $25B burning plasma international thermonuclear experimental reactor (ITER) experiment. While a variety of statistical methods have been used to address the problem of tokamak disruption prediction and control, recent approaches based on deep learning have proven particularly compelling. In the present paper, we introduce further improvements to the fusion recurrent neural network (FRNN) software suite, which delivered cross-machine disruption predictions with unprecedented accuracy using a large database of experimental signals from two major tokamaks. Up to now, FRNN was based on the long short-term memory (LSTM) variant of recurrent neural networks to leverage the temporal information in the data. Here, we implement and apply the "temporal convolutional neural network (TCN)" architecture to the time-dependent input signals. Furthermore, this allows highly optimized convolution operations to carry the majority of the computational load of training, thus enabling a reduction in training time, and the effective use of high-performance computing resources for hyperparameter tuning. At the same time, the TCN-based architecture achieves better predictive performance when compared with the LSTM architecture for various tasks for a representative fusion database.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

UIR-Net: Object Detection in Infrared Imaging of Thermomechanical Processes in Automotive Manufacturing

Thermomechanical processes (TMPs) such as resistance spot welding (RSW) and hot stamping are widely used in automotive manufacturing. Recent advancement in sensing technology has led to an increasing adoption of thermographic cameras to capture the infrared (IR) radiation of a metal part (or component of a part) during its thermomechanical processing or immediately after the process when the part is still hot. Detecting the object(s) of interest from raw IR images is an essential step in analyzing these data. Deep learning (DL) has been a recent success for object detection (OD), but the application of DL-based OD for industrial IR images in manufacturing is largely lagging behind. The major contribution of this work, which is also the distinction from previous OD studies, is the capability of building the OD model with unlabeled IR images, i.e., imaging data without accurate information indicating the object position. Here, the architecture of Unsupervised IR Image Net (UIR-Net) is designed to accommodate the unique characteristics of IR images from TMPs in manufacturing. This study presents a novel method for OD in unlabeled IR images from TMPs. The proposed method, called UIR-Net, consists of two components: label generation and DL model construction. Two case studies from automotive manufacturing, RSW and hot stamping, are reported to demonstrate the feasibility and effectiveness of the proposed method.

42 ENGINEERING↗

Sharpness-Aware Low-Dose CT Denoising Using Conditional Generative Adversarial Network

Low-dose computed tomography (LDCT) has offered tremendous benefits in radiation-restricted applications, but the quantum noise as resulted by the insufficient number of photons could potentially harm the diagnostic performance. Current image-based denoising methods tend to produce a blur effect on the final reconstructed results especially in high noise levels. In this paper, a deep learning-based approach was proposed to mitigate this problem. An adversarially trained network and a sharpness detection network were trained to guide the training process. Experiments on both simulated and real dataset show that the results of the proposed method have very small resolution loss and achieves better performance relative to state-of-the-art methods both quantitatively and visually.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Five-second STEM dislocation tomography for 300 nm thick specimen assisted by deep-learning-based noise filtering

Scanning transmission electron microscopy (STEM) is suitable for visualizing the inside of a relatively thick specimen than the conventional transmission electron microscopy, whose resolution is limited by the chromatic aberration of image forming lenses, and thus, the STEM mode has been employed frequently for computed electron tomography based three-dimensional (3D) structural characterization and combined with analytical methods such as annular dark field imaging or spectroscopies. However, the image quality of STEM is severely suffered by noise or artifacts especially when rapid imaging, in the order of millisecond per frame or faster, is pursued. Here we demonstrate a deep-learning-assisted rapid STEM tomography, which visualizes 3D dislocation arrangement only within five-second acquisition of all the tilt-series images even in a 300 nm thick steel specimen. The developed method offers a new platform for various in situ or operando 3D microanalyses in which dealing with relatively thick specimens or covering media like liquid cells are required.

47 OTHER INSTRUMENTATION↗

Boosting Noise2Inverse via enhanced model selection for denoising computed tomography data

Synchrotron-based x-ray tomographic imaging enables the examination of the internal structure of materials at high spatial and temporal resolution. Experimental constraints can impose dose and time limits on the measurements, introducing a higher level of noise and artifacts in the reconstructed images. Deep learning has emerged as a powerful tool to remove noise from reconstructed images. Recently, the Noise2Inverse method was designed specifically for denoising reconstructed images without requiring paired noisy and clean images. This method creates multiple statistically independent reconstructions used to pair the data in which training involves transforming one reconstruction into the other, and vice versa. Originally designed to be used after a fixed number of epochs, we see in practice that this approach may not produce the optimal model and may unnecessarily waste computational resources. Therefore, we propose an alternative method of identifying the best model during training that aligns with the Noise2Inverse method. During validation, we compare the model output of the multiple reconstructions among each other. We hypothesize that the best model is the one that produces images with the highest similarity, implying a convergence in the predicted material properties and absorption values. To compare model outputs, we consider the absolute error, square error, structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), and cosine similarity. We evaluate our method on two simulated tomography datasets and two, real-world, low-contrast, high-energy x-ray tomography datasets. We show our approach is more effective at determining the best model, up to an increase of 12.50% and 12.53% in SSIM and PSNR, respectively, while only requiring a fifth of the training time compared to the original approach.

CT↗

Bond order predictions using deep neural networks

Machine learning is an extremely powerful tool for the modern theoretical chemist since it provides a method for bypassing costly algorithms for solving the Schrödinger equation. Already, it has proven able to infer molecular and atomic properties such as charges, enthalpies, dipoles, excited state energies, and others. Most of these machine learning algorithms proceed by inferring properties of individual atoms, even breaking down total molecular energy into individual atomic contributions. In this paper, we introduce a modified version of the Hierarchically Interacting Particle Neural Network (HIP-NN) capable of making predictions on the bonds between atoms rather than on the atoms themselves. We train the modified HIP-NN to infer bond orders for a large number of small organic molecules as computed via the Natural Bond Orbital package. We demonstrate that the trained model is extensible to molecules much larger than those in the training set by studying its performance on the COMP6 dataset. This method has applications in cheminformatics and force field parameterization and opens a promising future for machine learning models to predict other quantities that are defined between atoms such as density matrix elements, Hamiltonian parameters, and molecular reactivities.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Deep Learning Based Superconducting Radio-Frequency Cavity Fault Classification at Jefferson Laboratory

This work investigates the efficacy of deep learning (DL) for classifying C100 superconducting radio-frequency (SRF) cavity faults in the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. CEBAF is a large, high-power continuous wave recirculating linac that utilizes 418 SRF cavities to accelerate electrons up to 12 GeV. Recent upgrades to CEBAF include installation of 11 new cryomodules (88 cavities) equipped with a low-level RF system that records RF time-series data from each cavity at the onset of an RF failure. Typically, subject matter experts (SME) analyze this data to determine the fault type and identify the cavity of origin. This information is subsequently utilized to identify failure trends and to implement corrective measures on the offending cavity. Manual inspection of large-scale, time-series data, generated by frequent system failures is tedious and time consuming, and thereby motivates the use of machine learning (ML) to automate the task. This study extends work on a previously developed system based on traditional ML methods (Tennant and Carpenter and Powers and Shabalina Solopova and Vidyaratne and Iftekharuddin, Phys. Rev. Accel. Beams, 2020, 23, 114601), and investigates the effectiveness of deep learning approaches. The transition to a DL model is driven by the goal of developing a system with sufficiently fast inference that it could be used to predict a fault event and take actionable information before the onset (on the order of a few hundred milliseconds). Because features are learned, rather than explicitly computed, DL offers a potential advantage over traditional ML. Specifically, two seminal DL architecture types are explored: deep recurrent neural networks (RNN) and deep convolutional neural networks (CNN). We provide a detailed analysis on the performance of individual models using an RF waveform dataset built from past operational runs of CEBAF. In particular, the performance of RNN models incorporating long short-term memory (LSTM) are analyzed along with the CNN performance. Furthermore, comparing these DL models with a state-of-the-art fault ML model shows that DL architectures obtain similar performance for cavity identification, do not perform quite as well for fault classification, but provide an advantage in inference speed.

97 MATHEMATICS AND COMPUTING↗

Novel Fault Location Method for Power Systems Based on Attention Mechanism and Double Structure GRU Neural Network

Fault location is one of the most essential techniques to maintain the stable operation of power systems. A fast and accurate fault location allows operators to restore power grids faster and avoid economic losses. Conventional methods rely on expert knowledge to extract the necessary features (e.g. DWT, DFT). For large systems, more coupling effects of transmission lines require more complex feature engineering, and incomplete features can easily introduce large errors. To overcome this, a deep learning approach without manual feature extraction is introduced to the fault location model under big data application. Towards this end, in the proposed method, the attention mechanism, the Bi-GRU and a dual structure network are applied to analyze the current data from different perspectives. Complete information for the fault features is extracted for the fault location. Based on a time series model and benefit from the ability to internally acquire the information architecture of faulty line, the established model is adaptive to the power grids with very complex topologies. Simulation results indicate that the proposed double-structure model reduces the maximum error and is less affected by noise. In comparison with different structures and different models, the proposed method shows better performance in IEEE 39-bus system.

42 ENGINEERING↗

Physics-Guided Deep Learning for Time-Series State Estimation Against False Data Injection Attacks

The modern power grid is a cyber-physical system. While the grid is becoming more intelligent with emerging sensing and communication techniques, new vulnerabilities are introduced and cyber security becomes a major concern. One type of cyber-attacks - False Data Injection Attacks (FDIAs) - exploits the limitations in traditional power system state estimation, and modifies system states without being detected. In this paper, we propose a physics-guided deep learning (PGDL) approach to defend against FDIAs. The PGDL takes real-time measurements as inputs to neural networks, outputs the estimated states, and reconstructs measurements considering power system physics. A deep recurrent neural network - Long Short-Term Memory (LSTM) - is employed to learn the temporal correlations among states. This hybrid learning model leads to a time-series state estimation method to defend against FDIAs. The simulation results using IEEE 14-bus test system demonstrate the accuracy and robustness of the proposed time-series state estimation under FDIAs.

Cyber security, state estimation, physics-guided d↗

A Deep Learning-Based Workflow for Fast Prediction of 3D State Variables in Geological Carbon Storage: A Dimension Reduction Approach

In this study, we used deep learning techniques, which are a form of artificial intelligence, to create fast and effective models for predicting how fluids flow in underground geological formations. This is important for managing geological carbon storage, a method used to fight climate change by storing carbon dioxide underground. The challenge lies in the complex nature of these underground spaces and the large amount of data needed to accurately simulate them. To overcome these issues, we developed a new workflow that reduces the data’s complexity before training the deep learning model and then reconstructs the predicted results in their original form. We also proposed a unique approach to handle the specific complexities found in 3D saturation fields, a crucial aspect of fluid flow prediction. We tested our method using real-world data from the Gulf of Mexico. Our results show that our approach not only accurately predicts fluid behavior but also significantly reduces computation time. This will greatly improve real-time decision-making and risk assessment in large-scale geological carbon storage operations.

Wang, Hongsheng↗

Hydrodynamic and Radiographic Toolbox (HART)

With a multi-lab and university team, we propose to develop new methods for a Hydrodynamic and Radiographic Toolbox (HART) that will enable a fuller and more extensive use of experimental radiographic data towards better characterizing and reducing uncertainties in predictive modeling of weapons performance. We will achieve this by leveraging recent developments in areas of computational imaging, statistical and machine learning, and reduced order modeling of hydrodynamics. In terms of software practices, by partnering with XCP (RISTRA Project) we will conform to recent XCP standards that are inline with modern software practices and standards and ensure compatibility and ease of inter-operability with existing codes. Three new activities under this proposal include (a) the development and use of deep learning-based surrogates to accelerate reconstruction and variational inference of density fields from radiographs of hydrotests, (b) a model-data fusion strategy that couples deep learning-based density reconstructions with fast hydrodynamics simulators to better constrain the reconstruction, and (c) a method for treating asymmetries using techniques adopted from limited view tomography. All three new activities will be based on improved treatment of scatter, noise, beam spot movement, detector blur, and flat fielding in the forward model, and a use of sophisticated priors to aid in the re construction. The improvements to the forward model and improved algorithmic design of the reconstruction when complete will be contained in the iterative reconstruction code SHIVA—a code project that we have recently initiated. The many ways in which machine learning can be used in the reconstruction work will be contained in a code HERMES that has been initiated with DTRA support. For example, the significant levels of acceleration that will likely be achieved by the use of machine learning techniques will permit us (and are required) to quantify uncertainties in density retrievals. Next, the two-way coupling between density reconstruction and model-based simulation of the hydrodynamics will be contained in code EREBUS, and will permit a fuller realization of the potential of the data to constrain the hydrodynamic model and better address issues related to asymmetries in the problem. Finally, we anticipate that the better consistency with physics achieved in our reconstructions will allow them to be used by X-Division more so than today.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

DeepCare: Improving Patient Care using Deep Learning on Electronic Health Records

Coordinating patient care using electronic health records (EHR) data presents an exciting but formidable opportunity in data extraction, analysis and modeling. Traditional methods use a manual feature driven approach to model patients with age, family history and symptoms to predict disease outcomes. We propose a novel approach to model patients based on their streaming electronic health records data combined with information from medical knowledge bases, which has been gained over years of medical research. Using a combination of representation learning and long short term memory (LSTM) networks we plan to model patient evolution over time, leading to more accurate and individualized predictive models for patient’s diseases. Our approach will be transformative in providing critical decision support for patient care, enabling accurate understanding and evolution of diseases in patients.

60 APPLIED LIFE SCIENCES↗

Safe Reinforcement Learning for Emergency Load Shedding of Power Systems

The paradigm shift in the electric power grid necessitates a revisit of existing control methods to ensure the grid’s security and resilience. In particular, the increased uncertainties and rapidly changing operational conditions in power systems have revealed outstanding issues in terms of either speed, adaptiveness, or scalability of the existing control methods for power systems. On the other hand, the availability of massive real-time data can provide a clearer picture of what is happening in the grid. Recently, deep reinforcement learning (RL) has been regarded and adopted as a promising approach leveraging massive data for fast and adaptive grid control. However, like most existing machine learning (ML)- based control techniques, RL control usually cannot guarantee the safety of the power systems. In this paper, we introduce a novel method for safe RL-based load shedding of power systems that can enhance the safe voltage recovery of the electric power grid after experiencing faults. Numerical simulation on the IEEE 39-bus testcase is performed to demonstrate the effectiveness of the proposed safe RL emergency control, as well as its adaptive capability to faults not seen in the training.

reinforcement learning, emergency control, power g↗

Automatic detection of methane emissions in multispectral satellite imagery using a vision transformer

Curbing methane emissions is among the most effective actions that can be taken to slow down global warming. However, monitoring emissions remains challenging, as detection methods have a limited quantification completeness due to trade-offs that have to be made between coverage, resolution and detection accuracy. Here we show that deep learning can overcome the trade-off in terms of spectral resolution that comes with multi-spectral satellite data, resulting in a methane detection tool with global coverage and high temporal and spatial resolution. We compare our detections with airborne methane measurement campaigns, which suggests that our method can detect methane point sources in Sentinel-2 data down to plumes of 0.01 km 2 , corresponding to 200 to 300 kg CH 4 /h -1 sources. Our model shows an order of magnitude improvement over the state-of-the-art, providing a significant step towards the automated, high resolution detection of methane emissions at a global scale, every few days.

54 ENVIRONMENTAL SCIENCES↗

Assessing and benchmarking the fidelity of posterior inference methods for astrophysics data analysis

In this era of large and complex astronomical survey data, interpreting, validating, and comparing inference techniques becomes increasingly difficult. This is particularly critical for emerging inference methods like Simulation-Based Inference (SBI), which offer significant speedup potential and posterior modeling flexibility, especially when deep learning is incorporated. We present a study to assess and compare the performance and uncertainty prediction capability of Bayesian inference algorithms – from traditional MCMC sampling of analytic functions to deep learning-enabled SBI. We focus on testing the capacity of hierarchical inference modeling in those scenarios. Before we extend this study to cosmology, we first use astrophysical simulation data to ensure interpretability. We demonstrate a probabilistic programming implementation of hierarchical and non-hierarchical Bayesian inference using simulations derived from the DeepBench software library, a benchmarking tool developed by our group that generates simple and controllable astrophysical objects from first principles. This study will enable astronomers and physicists to harness the inference potential of these methods with confidence.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A Deep Learning Approach for TNC Trip Demand Prediction Considering Spatial-Temporal Features: Preprint

Ride-hailing or transportation network companies (TNCs), such as Uber, Lyft, DiDi Chuxing, or RideAustin, are emerging as a new and disruptive on-demand mobility service in recent years. However, the methods for developing predictive analytics to explore the nature and dynamics of TNCs across cities in the United States are still nascent due to the lack of publicly available data. Recently available public datasets on TNCs by RideAustin offer a unique opportunity to examine spatial, temporal, environmental, and special event factors associated with TNC trip demand. This study explores the use of a deep learning approach - Long Short-Term Memory (LSTM) - to predict TNC trip demand at the ZIP Code level using data from Austin, Texas. The analysis includes key predictive factors such as time of day, day of week, precipitation, and temperature, indicating their corresponding associations with TNC trip demand. Results from initial analysis show that LSTM is able to predict the TNC trip demand for the upcoming hour accurately. LSTM, when compared to other prediction methods, such as historical average and instantaneous trip demand, reduces the mean absolute error (MAE) of the model predictions by 37% and 24%, respectively. This novel method offers significant potential for scaling up and/or replicability across cities where data are available for understanding TNC trip demand to inform emerging mobility system operators. Predicting trip demand can help TNC drivers make informed decisions on how to be more efficient by maximizing passenger pickups and minimizing wait times and deadheading.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Robust, automated sleep scoring by a compact neural network with distributional shift correction

Studying the biology of sleep requires the accurate assessment of the state of experimental subjects, and manual analysis of relevant data is a major bottleneck. Recently, deep learning applied to electroencephalogram and electromyogram data has shown great promise as a sleep scoring method, approaching the limits of inter-rater reliability. As with any machine learning algorithm, the inputs to a sleep scoring classifier are typically standardized in order to remove distributional shift caused by variability in the signal collection process. However, in scientific data, experimental manipulations introduce variability that should not be removed. For example, in sleep scoring, the fraction of time spent in each arousal state can vary between control and experimental subjects. We introduce a standardization method, mixture z-scoring, that preserves this crucial form of distributional shift. Using both a simulated experiment and mouse in vivo data, we demonstrate that a common standardization method used by state-of-the-art sleep scoring algorithms introduces systematic bias, but that mixture z-scoring does not. We present a free, open-source user interface that uses a compact neural network and mixture z-scoring to allow for rapid sleep scoring with accuracy that compares well to contemporary methods. This work provides a set of computational tools for the robust automation of sleep scoring.

60 APPLIED LIFE SCIENCES↗

Unifying Quantum Materials Modeling and Experiments: The Role of Machine Learning Interatomic Potentials

Computational experiments have emerged as a powerful complement to traditional experiments in the design of new materials. The development of machine learning (ML) and deep learning techniques, combined with database construction and data mining, has significantly enhanced traditional quantum mechanical methods. This synergy enables the rapid development of structure-property relationships. In this talk, I will discuss our recent efforts in applying Machine Learning Interatomic Potentials (MLIAPs) to accelerate materials modeling across various material classes and challenging applications where traditional methods fall short. First, I will highlight the success of MLIAPs in accurately modeling the melting behavior of complex materials. Our results demonstrate high fidelity with experimental observations and also with calculated reference melting temperatures. In the second application, I will discuss how MLIAPs are trained and applied to elucidate the interplay between segregation tendencies and surface reconstructions in CuNi alloys under oxidizing conditions. A key factor in the success of these MLIAP applications is the design of minimalistic yet flexible datasets along with a computational framework for training MLIAPs.

Saidi, Wissam↗