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 1,495 records · Page 83

Performance of Pulsed Thermal Tomography Imaging with Machine Learning-Based Classification of Defects in Additively Manufactured Structures

Additive manufacturing (AM) is an emerging method for cost-efficient fabrication of complex topology nuclear reactor parts from high-strength corrosion resistance alloys, such as stainless steel and Inconel. AM of metallic structures for nuclear energy applications is currently based on laser powder bed fusion (LPBF) process. Some of the challenges with using LPBF method for nuclear manufacturing include the possibility of introducing pores into metallic structures. Integrity of AM structures needs to be evaluated nondestructively because material flaws could lead to premature failures in high temperature nuclear reactor environment. Currently, there exist limited capabilities to evaluate actual AM structures non-destructively. Pulsed Thermal Tomography Imaging (PTT) provides a capability for non-destructive evaluation (NDE) of subsurface defects in arbitrary size structures. The PTT method is based on recording material surface temperature transients with infrared (IR) camera following thermal pulse delivered on material surface with flash light. The PTT method has advantages for NDE of actual AM structures because the method involves one-sided non-contact measurements and fast processing of large sample areas captured in one image. Following initial qualification of an AM component for deployment in a nuclear reactor, a PTT system can also be used for in-service nondestructive evaluation (NDE) applications. In this report, we describe recent progress in enhancing PTT capabilities in detecting and visualizing microscopic defects in metallic specimens. The thermal tomography (TT) algorithm obtains depth reconstructions of spatial effusivity from the data cube of sequentially recorded surface temperatures. However, interpretation of TT images is non-trivial because of blurring of images with increasing depth. To address this challenge, we have developed a deep learning convolutional neural network (CNN) to classify size and orientation subsurface defects in simulated TT images. CNN is trained on a database of TT images created for a set of simulated metallic structures with elliptical subsurface voids. Test of CNN performance demonstrate the ability to classify radii and angular orientation of subsurface defects in TT images. In addition, we have shown that CNN trained on elliptical defects is capable of classifying irregular-shaped defects obtained from scanning electron microscopy (SEM) of stainless steel sections printed with LPBF.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

An interactive machine learning platform for analyzing multi-particle coincidence data from cold target recoil ion momentum spectroscopy

We present SCULPT (Supervised Clustering and Uncovering Latent Patterns with Training), a comprehensive software platform for analyzing tabulated high-dimensional multi-particle coincidence data from Cold Target Recoil Ion Momentum Spectroscopy (COLTRIMS) experiments. The software addresses critical challenges in modern momentum spectroscopy by integrating advanced machine learning techniques with physics-informed analysis in an interactive web-based environment. SCULPT implements uniform manifold approximation and projection for non-linear dimensionality reduction to reveal correlations in high-dimensional data. We also discuss potential extensions to deep autoencoders for feature learning and genetic programming for automated discovery of physically meaningful observables. A novel adaptive confidence scoring system provides quantitative reliability assessments by evaluating user-selected clustering quality metrics with predefined weights that reflect each metric’s robustness. The platform features configurable molecular profiles for different experimental systems, interactive visualization with selection tools, and comprehensive data filtering capabilities. Utilizing a subset of SCULPT’s capabilities, we analyze photo-double-ionization data measured using the COLTRIMS method for three-body dissociation of the D 2 O molecule, revealing distinct fragmentation channels and their correlations with physics parameters. The software’s modular architecture and web-based implementation make it accessible to the broader atomic and molecular physics community, significantly reducing the time required for complex multi-dimensional analyses. This opens the door to finding and isolating rare events exhibiting non-linear correlations on the fly during experimental measurements, which can help steer exploration and improve the efficiency of experiments.

Artificial neural networks↗

A Reinforcement Learning Approach to Augment Conventional PID Control in Nuclear Power Plant Transient Operation

The ability of nuclear reactors to operate their power conversion cycles more flexibly will enhance their value to energy grids with variable pricing. Current nuclear control systems are typically classical controllers that are often based on proportional-integral-derivative (PID) control. This paper presents a method of augmenting the existing PID control for difficult transient operations in nuclear power plants using a reinforcement learning–derived feedforward signal applied in real time. The agents, which are trained on a test thermal load-following problem, are designed to improve steam generator outlet temperature control for a range of fast load-following scenarios covering ramp rates from 9%/min to 15%/min. Several reinforcement learning algorithms were initially investigated for the training of the feedforward agents with deep Q-learning (DQN) and proximal policy optimization (PPO) networks, which were found to be the most promising. The DQN controllers utilize discrete actions, giving them a better disturbance rejection at steady state but inconsistent response to initial temperature deviations. In contrast, PPO-trained agents, which take continuous actions except for a dead zone around zero, were shown to have the best combination of high disturbance rejection at steady state and good tracking of the desired temperature value. The ability of the PPO agent was also examined, with the average time of decision making found to be on the order of 1 ms. The fault properties of the controller under the loss of the reinforcement learning agent feedforward signal were also examined. The controller showed strong performance in situations of “no-signal” faults. but was less good at handling “stuck-at” faults, where the feedforward signal remains at a set value. In both cases, however, the PID was able to successfully maintain stability, eventually returning the system to a steady state. It is hoped that this work will allow for the proposed control architecture to be examined for more difficult control problems such that it may eventually be used to adapt existing nuclear plants for more aggressive load-following on grids of the future.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

LSBGnet: an improved detection model for low-surface brightness galaxies

ABSTRACT The Chinese Space Station Telescope (CSST) is scheduled to launch soon, which is expected to provide a vast amount of image potentially containing low-surface brightness galaxies (LSBGs). However, detecting and characterizing LSBGs is known to be challenging due to their faint surface brightness, posing a significant hurdle for traditional detection methods. In this paper, we propose LSBGnet, a deep neural network specifically designed for automatic detection of LSBGs. We established LSBGnet-SDSS model using data set from the Sloan Digital Sky Survey (SDSS). The results demonstrate a significant improvement compared to our previous work, achieving a recall of 97.22 per cent and a precision of 97.27 per cent on the SDSS test set. Furthermore, we use the LSBGnet-SDSS model as a pre-training model, employing transfer learning to retrain the model with LSBGs from Dark Energy Survey (DES), and establish the LSBGnet-DES model. Remarkably, after retraining the model on a small DES sample, it achieves over 90 per cent precision and recall. To validate the model’s capabilities, we utilize the trained LSBGnet-DES model to detect LSBG candidates within a selected 5 sq. deg area in the DES footprint. Our analysis reveals the detection of 204 LSBG candidates, characterized by a mean surface brightness range of $23.5\ \mathrm{ mag}\ \mathrm{ arcsec}^{-2}\le \bar{\mu }_{\text{eff}}(g)\le 26.8\ \mathrm{ mag}\ \mathrm{ arcsec}^{-2}$ and a half-light radius range of 1.4 arcsec ≤ r1/2 ≤ 8.3 arcsec. Notably, 116 LSBG candidates exhibit a half-light radius ≥2.5 arcsec. These results affirm the remarkable performance of our model in detecting LSBGs, making it a promising tool for the upcoming CSST.

Su, Hao (ORCID:0009000778888301)↗

Grid-Interactive Multi-Zone Building Control Using Reinforcement Learning with Global-Local Policy Search

In this paper, we develop a grid-interactive multi-zone building controller based on a deep reinforcement learning (RL) approach. The controller is designed to facilitate building operation during normal conditions and demand response events, while ensuring occupants comfort and energy efficiency. We leverage a continuous action space RL formulation, and devise a two-stage global-local RL training framework. In the first stage, a global fast policy search is performed using a gradient-free RL algorithm. In the second stage, a local fine-tuning is conducted using a policy gradient method. In contrast to the state-of-the-art model predictive control (MPC) approach, the proposed RL controller does not require complex computation during real-time operation and can adapt to nonlinear building models. We illustrate the controller performance numerically using a five-zone commercial building.

30 DIRECT ENERGY CONVERSION↗

Digital Technologies at NASA for Science and Engineering

While scientific and engineering advancements used to rely primarily on theoretical studies and physical experiments, today digital technology enabled by petaflops-scale supercomputers is an equal, if not a greater, contributor to such achievements. In addition, computational modeling and simulation serves as a predictive tool that is not otherwise available. As a result, the use of high performance computing is integral to NASA's work in all mission areas such as space exploration, aeronautics, and scientific discovery. But traditional supercomputing alone is not sufficient for all of the space agency's needs. The success of many NASA missions depends on solving complex computing challenges, some of which are NP-hard (decision theory) if using classical solution methods. Quantum computing promises an unprecedented ability to solve such intractable problems by harnessing quantum mechanical effects such as tunneling, superposition, and entanglement. Another disruptive digital technology is neuromorphic computing that uses brain-inspired lessons to generate new architectures that are much more energy efficient, and capable of massive parallel processing and learning in-situ. Finally, with large amounts of observational and computational data sets, the opportunities of big data and data analytics can be leveraged to enable deep learning and knowledge discovery - it's all a massive digital transformation. This talk will be an overview how NASA utilizes digital technologies for its science and engineering efforts.

Biswas, Rupak↗

Grid-Interactive Multi-Zone Building Control Using Reinforcement Learning with Global-Local Policy Search: Preprint

In this paper, we develop a grid-interactive multi-zone building controller based on a deep reinforcement learning (RL) approach. The controller is designed to facilitate building operation during normal conditions and demand response events, while ensuring occupants comfort and energy efficiency. We leverage a continuous action space RL formulation, and devise a two-stage global-local RL training framework. In the first stage, a global fast policy search is performed using a gradient-free RL algorithm. In the second stage, a local fine-tuning is conducted using a policy gradient method. In contrast to the state-of-the-art model predictive control (MPC) approach, the proposed RL controller does not require complex computation during real-time operation and can adapt to non-linear building models. We illustrate the controller performance numerically using a five-zone commercial building.

30 DIRECT ENERGY CONVERSION↗

Understanding hadronic interactions using advanced reconstruction techniques

ProtoDUNE-SP served as the prototype of the future Deep Underground Neutrino Experiment (DUNE). With a total liquid argon mass of 0.77 kt, it stood as the most extensive monolithic single-phase Liquid Argon Time Projection Chamber (LArTPC) ever constructed. Strategically located and operated at CERN, ProtoDUNE-SP benefited from a specialized charged-particle test beam, offering a momentum range of 0.3 - 7 GeV/c. This setup provided a unique opportunity to conduct in-depth studies on hadronic interactions within argon. Hadronic interactions play a fundamental role in neutrino physics, particularly in the intricacies of neutrino detection. We depend on analyzing the outcomes of these interactions to accurately determine both the flavor and energy of the interacting neutrinos. Pions, as one of the primary byproducts of neutrino interactions, are crucial for understanding the overall dynamics and kinematics of these processes. Accurately characterizing and understanding the interaction of pions with argon can significantly enhance the precision of neutrino simulations and measurements.Being a Liquid Argon Time Projection Chamber (LArTPC), ProtoDUNE-SP is distinguished by its ability to produce high-definition images of charged particles as they traverse through the detector's active volume. Yet, accurately reconstructing these particle interactions and determining their kinematic attributes remains a challenging task. The successes of deep learning in diverse domains, especially in image recognition, provide a promising approach for addressing this challenge in ProtoDUNE. This thesis discusses a measurement of the Piplus - argon inelastic cross section in the energy range of 2400 to 3000 MeV, using data taken by protoDUNE during the fall of 2018. Additionally, panoptic segmentation, a machine learning technique, is introduced and validated through its application in the reconstruction of the neutral pion rest mass. The aim is to showcase how these advanced methods can enhance the quality of event reconstruction.

Sarasty Segura, Carlos Eduardo↗

AI-enabled traveling wave protection for microgrids

Grid forming inverters provide voltage and frequency regulations for microgrids; in the meantime, new challenges are introduced for microgrid protections. For instance, inverters’ control strategies can affect protection behaviors, and low short-circuit ratios and bi-directional power flows also make protection operations complex. Protection schemes based on conventional principles such as overcurrent and distance relays do not always provide reliable, sensitive, or selective operations. Here, we devise a traveling wave protection approach for microgrids using a wavelet-driven deep neural network named WaveletKernelNet (WKN). Compared with conventional methods, the presented approach provides enhanced sensitivity, higher selectivity, and better identification of various faults in microgrids. Extensive case studies validate the efficacy and excellent performance of the devised approach.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Wattile: Probabilistic Deep Learning-based Forecasting of Building Energy Consumption [SWR-20-94]

Accurate energy forecasting is becoming critical due to many reasons: i ) optimal distributed energy resources operations and dispatch, ii) fault detection and diagnostics, and iii) meeting operational energy efficiency targets. Wattile uses deep learning (DL) for the building's short-term load forecasting application. Two specific types of neural networks called, Long Short Term Memory (LSTM) and Sequence-to-Sequence (S2S) models are used to make predictions. Forecasting models are trained using online historical weather and occupancy indicator data streams from the Intelligent Campus Program's data acquisition systems at the National Renewable Energy Laboratory (NREL) for main meters and sub-meters of multiple building types. These models use probabilistic methods to provide quantile-based forecasts in addition to nominal conditional median predictions of electricity consumption.

Frank, Stephen↗

Deep Learning-based Non-Stationary Bias Correction (NSBC)

This work develops the NSBC (non-stationary bias correction) methodology to correct temperature projection bias from E3SM. The NSBC deep learning framework consists of a three-part architecture: an auto-encoder for compressing the spatial information, an LSTM for predicting annual temperature mean, and a U-Net for capturing the residual bias in temperature. The non-stationary bias correction (NSBC) framework can correct the non-stationarity of the biases of the climate models, which significantly improves the accuracy of future temperature prediction and improves the overestimation of extreme high temperatures that many existing bias correction methods suffer from. Getting started 1. Obtain the historical climate simulation and observation data. The E3SM simulation data are available through https://aims2.llnl.gov/search/cmip6/. The pseudo observations, the Geophysical Fluid Dynamics Laboratory (GFDL)-ESM4 model (Krasting et al., 2018) are available through https://aims2.llnl.gov/search/cmip6/. The spatial resolution of E3SM and pseudo observation datasets are both regridded to a common 1° resolution grid using conservative interpolation. The regridded E3SM and pseudo observation with 1° resolution can be found throught ./data/. 2. Train the Auto-encoder model. Python 0-autoencoder.py 3. Train the LSTM Python 1-LSTM.py 4. Generate the annual mean temperature based on trained LSTM Python 2-generate_annual_mean_LSTM.py 5. Train the U-Net. Python 3-unet.py 6. Evaluation and compared with the baseline Python 4_evaluation.py Is there a deadline approaching that requires the release of yo

Lucas, Donald↗

Investigating Carboxysome Morphology Dynamics with a Rotationally Invariant Variational Autoencoder

Carboxysomes are a class of bacterial microcompartments that form proteinaceous organelles within the cytoplasm of cyanobacteria and play a central role in photosynthetic metabolism by defining a cellular microenvironment permissive to CO2 fixation. Critical aspects of the assembly of the carboxysomes remain relatively unknown, especially with regard to the dynamics of this microcompartment. Progress in understanding carboxysome dynamics is impeded in part because analysis of the subtle changes in carboxysome morphology with microscopy remains a low-throughput and subjective process. Here we use deep learning techniques, specifically a Rotationally Invariant Variational Autoencoder (rVAE), to analyze fluorescence microscopy images of cyanobacteria bearing a carboxysome reporter and quantitatively evaluate how carboxysome shell remodelling impacts subtle trends in the morphology of the microcompartment over time. Toward this goal, we use a recently developed tool to control endogenous protein levels, including carboxysomal components, in the model cyanobacterium Synechococcous elongatus PCC 7942. By utilization of this system, proteins that compose the carboxysome can be tuned in real time as a method to examine carboxysome dynamics. We find that rVAEs are able to assist in the quantitative evaluation of changes in carboxysome numbers, shape, and size over time. Further, we propose that rVAEs may be a useful tool to accelerate the analysis of carboxysome assembly and dynamics in response to genetic or environmental perturbation and may be more generally useful to probe regulatory processes involving a broader array of bacterial microcompartments.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Reinforced dynamics for enhanced sampling in large atomic and molecular systems

A new approach for efficiently exploring the configuration space and computing the free energy of large atomic and molecular systems is proposed, motivated by an analogy with reinforcement learning. There are two major components in this new approach. Like metadynamics, it allows for an efficient exploration of the configuration space by adding an adaptively computed biasing potential to the original dynamics. Like deep reinforcement learning, this biasing potential is trained on the fly using deep neural networks, with data collected judiciously from the exploration and an uncertainty indicator from the neural network model playing the role of the reward function. Parameterization using neural networks makes it feasible to handle cases with a large set of collective variables. This has the potential advantage that selecting precisely the right set of collective variables has now become less critical for capturing the structural transformations of the system. Finally, the method is illustrated by studying the full-atom explicit solvent models of alanine dipeptide and tripeptide, as well as the system of a polyalanine-10 molecule with 20 collective variables.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A scalable algorithm for the optimization of neural network architectures

In this work, we propose a new scalable method to optimize the architecture of an artificial neural network. The proposed algorithm, called Greedy Search for Neural Network Architecture, aims to determine a neural network with minimal number of layers that is at least as performant as neural networks of the same structure identified by other hyperparameter search algorithms in terms of accuracy and computational cost. Numerical results performed on benchmark datasets show that, for these datasets, our method outperforms state-of-the-art hyperparameter optimization algorithms in terms of attainable predictive performance by the selected neural network architecture, and time-to-solution for the hyperparameter optimization to complete.

97 MATHEMATICS AND COMPUTING↗

Understanding the surface wave characteristics using 2D particle-in-cell simulation and deep neural network

Here, the characteristics of the surface waves along the interface between a plasma and a dielectric material have been investigated using kinetic particle-in-cell simulations. A microwave source of GHz frequency has been used to trigger the surface wave in the system. The outcome indicates that the surface wave gets excited along the interface of plasma and the dielectric tube and appears as light and dark patterns in the electric field profiles. The dependency of radiation pressure on the dielectric permittivity and supplied input frequency has been investigated. Further, we assessed the capabilities of neural networks to predict the radiation pressure for a given system. The proposed deep neural network model is aimed at developing accurate and efficient data-driven plasma surface wave devices.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Data-driven modeling of power generation for a coal power plant under cycling

Increased penetration of renewables for power generation has negatively impacted the dynamics of conventional fossil fuel-based power plants. The power plants operating on the base load are forced to cycle, to adjust to the fluctuating power demands. This results in an inefficient operation of the coal power plants, which leads up to higher operating losses. To overcome such operational challenge associated with cycling and to develop an optimal process control, this work analyzes a set of models for predicting power generation. Moreover, the power generation is intrinsically affected by the state of the power plant components, and therefore our model development also incorporates additional power plant process variables while forecasting the power generation. We present and compare multiple state-of-the-art forecasting data-driven methods for power generation to determine the most adequate and accurate model. We also develop an interpretable attention-based transformer model to explain the importance of process variables during training and forecasting. The trained deep neural network (DNN) LSTM model has good accuracy in predicting gross power generation under various prediction horizons with/without cycling events and outperforms the other models for long-term forecasting. The DNN memory-based models show significant superiority over other state-of-the-art machine learning models for short, medium and long range predictions. The transformer-based model with attention enhances the selection of historical data for multi-horizon forecasting, and also allows to interpret the significance of internal power plant components on the power generation. This newly gained insights can be used by operation engineers to anticipate and monitor the health of power plant equipment during high cycling periods.

01 COAL, LIGNITE, AND PEAT↗

Multi-resolution partial differential equations preserved learning framework for spatiotemporal dynamics

Traditional data-driven deep learning models often struggle with high training costs, error accumulation, and poor generalizability in complex physical processes. Physics-informed deep learning (PiDL) addresses these challenges by incorporating physical principles into the model. Most PiDL approaches regularize training by embedding governing equations into the loss function, yet this depends heavily on extensive hyperparameter tuning to weigh each loss term. To this end, we propose to leverage physics prior knowledge by “baking” the discretized governing equations into the neural network architecture via the connection between the partial differential equations (PDE) operators and network structures, resulting in a PDE-preserved neural network (PPNN). This method, embedding discretized PDEs through convolutional residual networks in a multi-resolution setting, largely improves the generalizability and long-term prediction accuracy, outperforming conventional black-box models. The effectiveness and merit of the proposed methods have been demonstrated across various spatiotemporal dynamical systems governed by spatiotemporal PDEs, including reaction-diffusion, Burgers’, and Navier-Stokes equations.

97 MATHEMATICS AND COMPUTING↗

Catalysis, dynamics and stability of enzymes under extreme conditions

This project led to an unprecedented understanding of how enzymes can catalyze chemical reactions with such high catalytic turnover numbers. This was achieved by a detailed description of the free energy landscape of adenylate kinase using a combination of experimental atomic resolution techniques and computation. In this project, a systematic approach was developed how to characterize free energy landscapes with these biophysical methods. These approaches are directly transferable to other enzymes. In addition, we recreated the evolutionary pathway of enzyme evolution over about 3.5 billion years. The results shed light into how enzymes adapted to the decrease in temperature by increasing their catalytic power at lower temperatures. The data demonstrate fundamental principles of evolution such as active pressure versus passive drift. The adaptation of biocatalyst to extreme conditions such as high or low temperature, pressure (living deep at sea) or high salt was studied. Lessons learned from this project have major implications for catalysis in biotechnology.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗