Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “generative machine 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.

At least 1,441 records · Page 80

Advancing Geothermal Research: Fiscal Year 2025 Accomplishments Report

This is a summary of geothermal work done at the National Laboratory of the Rockies in Fiscal Year 2025. This year brought increased attention to the geothermal industry and NLR's geothermal research portfolio. With more than 70 active projects, NLR research spanned the areas of resource exploration and characterization; conventional and next-generation geothermal technologies; subsurface thermal energy storage; heating and cooling; co-production of geothermal with critical minerals and oil and gas; modeling and analysis leveraging expertise in data science and machine learning; and more.

15 GEOTHERMAL ENERGY↗

TRANSFER LEARNING FOR FIELD EMISSION MITIGATION IN CEBAF SRF CAVITIES

The Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab operates hundreds of super-conducting radio frequency (SRF) cavities in its two linear accelerators (linacs). Field emission (FE) is an ongoing operational challenge in higher gradient SRF cavities. FE generates high levels of neutron and gamma radiation leading to damaged accelerator hardware and a radiation hazard environment. During machine development periods, we performed gradient scans to record data capturing the relationship between cavity gradients and radiation levels measured throughout the linacs. However, the field emission environment at CEBAF varies considerably over time as the configuration of the radio frequency (RF) gradients changes and due to the changing behaviour of field emitters. An artificial intelligence/machine learning (AI/ML) approach with transfer learning could be a valuable tool to mitigate FE and lower the radiation levels. In this work, we mainly focus on leveraging the RF trip data gathered during CEBAF operations. We develop a transfer learning-based surrogate model for radiation detector readings given RF cavity gradients to track the CEBAF?s changing configuration and environment. Then, we could use the developed model as an optimization process for redistributing the RF gradients within a linac to minimize radiation levels.

Ahammed, K.↗

TRANSFER LEARNING FOR FIELD EMISSION MITIGATION IN CEBAF SRF CAVITIES

The Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab operates hundreds of super-conducting radio frequency (SRF) cavities in its two linear accelerators (linacs). Field emission (FE) is an ongoing operational challenge in higher gradient SRF cavities. FE generates high levels of neutron and gamma radiation leading to damaged accelerator hardware and a radiation hazard environment. During machine development periods, we performed gradient scans to record data capturing the relationship between cavity gradients and radiation levels measured throughout the linacs. However, the field emission environment at CEBAF varies considerably over time as the configuration of the radio frequency (RF) gradients changes and due to the changing behaviour of field emitters. An artificial intelligence/machine learning (AI/ML) approach with transfer learning could be a valuable tool to mitigate FE and lower the radiation levels. In this work, we mainly focus on leveraging the RF trip data gathered during CEBAF operations. We develop a transfer learning-based surrogate model for radiation detector readings given RF cavity gradients to track the CEBAF?s changing configuration and environment. Then, we could use the developed model as an optimization process for redistributing the RF gradients within a linac to minimize radiation levels.

Ahammed, K.↗

Improvement of machine learning enhanced genetic algorithm for nonlinear beam dynamics optimization

On top of the genetic algorithm enhanced by machine learning for nonlinear lattice optimization, as proposed in Li et al. (2018), an improved repopulation technique has been developed. Different weight coefficients for defining the “elite cluster” were compared to discern the fastest convergence in two classic optimization test problems. Additonally, the volume of the parameter space for generating potentially competitive candidates was further confined by repopulation in the vicinity of randomly selected “elite seeds”. The new repopulation technique significantly improves the quality of newly populated candidates by excluding the less competitive seeds introduced by the old repopulation algorithm. Finally, this technique has been validated, having a faster convergence for test problems first, and then applied to the nonlinear lattice optimization for the High Energy Photon Source storage ring.

36 MATERIALS SCIENCE↗

RU-net for automatic characterization of TRISO fuel cross sections

During irradiation, phenomena such as kernel swelling and buffer densification may impact the performance of tristructural isotropic (TRISO) particle fuel. Post-irradiation microscopy is often used to identify these irradiation-induced morphologic changes. However, each fuel compact generally contains thousands of TRISO particles. Manually performing the work to get statistical information on these phenomena is cumbersome and subjective. Here, to reduce the subjectivity inherent in that process and to accelerate data analysis, we used convolutional neural networks (CNNs) to automatically segment cross-sectional images of microscopic TRISO layers. CNNs are a class of machine-learning algorithms specifically designed for processing structured grid data. They have gained popularity in recent years due to their remarkable performance in various computer vision tasks, including image classification, object detection, and image segmentation. In this research, we generated a large irradiated TRISO layer dataset with more than 2,000 microscopic images of cross-sectional TRISO particles and the corresponding annotated images. Based on these annotated images, we used different CNNs to automatically segment different TRISO layers. These CNNs include RU-Net (developed in this study), as well as three existing architectures: U-Net, Residual Network (ResNet), and Attention U-Net. The preliminary results show that the model based on RU-Net performs best in terms of Intersection over Union (IoU). Using CNN models, we can expedite the analysis of TRISO particle cross sections, significantly reducing the manual labor involved and improving the objectivity of the segmentation results.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

ERRATUM: “Automated Transient Identification in the Dark Energy Survey” (2015, AJ, 150, 82)

Here, we describe an algorithm for identifying point-source transients and moving objects on reference-subtracted optical images containing artifacts of processing and instrumentation. The algorithm makes use of the supervised machine learning technique known as Random Forest. We present results from its use in the Dark Energy Survey Supernova program (DES-SN), where it was trained using a sample of 898,963 signal and background events generated by the transient detection pipeline. After reprocessing the data collected during the first DES-SN observing season (2013 September through 2014 February) using the algorithm, the number of transient candidates eligible for human scanning decreased by a factor of 13.4, while only 1.0% of the artificial Type Ia supernovae (SNe) injected into search images to monitor survey efficiency were lost, most of which were very faint events. Here we characterize the algorithm's performance in detail, and we discuss how it can inform pipeline design decisions for future time-domain imaging surveys, such as the Large Synoptic Survey Telescope and the Zwicky Transient Facility. An implementation of the algorithm and the training data used in this paper are available at at http://portal.nersc.gov/project/dessn/autoscan.

79 ASTRONOMY AND ASTROPHYSICS↗

Winners are not keepers: Characterizing household engagement, gains, and energy patterns in demand response using machine learning in the United States

Demand-response programs can help utilities manage rapidly evolving electric grids, but these programs are subject to the complexities of human behavior. This paper explores a novel method for uncovering heterogeneity in households. In this work, we use a machine-learning method known as a Conditional Inference Tree (c-tree) algorithm to categorize households based on their energy behavior characteristics collected via smart meters, and explore how this translates through into heterogeneity in their real-world response to a DR program. Using data from randomized controlled trial, we generate estimates of the changes in energy use caused by the program within each household group. Our results show that the c-tree approach differentiates households by their energy-use characteristics in a way that increases the spread in enrollment rates and critical peak reduction among household groups, compared with the spreads achieved via several conventional segmentation methods. Thus, the c-tree analysis enables the most tailored targeting of major potential energy savers and could provide the greatest increase in cost-effectiveness of household recruitment into DR programs. Our results also offer fresh insights into the relationships between household energy behavior characteristics – such as peak energy use and “structural winningness” (the ability to save money under a DR program without changing energy-use behaviors) – and household decisions about enrolling in DR programs and reducing energy use. Our research also demonstrates the potential of smart meter data, combined with machine learning and econometric methods, to provide significant value to utilities, program implementers, researchers, and other stakeholders.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Machine learning for Deep Space Network antenna motions detection

Highly stable frequency and timing standards are essential for deep-space missions and radio science. At the NASA Deep Space Network (DSN), these standards are distributed through a network of underground fiber cables to support several Goldstone antennas. Independently developed frequency-measuring instruments generate tremendous quantities of data to monitor and validate the antennas’ stringent frequency requirements. In this paper, we propose a lightweight processing tool capable of detecting disturbances on the frequency signal caused by DSN antenna motions. Our training data is sampled from the movement log of the antenna of interest and the generated data from the fiber optic metrology instrument linked to the antenna. We demonstrate that a convolutional neural network (CNN) model can achieve high accuracies on classifying instances of antenna movements and is an effective predictor when used iteratively on longer, variable stretches of metrology data. The simplicity, low training cost, and high accuracies of our model strongly suggest its efficacy in identifying and troubleshooting frequency disturbances caused by the antenna.

Yi, Lin↗

Multi-omics Characterization of the Host Response to COVID-19

This project is a multi-disciplinary collaboration between investigators at PNNL with expertise in mass spectrometry (MS)-based omics technology development, omics measurement methods development and application, statistics, machine learning and integration of disparate datasets for a systems-level understanding, and expertise in pathogenic coronaviruses, and investigators at the University of Wisconsin-Madison (UW-Madison) with expertise in pathogenic respiratory viruses (e.g. influenza). The goal of this project is to obtain a comprehensive picture of the human host factors critical for the outcome of SARS-CoV-2 infection. We will generate broad untargeted multi-omics profiles using both state-of-the-art and novel instrumentation and approaches to enable the identification of the molecular mechanisms and host response pathways that impact human COVID-19 outcomes. We anticipate these results will lead to the generation of biomarker panels that are predictive of disease outcomes and mechanistic hypotheses that can be further interrogated in future studies and will provide the basis for vaccine or therapeutic development. To do so, we are obtaining and analyzing blood samples from COVID-19 patients with a range of disease outcomes that were treated at the Center Hospital of the National Center for Global Health and Medicine in Tokyo, Japan and other collaborating hospitals in our network. Specifically, this project will fund proteomics and metabolomics analyses of clinical COVID samples, machine learning-based integration of the data, and pathway-based interpretation of the data. This project was funded in June 2020. In the time span of June to September 2020, the project team developed an analytically and statistically robust analysis plan and made various preparations to facilitate sample receipt from our UW-Madison collaborators. This included blocking and randomization of sample prep orders, ordering of reagents and reference materials, and shipping of materials needed for preparation of the samples under BSL3 conditions to our collaborators at UW-Madison. As of FY21, this project has been picked up via a sponsor, the Naval Medical Research Center, which will cover the remainder of the proposed scope of work.

60 APPLIED LIFE SCIENCES↗

Physics-Informed Machine Learning Models for Predicting the Progress of Reactive-Mixing

This paper presents a physics-informed machine learning (ML) framework to construct reduced-order models (ROMs) for reactive-transport quantities of interest (QoIs) based on high-fidelity numerical simu-lations. QoIs include species decay, product yield, and degree of mixing. The ROMs for QoIs are applied to quantify and understand how the chemical species evolve over time. First, high-resolution datasets for constructing ROMs are generated by solving anisotropic reaction-di?usion equations using a non-negative finite element formulation for di?erent input parameters. The reactive-mixing model input parameters are: time-scale associated with flipping of velocity, spatial-scale controlling small/large vortex structures of velocity, perturbation parameter of the vortex-based velocity, anisotropic dispersion strength/contrast, and molecular diffusion. Second, random forests, F-test, and mutual information criterion are used to evaluate the importance of model inputs/features with respect to QoIs. We observed that anisotropic dispersion strength/contrast is the most important feature and time-scale associated with flipping of velocity is the least important feature. Third, Support Vector Machines (SVM) and Support Vector Regression (SVR) are used to construct ROMs based on the model inputs. The constructed SVR-ROMs are then used to predict scaling of QoIs. We also present estimates and inequalities on the QoIs, which inform that the species decay, mix, and produce in an exponential fashion. These inequalities also inform that a radial basis function is the most suitable kernel for the SVM/SVR models for QoIs. It is observed that R2-score for SVR-ROMs on unseen data is greater than 0.9, implying that the SVR-ROMs are able to predict the reaction-diffusion system state reasonably well. Finally, in terms of the computational cost, the proposed SVM-ROMs are O(107) times faster than running a high-fidelity finite element simulation for evaluating QoIs. This makes the proposed ML-based ROMs attractive for reactive-transport sensing and real-time monitoring applications as they are significantly faster yet reasonably accurate.

Mudunuru, Maruti K.↗

Benchmarking structural evolution methods for training of machine learned interatomic potentials

When creating training data for machine-learned interatomic potentials (MLIPs), it is common to create initial structures and evolve them using molecular dynamics (MD) to sample a larger configuration space. Here, we benchmark two other modalities of evolving structures, contour exploration (CE) and dimer-method (DM) searches against MD for their ability to produce diverse and robust density functional theory training data sets for MLIPs. We also discuss the generation of initial structures which are either from known structures or from random structures in detail to further formalize the structure-sourcing processes in the future. The polymorph-rich zirconium-oxygen composition space is used as a rigorous benchmark system for comparing the performance of MLIPs trained on structures generated from these structural evolution methods. Using Behler–Parrinello neural networks as our MLIP models, we find that CE and the DM searches are generally superior to MD in terms of spatial descriptor diversity and statistical accuracy.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Neural network modeling of in situ fluid-filled pore size distributions in subsurface shale reservoirs under data constraints

Subsurface nuclear magnetic resonance (NMR) logs acquired in the wellbore environment are sensitive to fluid-filled pore size distribution, fluid mobility, permeability, and porosity in the near-wellbore reservoir volume. NMR response of a formation layer is processed to extract the T2 distribution, which approximates the fluid-filled pore size distribution. NMR logs are acquired in limited number of wells due to financial and operational challenges, which adversely affects reservoir characterization. We developed here two neural-network-based machine learning techniques, long short-term memory (LSTM) network and variational autoencoder with a convolutional layer (VAEc) network, to process the ‘easy-to-acquire’ formation mineral and fluid saturation logs to generate synthetic NMR T2 distributions in the absence of ‘hard-to-acquire’ NMR T2 distribution log. Both the predictive models are trained and tested on limited wireline log measurements randomly selected from a 300-ft depth interval of the Bakken shale formation. Synthesis performances of LSTM and VAEc models in terms of R 2 are 0.78 and 0.75, respectively. Noise is inevitable in logging data due to the complex wellbore and formation conditions. Notably, both the predictive models robustly synthesize the fluid-filled pore size distributions in the presence of 50% noise in input logs and 30% noise in training T2 data. The performance of the proposed methodology improves with access to larger volume of training data from other formation types. The proposed method is critical to the synthesis of in situ fluid-filled pore size distributions in shale formations under data constraints due to financial and operational challenges.

58 GEOSCIENCES↗

DeepCMB: Lensing reconstruction of the cosmic microwave background with deep neural networks

Next-generation cosmic microwave background (CMB) experiments will have lower noise and therefore increased sensitivity, enabling improved constraints on fundamental physics parameters such as the sum of neutrino masses and the tensor-to-scalar ratio r. Achieving competitive constraints on these parameters requires high signal-to-noise extraction of the projected gravitational potential from the CMB maps. Standard methods for reconstructing the lensing potential employ the quadratic estimator (QE). However, the QE performs suboptimally at the low noise levels expected in upcoming experiments. Other methods, like maximum likelihood estimators (MLE), are under active development. In this work, we demonstrate reconstruction of the CMB lensing potential with deep convolutional neural networks (CNN) - ie, a ResUNet. The network is trained and tested on simulated data, and otherwise has no physical parametrization related to the physical processes of the CMB and gravitational lensing. We show that, over a wide range of angular scales, ResUNets recover the input gravitational potential with a higher signal-to-noise ratio than the QE method, reaching levels comparable to analytic approximations of MLE methods. We demonstrate that the network outputs quantifiably different lensing maps when given input CMB maps generated with different cosmologies. We also show we can use the reconstructed lensing map for cosmological parameter estimation. This application of CNN provides a few innovations at the intersection of cosmology and machine learning. First, while training and regressing on images, we predict a continuous-variable field rather than discrete classes. Second, we are able to establish uncertainty measures for the network output that are analogous to standard methods. We expect this approach to excel in capturing hard-to-model non-Gaussian astrophysical foreground and noise contributions.

79 ASTRONOMY AND ASTROPHYSICS↗

Interpretable artificial intelligence and exascale molecular dynamics simulations to reveal kinetics: Applications to Alzheimer's disease

The rapid increase in computing power, especially with the integration of graphics processing units, has dramatically increased the capabilities of molecular dynamics simulations. To date, these capabilities extend from running very long simulations (tens to hundreds of microseconds) to thousands of short simulations. However, the expansive data generated in these simulations must be made interpretable not only by the investigator who performs them but also by others as well. Here, we demonstrate how integrating learning techniques, such as artificial intelligence, machine learning, and neural networks, into analysis pipelines can reveal the kinetics of Alzheimer's disease (AD) protein aggregation. Finally, we review select AD targets, describe current simulation methods, and introduce learning concepts and their application in AD, highlighting limitations and potential solutions.

59 BASIC BIOLOGICAL SCIENCES↗

Assessment of fine-tuned large language models for real-world chemistry and material science applications

The current generation of large language models (LLMs) has limited chemical knowledge. Recently, it has been shown that these LLMs can learn and predict chemical properties through fine-tuning. Using natural language to train machine learning models opens doors to a wider chemical audience, as field-specific featurization techniques can be omitted. In this work, we explore the potential and limitations of this approach. We studied the performance of fine-tuning three open-source LLMs (GPT-J-6B, Llama-3.1-8B, and Mistral-7B) for a range of different chemical questions. We benchmark their performances against “traditional” machine learning models and find that, in most cases, the fine-tuning approach is superior for a simple classification problem. Depending on the size of the dataset and the type of questions, we also successfully address more sophisticated problems. The most important conclusions of this work are that, for all datasets considered, their conversion into an LLM fine-tuning training set is straightforward and that fine-tuning with even relatively small datasets leads to predictive models. These results suggest that the systematic use of LLMs to guide experiments and simulations will be a powerful technique in any research study, significantly reducing unnecessary experiments or computations.

Van Herck, Joren↗

Adaptive PV Frequency Control Strategy Based on Real-time Inertia Estimation

The declining cost of solar Photovoltaics (PV) generation is driving its worldwide deployment. As conventional generation with large rotating masses is being replaced by renewable energy such as PV, the power system’s inertia will be affected. As a result, the system’s frequency may vary more dramatically in the case of a disturbance, and the frequency nadir may be low enough to trigger protection relays such as under-frequency load shedding. The existing frequency-watt function mandated in power inverters cannot provide grid frequency support in a loss-of-generation event, as PV plants usually do not have power reserves. Here, a novel adaptive PV frequency control strategy is proposed to reserve the minimum power required for grid frequency support. A machine learning model is trained to predict system frequency response under varying system conditions, and an adaptive allocation of PV headroom reserves is made based on the machine learning model as well as real-time system conditions including inertia. Case studies show the proposed control method meets the frequency nadir requirements using minimal power reserves compared to a fixed headroom control approach.

14 SOLAR ENERGY↗

ASAP: Automatic Synthesis of Area-Efficient and Precision-Aware CGRAs

Coarse-grained reconfigurable accelerators (CGRAs) are a promising accelerator design choice that strikes a balance between performance and adaptability to different computing patterns across various applications domains. Designing a CGRA for a specific application domain involves enormous software/hardware engineering effort. Recent research works explore loop transformations, functional unit types, network topology, and memory size to identify optimal CGRA designs given a set of kernels from a specific application do- main. Unfortunately, the impact of functional units with different precision support has rarely been investigated. To address this gap, we propose ASAP – a hardware/software co-design framework that automatically identifies and synthesizes optimal precision-aware CGRA for a set of applications of interest. Our evaluation shows that ASAP generates specialized designs 3.2×, 4.21×, and 5.8× more efficient (in terms of performance per unit of energy or area) than non-specialized homogeneous CGRAs, for the scientific computing, embedded, and edge machine learning domains, respectively, with limited accuracy loss. Moreover, ASAP provides more efficient designs than other state-of-the-art synthesis frameworks for specialized CGRAs.

artificial intelligence↗

Halide Perovskite Solar Photovoltaics

Technological progress in photovoltaic (PV) technologies provides hope that a comprehensive and desperately needed decarbonization of the energy sector is possible. Commercially successful PV technologies based predominantly on silicon wafer technology are reliable and cost-effective, but remain capital- and carbon-intensive. In this context, emerging PV technologies, such as metal-halide perovskites (MHPs), could further catalyze the energy transition by providing technological opportunities for even lower-cost, mass-producible, high-efficiency solar cells with a significantly reduced "carbon footprint." This themed issue of MRS Bulletin on "Halide perovskite solar photovoltaics summarizes the current state of the art, challenges, and opportunities of perovskite photovoltaics with contributions and perspectives from six expert teams worldwide. The topics covered provide a status update on perovskite PV, remaining hurdles to their deployment, and challenges to realizing the potential of this technology to impact climate goals. Articles in this collection address scalability of perovskite PV and prospects for industrial manufacturing; perovskite PV as an add-on technology on top of commercial silicon PV; environmental and sustainability considerations; and durability and reliability considerations. Further considerations include prospects of automation, coupled to artificial intelligence and machine learning, for accelerating material-based solutions to these outstanding challenges including the possibilities of discovering new absorber and device component materials to enable success and ultimately deployment of these next-generation PVs.

metal-halide perovskites↗