SEARCH · Engineering Papers

Results for “training data quality”

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.

SourceResearch connectionFrom yearThrough year

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

At least 199 records · Page 11

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Dynamic Monkey Bar Mechanism of Superionic Li‐ion Transport in LiTaCl 6

Abstract The LiTaCl 6 solid electrolyte has the lowest activation energy of ionic conduction at ambient conditions (0.165 eV), with a record high ionic conductivity for a ternary compound (11 mS cm −1 ). However, the mechanism has been unclear. We train machine‐learning force fields (MLFF) on ab initio molecular dynamics (AIMD) data on‐the‐fly and perform MLFF MD simulations of AIMD quality up to the nanosecond scale at the experimental temperatures, which allows us to predict accurate activation energy for Li‐ion diffusion (at 0.164 eV). Detailed analyses of trajectories and vibrational density of states show that the large‐amplitude vibrations of Cl − ions in TaCl 6 − enable the fast Li‐ion transport by allowing dynamic breaking and reforming of Li−Cl bonds across the space in between the TaCl 6 − octahedra. We term this process the dynamic‐monkey‐bar mechanism of superionic Li + transport which could aid the development of new solid electrolytes for all‐solid‐state lithium batteries.

Lei, Ming↗

DOE OSTI2024

Active and Transfer Learning of High-Dimensional Neural Network Potentials for Transition Metals

Classical molecular dynamics (MD) simulations represent a very popular and powerful tool for materials modeling and design. The predictive power of MD hinges on the ability of the interatomic potential to capture the underlying physics and chemistry. There have been decades of seminal work on developing interatomic potentials, albeit with a focus predominantly on capturing the properties of bulk materials. Such physics-based models, while extensively deployed for predicting the dynamics and properties of nanoscale systems over the past two decades, tend to perform poorly in predicting nanoscale potential energy surfaces (PESs) when compared to high-fidelity first-principles calculations. These limitations stem from the lack of flexibility in such models, which rely on a predefined functional form. Machine learning (ML) models and approaches have emerged as a viable alternative to capture the diverse size-dependent cluster geometries, nanoscale dynamics, and the complex nanoscale PESs, without sacrificing the bulk properties. Here, in this study, we introduce an ML workflow that combines transfer and active learning strategies to develop high-dimensional neural networks (NNs) for capturing the cluster and bulk properties for several different transition metals with applications in catalysis, microelectronics, and energy storage, to name a few. Our NN first learns the bulk PES from the high-quality physics-based models in literature and subsequently augments this learning via retraining with a higher-fidelity first-principles training data set to concurrently capture both the nanoscale and bulk PES. Our workflow departs from status-quo in its ability to learn from a sparsely sampled data set that nonetheless covers a diverse range of cluster configurations from near-equilibrium to highly nonequilibrium as well as learning strategies that iteratively improve the fingerprinting depending on model fidelity. All the developed models are rigorously tested against an extensive first-principles data set of energies and forces of cluster configurations as well as several properties of bulk configurations for 10 different transition metals. Our approach is material agnostic and provides a methodology to transfer and build upon the learnings from decades of seminal work in molecular simulations on to a new generation of ML-trained potentials to accelerate materials discovery and design.

36 MATERIALS SCIENCE↗

NASA NTRS1977

Remote sensing data processing: Two years ago, today, and two years from today

Beginning with a survey of the state-of-the-art of processing remotely sensed data in early 1975, significant developments between that time and the present are chronicled, and technologies for early 1979 are projected. Current technical issues discussed include: training selection and labeling; classification and mensuration; use of satellite indicators to supplement predictions; small scale field structures; physical factors; ancillary data; geometric quality; and the cost of processing.

Holmes, Q. A.↗

DOE OSTI2023

Data-Centric AI and the Open Energy Data Initiative (OEDI)

This presentation emphasizes the critical importance of data-centric AI. The limitations of model-centric AI when dealing with poor or insufficient data are highlighted, and it is illustrated how training models on inaccurate or noisy data leads to suboptimal results. This talk advocates for a hybrid approach that combines a focus on data quality and model parameters to achieve optimal results. The Open Energy Data Initiative (OEDI) is introduced as a valuable resource for obtaining high-quality energy-related datasets, hosting nearly 2,000 publicly accessible datasets, including 99 solar-related datasets, totaling over 2.7 petabytes of data. OEDI's data lakes enable users to query and work with data without extensive transfers. In conclusion, the significance of data-centric AI and adherence to data curation best practices is emphasized, positioning OEDI as a prime source of high-quality data for AI and machine learning in the renewable energy sector.

AI↗

DOE OSTI2024

PMU Data Quality and Sensor Health Monitoring

Phasor Measurement Units (PMUs) play a critical role in the evolution of the electric power industry by providing high-precision, real-time monitoring of essential power system metrics. However, effectively detecting abnormalities and critical events from PMU data is a complex task, complicated by intricate temporal patterns, a scarcity of labeled data for training algo- rithms, and constraints on online computational power. In this study, we apply TranAD, an innovative algorithm that combines transformer architectures with the refinement of adversarial learning, to both synthetic and real-world PMU datasets for developing a data quality and sensor online health monitoring platform for utilities. Our findings reveal that TranAD not only provides efficient detection and localization but also enhances the detail with which abnormalities are detected, marking a a significant step forward in the field of clean data acquisition processes for power system monitoring

deep neural network, machine learning (ML)↗

DOE OSTI2021

A study of real-world micrograph data quality and machine learning model robustness

Abstract Machine-learning (ML) techniques hold the potential of enabling efficient quantitative micrograph analysis, but the robustness of ML models with respect to real-world micrograph quality variations has not been carefully evaluated. We collected thousands of scanning electron microscopy (SEM) micrographs for molecular solid materials, in which image pixel intensities vary due to both the microstructure content and microscope instrument conditions. We then built ML models to predict the ultimate compressive strength (UCS) of consolidated molecular solids, by encoding micrographs with different image feature descriptors and training a random forest regressor, and by training an end-to-end deep-learning (DL) model. Results show that instrument-induced pixel intensity signals can affect ML model predictions in a consistently negative way. As a remedy, we explored intensity normalization techniques. It is seen that intensity normalization helps to improve micrograph data quality and ML model robustness, but microscope-induced intensity variations can be difficult to eliminate.

36 MATERIALS SCIENCE↗

DOE OSTI2024

Missing Wedge Completion via Unsupervised Learning with Coordinate Networks

Cryogenic electron tomography (cryoET) is a powerful tool in structural biology, enabling detailed 3D imaging of biological specimens at a resolution of nanometers. Despite its potential, cryoET faces challenges such as the missing wedge problem, which limits reconstruction quality due to incomplete data collection angles. Recently, supervised deep learning methods leveraging convolutional neural networks (CNNs) have considerably addressed this issue; however, their pretraining requirements render them susceptible to inaccuracies and artifacts, particularly when representative training data is scarce. To overcome these limitations, we introduce a proof-of-concept unsupervised learning approach using coordinate networks (CNs) that optimizes network weights directly against input projections. This eliminates the need for pretraining, reducing reconstruction runtime by 3–20× compared to supervised methods. Our in silico results show improved shape completion and reduction of missing wedge artifacts, assessed through several voxel-based image quality metrics in real space and a novel directional Fourier Shell Correlation (FSC) metric. Our study illuminates benefits and considerations of both supervised and unsupervised approaches, guiding the development of improved reconstruction strategies.

42 ENGINEERING↗

DOE OSTI2022

AutoPhaseNN: unsupervised physics-aware deep learning of 3D nanoscale Bragg coherent diffraction imaging

Abstract The problem of phase retrieval underlies various imaging methods from astronomy to nanoscale imaging. Traditional phase retrieval methods are iterative and are therefore computationally expensive. Deep learning (DL) models have been developed to either provide learned priors or completely replace phase retrieval. However, such models require vast amounts of labeled data, which can only be obtained through simulation or performing computationally prohibitive phase retrieval on experimental datasets. Using 3D X-ray Bragg coherent diffraction imaging (BCDI) as a representative technique, we demonstrate AutoPhaseNN, a DL-based approach which learns to solve the phase problem without labeled data. By incorporating the imaging physics into the DL model during training, AutoPhaseNN learns to invert 3D BCDI data in a single shot without ever being shown real space images. Once trained, AutoPhaseNN can be effectively used in the 3D BCDI data inversion about 100× faster than iterative phase retrieval methods while providing comparable image quality.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Electrical Load Forecasting Over Multihop Smart Metering Networks With Federated Learning

Electric load forecasting is essential for power management and stability in smart grids. This is mainly achieved via advanced metering infrastructure, where smart meters (SMs) record household energy data. Traditional machine learning (ML) methods are often employed for load forecasting, but require data sharing, which raises data privacy concerns. Federated learning (FL) can address this issue by running distributed ML models at local SMs without data exchange. However, current FL-based approaches struggle to achieve efficient load forecasting due to imbalanced data distribution across heterogeneous SMs. Here, this article presents a novel personalized FL (PFL) method for high-quality load forecasting in metering networks. A meta-learning-based strategy is developed to address data heterogeneity at local SMs in the collaborative training of local load forecasting models. Moreover, to minimize the load forecasting delays in our PFL model, we study a new latency optimization problem based on optimal resource allocation at SMs. A theoretical convergence analysis is also conducted to provide insights into FL design for federated load forecasting. Extensive simulations from real-world datasets show that our method outperforms existing approaches regarding better load forecasting and reduced operational latency costs.

Rahman, Ratun [Univ. of Alabama, Huntsville, AL (U↗

NASA NTRS2005

Experimental Study of the Influence of Speed and Load on Thermal Behavior of High-Speed Helical Gear Trains

An experimental effort has been conducted on an aerospace-quality helical gear train to investigate the thermal behavior of the gear system as speed, load, and lubricant flow rate were varied. Temperature test data from a helical gear train at varying speeds and loads (to 5000 hp and 15000 rpm) was collected using thermocouple rakes and axial arrays. The instrumentation was able to capture the radial and axial expelled lubricant-air environment (fling-off lubricant) that is expelled during the gear meshing process. Effects of operational characteristics are presented.

Handschuh, R.↗

DOE OSTI2025

RINO: Renormalization Group Invariance with No Labels

A common challenge with supervised machine learning (ML) in high energy physics (HEP) is the reliance on simulations for labeled data, which can often mismodel the underlying collision or detector response. To help mitigate this problem of domain shift, we propose RINO (Renormalization Group Invariance with No Labels), a self-supervised learning approach that can instead pretrain models directly on collision data, learning embeddings invariant to renormalization group flow scales. In this work, we pretrain a transformer-based model on jets originating from quantum chromodynamic (QCD) interactions from the JetClass dataset, emulating real QCD-dominated experimental data, and then finetune on the JetNet dataset -- emulating simulations -- for the task of identifying jets originating from top quark decays. RINO demonstrates improved generalization from the JetNet training data to JetClass data compared to supervised training on JetNet from scratch, demonstrating the potential for RINO pretraining on real collision data followed by fine-tuning on small, high-quality MC datasets, to improve the robustness of ML models in HEP.

Hao, Zichun [Caltech] (ORCID:0000000256244907)↗

NASA NTRSDate not supplied

Elevating the Quality of Space Omics Sequencing Data: Innovations and Methodologies from NASA GeneLab Sample Processing Laboratory

NASA’s GeneLab, part of the NASA Open Science Data Repository, is a space-related database that hosts a diverse range of transcriptomics, proteomics, epigenomics and genomics data. The NASA GeneLab Sample Processing Laboratory (SPL) generates omics data from biological experiments conducted aboard the International Space Station, Space Shuttle and space related ground experiments, this omics data then hosted on the GeneLab repository. Samples generated such experiments pose numerous technical challenges such as small experimental sample size, variance in dissection times, limited tissue preservation methods, prolonged storage time, and more. GeneLab SPL team had developed specialized expertise in nucleic acid extraction, library preparation and sequencing of such biological samples via extensive training and years of experience. In order to ensure data accuracy and consistency across experiments, SPL has developed standardized protocols for each species and tissue type. These protocols in conjunction with quality control metrics and data standards are crucial in generating of high-quality data. SPL protocols and standards have been developed in collaboration with the scientific community and had been made publicly available on the GeneLab portal, guaranteeing comparability of datasets across spaceflight experiments. To ensure reliability of data generation, SPL leverages cutting-edge innovations in laboratory automation for sample processing. By leveraging these state-of-the-art platforms, SPL achieves high levels of data reproducibility while significantly minimizing sources of bias and variability, especially across experiments with large numbers of samples. Over the past few years, the space biology investigator community has accessed SPL-generated data from the Open Science Data Repository for a myriad of data re-analysis and re-use studies. We observe a trend that in-house SPL-generated data consistently outperforms outsourced sequencing data in terms of technical standards, quality control metrics, timeliness of data delivery, and sequencing and reagent efficiency. Superior data generation has and will continue to enable discoveries in disease, diagnostic tools, and the biological effects of long duration spaceflight.

GeneLab↗

NASA NTRSDate not supplied

Elevating the Quality of Space Omics Sequencing Data: Innovations and Methodologies from NASA GeneLab Sample Processing Laboratory

GeneLab↗

DOE OSTI2020

FAST: Continuing the Focus on Data Quality

This presentation provides an overview of fiscal year 2019 federal motor vehicle fleet data, collected at the individual vehicle level during the fall of 2019, how the the collecting project has reviewed that information for potential quality issues, how the quality of this year's data submission compare to the prior year, and recommendations for federal agencies in their efforts to continue to improve the quality of their submissions. This presentation will be given at the January 2020 FedFleet training event, hosted by the US General Services Administration in Washington, DC. The information is collected through the Federal Automotive Statistical Tool (FAST) project. FAST is a Web-based information system managed by the US Department of Energy, the US General Services Administration, and the Energy Information Administration. FAST is used to collect information about the fleet of motor vehicles used and managed by the Federal government. FAST is developed and maintained by DOE's Idaho National Laboratory (INL).

99 GENERAL AND MISCELLANEOUS↗

NASA NTRS2008

Guidance and Control Software Project Data - Volume 4: Configuration Management and Quality Assurance Documents

The Guidance and Control Software (GCS) project was the last in a series of software reliability studies conducted at Langley Research Center between 1977 and 1994. The technical results of the GCS project were recorded after the experiment was completed. Some of the support documentation produced as part of the experiment, however, is serving an unexpected role far beyond its original project context. Some of the software used as part of the GCS project was developed to conform to the RTCA/DO-178B software standard, "Software Considerations in Airborne Systems and Equipment Certification," used in the civil aviation industry. That standard requires extensive documentation throughout the software development life cycle, including plans, software requirements, design and source code, verification cases and results, and configuration management and quality control data. The project documentation that includes this information is open for public scrutiny without the legal or safety implications associated with comparable data from an avionics manufacturer. This public availability has afforded an opportunity to use the GCS project documents for DO-178B training. This report provides a brief overview of the GCS project, describes the 4-volume set of documents and the role they are playing in training, and includes configuration management and quality assurance documents from the GCS project. Volume 4 contains six appendices: A. Software Accomplishment Summary for the Guidance and Control Software Project; B. Software Configuration Index for the Guidance and Control Software Project; C. Configuration Management Records for the Guidance and Control Software Project; D. Software Quality Assurance Records for the Guidance and Control Software Project; E. Problem Report for the Pluto Implementation of the Guidance and Control Software Project; and F. Support Documentation Change Reports for the Guidance and Control Software Project.

Hayhurst, Kelly J.↗

DOE OSTI2025

Using Boosted Decision Trees to Select High Quality Measurements in the Mu2e Experiment at Fermilab

This thesis presents the implementation and evaluation of a Boosted Decision Tree (BDT) model to improve the selection of high-quality track measurements in the Mu2e experiment at Fermilab. The Mu2e experiment is a high-energy physics experiments seeking to observe a rare theoretical physics process known as Charged Lepton Flavor Violation. A significant challenge faced by the Mu2e experiment are so-called background events, which are events whose data mimics that of the rare physics process the experiment seeks to observe. Without a mechanism to reduce background, it would be impossible to know whether Charged Lepton Flavor Violation occurred or not. To this end, high-quality track measurements must be distinguished from low-quality track measurements. A track can be conceived of as the reconstructed path of a particle that traveled through the Mu2e detector. In addition to other data, data about such tracks is stored using a C++-based framework, specific to the domain of high-energy physics, known as ROOT. A boosted decision tree model was trained using ROOT’s Toolkit For Multivariate Analysis by leveraging variables ancillary to track quality. In evaluation, the BDT achieves a ROC-AUC of 0.927 in discriminating good-quality tracks from poor-quality tracks. Such a score is indicative of both strong discrimination and strong generalization. Subsequently, it is shown that applying a BDT-based quality cut to the distribution of particle momenta significantly enhances the signal-to-background distinction for signal electrons, paving the way for improved sensitivity to Charged Lepton Flavor Violation.

Mullany, Brendan T. [Drew U.] (ORCID:0009000818888↗

DOE OSTI2024

Underground hydrogen storage leakage detection and characterization based on machine learning of sparse seismic data

Underground hydrogen storage (UHS) is considered as a scalable approach for massive storage and seasonal extraction of hydrogen (H 2 ). Although conventional leakage detection and characterization methods based on time-lapse seismic imaging and inversion generally apply to H 2 leakage detection problem, a high-fidelity yet cost effective geophysics approach is still missing to reliably inform leakage location and properties based on very sparse data. In response, we develop a novel supervised machine learning method to detect and characterize H 2 leakage from UHS. The input to our neural network are sparse time-lapse seismic waveforms, while the output from the neural network includes the spatial location and physical properties of a H 2 leakage. Here, we generate high-quality time-lapse waveforms using the elastic-wave equations to train the neural network. We train and validate our machine learning model and find that it attains high accuracy in using extremely sparse time-lapse seismic data to detect and characterize H 2 leakage. Our investigation is the first systematic study that focuses on applying machine learning to subsurface H 2 leakage detection and characterization and could potentially serve as a cost-effective geophysical tool for underground hydrogen leakage detection and characterization with high fidelity.

08 HYDROGEN↗

DOE OSTI2023

Dynamic Monkey Bar Mechanism of Superionic Li–ion Transport in LiTaCl 6

Here, the LiTaCl 6 solid electrolyte has the lowest activation energy of ionic conduction at ambient conditions (0.165 eV), with a record high ionic conductivity for a ternary compound (11 mS cm –1 ). However, the mechanism has been unclear. We train machine-learning force fields (MLFF) on ab initio molecular dynamics (AIMD) data on-the-fly and perform MLFF MD simulations of AIMD quality up to the nanosecond scale at the experimental temperatures, which allows us to predict accurate activation energy for Li-ion diffusion (at 0.164 eV). Detailed analyses of trajectories and vibrational density of states show that the large-amplitude vibrations of Cl – ions in TaCl 6 – enable the fast Li-ion transport by allowing dynamic breaking and reforming of Li–Cl bonds across the space in between the TaCl 6 – octahedra. We term this process the dynamic-monkey-bar mechanism of superionic Li + transport which could aid the development of new solid electrolytes for all-solid-state lithium batteries.

25 ENERGY STORAGE↗