Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Labeling”

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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Deuterated water as a substrate-agnostic isotope tracer for investigating reversibility and thermodynamics of reactions in central carbon metabolism

Stable isotope tracers are a powerful tool for the quantitative analysis of microbial metabolism, enabling pathway elucidation, metabolic flux quantification, and assessment of reaction and pathway thermodynamics. 13 C and 2 H metabolic flux analysis commonly relies on isotopically labeled carbon substrates, such as glucose. However, the use of 2 H-labeled nutrient substrates faces limitations due to their high cost and limited availability in comparison to 13 C-tracers. Furthermore, isotope tracer studies in industrially relevant bacteria that metabolize complex substrates such as cellulose, hemicellulose, or lignocellulosic biomass, are challenging given the difficulty in obtaining these as isotopically labeled substrates. In this study, we examine the potential of deuterated water ( 2 H 2 O) as an affordable, substrate-neutral isotope tracer for studying central carbon metabolism. We apply 2 H 2 O labeling to investigate the reversibility of glycolytic reactions across three industrially relevant bacterial species -C. thermocellum, Z. mobilis, and E. coli-harboring distinct glycolytic pathways with unique thermodynamics. We demonstrate that 2 H 2 O labeling recapitulates previous reversibility and thermodynamic findings obtained with established 13 C and 2 H labeled nutrient substrates. Furthermore, we exemplify the utility of this 2 H 2 O labeling approach by applying it to high-substrate C. thermocellum fermentations -a setting in which the use of conventional tracers is impractical-thereby identifying the glycolytic enzyme phosphofructokinase as a major bottleneck during high-substrate fermentations and unveiling critical insights that will steer future engineering efforts to enhance ethanol production in this cellulolytic organism. This study demonstrates the utility of deuterated water as a substrate-agnostic isotope tracer for examining flux and reversibility of central carbon metabolic reactions, which yields biological insights comparable to those obtained using costly 2 H-labeled nutrient substrates.

09 BIOMASS FUELS↗
DOE OSTI2020

Large fluxes of fatty acids from membranes to triacylglycerol and back during N-deprivation and recovery in Chlamydomonas

Abstract Microalgae accumulate triacylglycerol (TAG) during nutrient deprivation and break it down after nutrient resupply, and these processes involve dramatic shifts in cellular carbon allocation. Due to the importance of algae in the global carbon cycle, and the potential of algal lipids as feedstock for chemical and fuel production, these processes are of both ecophysiological and biotechnological importance. However, the metabolism of TAG is not well understood, particularly the contributions of fatty acids (FAs) from different membrane lipids to TAG accumulation and the fate of TAG FAs during degradation. Here, we used isotopic labeling time course experiments on Chlamydomonas reinhardtii to track FA synthesis and transfer between lipid pools during nitrogen (N)-deprivation and resupply. When cells were labeled before N-deprivation, total levels of label in cellular FAs were unchanged during subsequent N-deprivation and later resupply, despite large fluxes into and out of TAG and membrane lipid pools. Detailed analyses of FA levels and labeling revealed that about one-third of acyl chains accumulating in TAG during N-deprivation derive from preexisting membrane lipids, and in total, at least 45% of TAG FAs passed through membrane lipids at one point. Notably, most acyl chains in membrane lipids during recovery after N-resupply come from TAG. Fluxes of polyunsaturated FAs from plastidic membranes into TAG during N-deprivation were particularly noteworthy. These findings demonstrate a high degree of integration of TAG and membrane lipid metabolism and highlight a role for TAG in storage and supply of membrane lipid components.

Plant Sciences↗
DOE OSTI2021

Deep active learning for classifying cancer pathology reports

Abstract Background Automated text classification has many important applications in the clinical setting; however, obtaining labelled data for training machine learning and deep learning models is often difficult and expensive. Active learning techniques may mitigate this challenge by reducing the amount of labelled data required to effectively train a model. In this study, we analyze the effectiveness of 11 active learning algorithms on classifying subsite and histology from cancer pathology reports using a Convolutional Neural Network as the text classification model. Results We compare the performance of each active learning strategy using two differently sized datasets and two different classification tasks. Our results show that on all tasks and dataset sizes, all active learning strategies except diversity-sampling strategies outperformed random sampling, i.e., no active learning. On our large dataset (15K initial labelled samples, adding 15K additional labelled samples each iteration of active learning), there was no clear winner between the different active learning strategies. On our small dataset (1K initial labelled samples, adding 1K additional labelled samples each iteration of active learning), marginal and ratio uncertainty sampling performed better than all other active learning techniques. We found that compared to random sampling, active learning strongly helps performance on rare classes by focusing on underrepresented classes. Conclusions Active learning can save annotation cost by helping human annotators efficiently and intelligently select which samples to label. Our results show that a dataset constructed using effective active learning techniques requires less than half the amount of labelled data to achieve the same performance as a dataset constructed using random sampling.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Developing an Active Learning algorithm for learning Bayesian classifiers under the Multiple Instance Learning scenario

In the Multiple Instance Learning scenario, the training data consists of instances grouped into bags, and each bag is labelled with whether it is positive, i.e. contains at least one positive instance. First, Active Learning, in which additional labels can be iteratively requested, has the potential to allow more accurate classifiers to be learned with less labels. Active Learning has been applied to the Multiple Instance Learning under two settings: when bag labels of unlabelled bags can be requested, and when instance labels within bags known to be positive can be requested. Second, Bayesian Active learning methods have the potential to learn accurate classifiers with few labels, because they explicitly track the classifier uncertainty and can thus address its knowledge gaps. Yet, there does not exist any Bayesian Active Learning method for the Multiple Instance Learning Scenario. In this work, we develop the first such method. We develop a Bayesian classifier for the Multiple Instance Learning scenario, show how it can be efficiently used for Bayesian Active Learning, and perform experiments assessing its performance. While its performance exceeds that when no Active Learning is used, it is sometimes better, sometimes worse than the naive baseline of uncertainty sampling, depending on the situation. This suggests future work: building more customizable Bayesian Active Learning methods for the Multiple Instance Scenario, customizable to whether bag or instance label accuracy is targeted, and the labeling budget.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

MindSynchro

This report presents the developments and results of MindSynchro project as part of DOE OE FOA 1861. DOE and Pacific Northwest National Laboratory (PNNL) have made available to FOA awardees datasets containing years of real historical data recorded from various phasor measurement units (PMUs) which are installed in three large US interconnections: Texas (IC A), Western (IC B), and Eastern (IC C). The main goal of the project, which was successfully achieved, was to develop methods for detection and identification of events which are relevant for power grid operation. Tasks performed for achieving the project goals included data exploration and pre-processing, the development and application of physics-based features, data analysis and labeling based on unsupervised learning approaches, training and testing of DSSL models for classification of events which are relevant for power grid operation, and deployment of solutions to cloud environments. The methods developed in the project can potentially provide relevant benefits to power grid asset owners/operators in general in terms of situational awareness. Two main types of outcomes can be provided by these tools: Identification of specific relevant power grid event types: Semi-supervised ML methods developed in the project can adequately employ not only the relatively scarce labeled data but also the large amount of available unlabeled data to train models for detection of specific event types. Such methods enable the application of trained models for the detection of events in a population of PMUs much larger than that associated to the labeled events. Support in data labeling / label validation: Labels are critical for training of models for identification of specific types of events. However, labeling large amounts of data is a manual and tedious process. This means that such process is error prone and is not scalable. Methods developed in the project, based on ensembles of clustering models, have been successfully employed for turning manual labeling into a scalable process. Accurate identification of specific relevant events can provide the operators with immediate situational awareness that could otherwise require hours or days of analysis from domain experts. We envision that such methods could be initially employed in support of post-mortem analysis of events and, as confidence is gained, they could be employed for online/real-time support, providing, among other benefits, insights for avoiding major events which could happen due to a combination of smaller ones. On the longer term, related methods could potentially be employed to improve protection and control.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Automated Grain Boundary (GB) Segmentation and Microstructural Analysis in 347H Stainless Steel Using Deep Learning and Multimodal Microscopy

Austenitic 347H stainless steel offers superior mechanical properties and corrosion resistance required for extreme operating conditions such as high temperature. The change in microstructure due to composition and process variations is expected to impact material properties. Identifying microstructural features such as grain boundaries thus becomes an important task in the process-microstructure-properties loop. Applying convolutional neural network (CNN)-based deep learning models is a powerful technique to detect features from material micrographs in an automated manner. In contrast to microstructural classification, supervised CNN models for segmentation tasks require pixel-wise annotation labels. However, manual labeling of the images for the segmentation task poses a major bottleneck for generating training data and labels in a reliable and reproducible way within a reasonable timeframe. Microstructural characterization especially needs to be expedited for faster material discovery by changing alloy compositions. Here, in this study, we attempt to overcome such limitations by utilizing multimodal microscopy to generate labels directly instead of manual labeling. We combine scanning electron microscopy images of 347H stainless steel as training data and electron backscatter diffraction micrographs as pixel-wise labels for grain boundary detection as a semantic segmentation task. The viability of our method is evaluated by considering a set of deep CNN architectures. We demonstrate that despite producing instrumentation drift during data collection between two modes of microscopy, this method performs comparably to similar segmentation tasks that used manual labeling. Additionally, we find that naïve pixel-wise segmentation results in small gaps and missing boundaries in the predicted grain boundary map. By incorporating topological information during model training, the connectivity of the grain boundary network and segmentation performance is improved. Finally, our approach is validated by accurate computation on downstream tasks of predicting the underlying grain morphology distributions which are the ultimate quantities of interest for microstructural characterization.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Soil organic matter is principally root derived in an Ultisol under oak forest

To a large degree, the sources and stability of soil organic carbon remain poorly constrained. A clear understanding of links among the components of the soil C cycle is hampered by the complexity of the system as well as challenges associated with partitioning bulk soil C into meaningful fractions. A large accidental 14 CO 2 release at the Oak Ridge Reservation in Tennessee, USA provided a strong label pulse into adjacent, well-studied oak forests, resulting in highly elevated Δ 14 C values in leaf litter (~1000‰) and roots (~260–450‰). A four-year manipulative study was conducted to determine the relative contribution of litter versus roots to the bulk mineral soil C pool, as well as to free light, occluded light and heavy fractions. The heavy fraction was further split into fractions with densities of 1.7–2.4 g cm –3 and >2.4 g cm –3 to test the homogeneity of the mineral-associated fraction of C. Substantial concentrations of label were detected in all soil fractions within a year of the 14 CO 2 release, indicating rapid incorporation of newly fixed photosynthates in all fractions of soil organic C, regardless of differences in stability inferred by previous work. This rapid incorporation of label occurred only in treatments where roots were labeled, indicating that roots are the major source of inputs to mineral soil C stocks at these sites. Separation of the heavy fraction into subfractions of intermediate (1.7–2.4 g cm –3 ) and high (>2.4 g cm –3 ) density indicated that both subfractions incorporated label at similar rates, despite significant differences in degree of microbial processing. In general, the rate of label incorporation suggested a much faster turnover for all fractions than indicated by natural radiocarbon abundance values. This study suggests that within each soil fraction there are portions of slow-cycling and fast-cycling materials, and the determination of an average turnover time or mean age is dependent on experimental approach. The rapid incorporation of label into all fractions within a year regardless of inferred stability implies a high degree of heterogeneity in all fractions regardless of how finely the soils are partitioned. Further refinement of the nature and drivers of this heterogeneity could yield important insights into the soil C cycle.

58 GEOSCIENCES↗
DOE OSTI2020

Source of 12 C in Calvin–Benson cycle intermediates and isoprene emitted from plant leaves fed with 13 CO 2

Feeding 14 CO 2 was crucial to uncovering the path of carbon in photosynthesis. Feeding 13 CO 2 to photosynthesizing leaves emitting isoprene has been used to develop hypotheses about the sources of carbon for the methylerythritol 4-phosphate pathway, which makes the precursors for terpene synthesis in chloroplasts and bacteria. Both photosynthesis and isoprene studies found that products label very quickly (<10 min) up to 80–90% but the last 10–20% of labeling requires hours indicating a source of 12 C during photosynthesis and isoprene emission. Furthermore, studies with isoprene showed that the proportion of slow label could vary significantly. This was interpreted as a variable contribution of carbon from sources other than the Calvin–Benson cycle (CBC) feeding the methylerythritol 4-phosphate pathway. Here, we measured the degree of label in isoprene and photosynthetic metabolites 20 min after beginning to feed 13 CO 2 . Isoprene labeling was the same as labeling of photosynthesis intermediates. High temperature reduced the label in isoprene and photosynthesis intermediates by the same amount indicating no role for alternative carbon sources for isoprene. A model assuming glucose, fructose, and/or sucrose reenters the CBC as ribulose 5-phosphate through a cytosolic shunt involving glucose 6-phosphate dehydrogenase was consistent with the observations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Automated annotation of scientific texts for ML-based keyphrase extraction and validation

Advanced omics technologies and facilities generate a wealth of valuable data daily; however, the data often lack the essential metadata required for researchers to find, curate, and search them effectively. The lack of metadata poses a significant challenge in the utilization of these data sets. Machine learning (ML)–based metadata extraction techniques have emerged as a potentially viable approach to automatically annotating scientific data sets with the metadata necessary for enabling effective search. Text labeling, usually performed manually, plays a crucial role in validating machine-extracted metadata. However, manual labeling is time-consuming and not always feasible; thus, there is a need to develop automated text labeling techniques in order to accelerate the process of scientific innovation. This need is particularly urgent in fields such as environmental genomics and microbiome science, which have historically received less attention in terms of metadata curation and creation of gold-standard text mining data sets. In this paper, we present two novel automated text labeling approaches for the validation of ML-generated metadata for unlabeled texts, with specific applications in environmental genomics. Our techniques show the potential of two new ways to leverage existing information that is only available for select documents within a corpus to validate ML models, which can then be used to describe the remaining documents in the corpus. The first technique exploits relationships between different types of data sources related to the same research study, such as publications and proposals. The second technique takes advantage of domain-specific controlled vocabularies or ontologies. In this paper, we detail applying these approaches in the context of environmental genomics research for ML-generated metadata validation. Our results show that the proposed label assignment approaches can generate both generic and highly specific text labels for the unlabeled texts, with up to 44% of the labels matching with those suggested by a ML keyword extraction algorithm.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2026

Advanced Semi-Supervised Learning with Uncertainty Estimation for Phase Identification in Distribution Systems

The integration of advanced metering infrastructure (AMI) into power distribution networks generates valuable data for tasks such as phase identification; however, the limited and unreliable availability of labeled data in the form of customer phase connectivity presents challenges. To address this issue, we propose a semi-supervised learning (SSL) framework that effectively leverages labeled and unlabeled data. Our approach incorporates self-training, label spreading, and Bayesian neural networks (BNNs) to enhance phase identification with AMI data. Our method uses an ensemble of multilayer perceptron classifiers in a self-training setup, iteratively adding high-confidence pseudo-labels to improve robustness. We also apply label spread to propagate labels based on data similarity, which enhances generalization across diverse distributions. In addition, we employ a BNNs with uncertainty estimation, boosting confidence in predictions and reducing phase identification errors. In our case study, we achieved approximately 98% +/- 0.08 accuracy with uncertainty using minimal and unreliable labeled data from a real U.S. utility, Duquesne Light Company. Our SSL approach, combined with uncertainty estimation, provides an efficient solution for phase identification in AMI data, ultimately improving the reliability of smart grid applications.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

Advanced Semi-Supervised Learning With Uncertainty Estimation for Phase Identification in Distribution Systems

The integration of advanced metering infrastructure (AMI) into power distribution networks generates valuable data for tasks such as phase identification; however, the limited and unreliable availability of labeled data in the form of customer phase connectivity presents challenges. To address this issue, we propose a semi-supervised learning (SSL) framework that effectively leverages labeled and unlabeled data. Our approach incorporates self-training, label spreading, and Bayesian neural networks (BNNs) to enhance phase identification with AMI data. Our method uses an ensemble of multilayer perceptron classifiers in a self-training setup, iteratively adding high-confidence pseudo-labels to improve robustness. We also apply label spread to propagate labels based on data similarity, which enhances generalization across diverse distributions. In addition, we employ a BNNs with uncertainty estimation, boosting confidence in predictions and reducing phase identification errors. In our case study, we achieved approximately 98% +/- 0.08 accuracy with uncertainty using minimal and unreliable labeled data from a real U.S. utility, Duquesne Light Company. Our SSL approach, combined with uncertainty estimation, provides an efficient solution for phase identification in AMI data, ultimately improving the reliability of smart grid applications.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2020

Automated Identification of Molecular Crystals’ Packing Motifs

Packing motifs—patterns in how molecules orient relative to one another in a crystal structure—are an important concept in many subdisciplines of materials science because of correlations observed between specific packing motifs and properties of interest. That said, packing motif data sets have remained small and noisy due to intensive manual labeling processes and insufficient labeling schemes. The most prominent labeling algorithms calculate relative interplanar angles of nearest neighbor molecules to determine the packing motif of a molecular crystal, but this simple approach can fail when neighbors are naively sampled isotropically around the crystal structure. To remedy this issue, here we propose an optimization algorithm, which rotates the molecular crystal structure to find representative molecules that inform the packing motif. We package this algorithm into an automated framework—Autopack—which both optimally rotates the crystal structure and labels the packing motif based on the appropriate neighboring molecules. In this work, we detail the Autopack framework and its performance, which shows improvements compared to previous state-of-the-art labeling methods, providing the first quantitative point of comparison for packing motif labeling algorithms. Furthermore, using Autopack (available at https://ipo.llnl.gov/technologies/software/autopack), we perform the first large-scale study of potential relationships between chemicals’ compositions and packing motifs, which shows that these relationships are more complex than previously hypothesized from studies that used only tens of polycyclic aromatic hydrocarbon molecules. Autopack’s capabilities help pose next steps for crystal engineering research focusing not only on a molecule’s adoption of a specific packing motif but also on new structure–property relationships.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Isotope ratio-based quantification of carbon assimilation highlights the role of plastidial isoprenoid precursor availability in photosynthesis

Abstract Background We report a method to estimate carbon assimilation based on isotope ratio-mass spectrometry (IRMS) of 13 CO 2 labeled plant tissue. Photosynthetic carbon assimilation is the principal experimental observable which integrates important aspects of primary plant metabolism. It is traditionally measured through gas exchange. Despite its centrality in plant research, gas exchange performs poorly with rosette growth habits typical of Arabidopsis thaliana , mutant lines with limited biomass, and accounts poorly for leaf shading. Results IRMS-based carbon assimilation values from plants labeled at different light intensities were compared to those obtained by gas exchange, and the two methods yielded similar values. Using this method, we observed a strong correlation between 13 C content and labeling time (R 2 = 0.999) for 158 wild-type plants labeled for 6 to 42 min. Plants cultivated under different light regimes showed a linear response with respect to carbon assimilation, varying from 7.38 nmol 13 C mg −1 leaf tissue min −1 at 80 PAR to 19.27 nmol 13 C mg −1 leaf tissue min −1 at 500 PAR. We applied this method to examine the link between inhibition of the 2 C -methyl- d -erythritol-4-phosphate (MEP) pathway and suppression of photosynthesis. A significant decrease in carbon assimilation was observed when metabolic activity in the MEP pathway was compromised by mutation or herbicides targeting the MEP pathway. Mutants affected in MEP pathway genes 1-DEOXY- d -XYLULOSE 5-PHOSPHATE SYNTHASE (DXS) or 1-HYDROXY-2-METHYL-2-(E)-BUTENYL 4-DIPHOSPHATE SYNTHASE (HDS) showed assimilation rates 36% and 61% lower than wild type. Similarly, wild type plants treated with the MEP pathway inhibitors clomazone or fosmidomycin showed reductions of 52% and 43%, respectively, while inhibition of the analogous mevalonic acid pathway, which supplies the same isoprenoid intermediates in the cytosol, did not, suggesting inhibition of photosynthesis was specific to disruption of the MEP pathway. Conclusions This method provides an alternative to gas exchange that offers several advantages: resilience to differences in leaf overlap, measurements based on tissue mass rather than leaf surface area, and compatibility with mutant Arabidopsis lines which are not amenable to gas exchange measurements due to low biomass and limited leaf surface area. It is suitable for screening large numbers of replicates simultaneously as well as post-hoc analysis of previously labeled plant tissue and is complementary to downstream detection of isotopic label in targeted metabolite pools.

2C‑methyl‑d‑erythritol 4‑phosphate pathway↗
DOE OSTI2022

A data-centric weak supervised learning for highway traffic incident detection

Using the data from loop detector sensors for near-real-time detection of traffic incidents on highways is crucial to averting major traffic congestion. While recent supervised machine learning methods offer solutions to incident detection by leveraging human-labeled incident data, the false alarm rate is often too high to be used in practice. Specifically, the inconsistency in the human labeling of the incidents significantly affects the performance of supervised learning models. To that end, we focus on a data-centric approach to improve the accuracy and reduce the false alarm rate of traffic incident detection on highways. We develop a weak supervised learning workflow to generate high-quality training labels for the incident data without the ground truth labels, and we use those generated labels in the supervised learning setup for final detection. This approach comprises three stages. First, we introduce a data preprocessing and curation pipeline that processes traffic sensor data to generate high-quality training data through leveraging labeling functions, which can be domain knowledge-related or simple heuristic rules. Second, we evaluate the training data generated by weak supervision using three supervised learning models-random forest, k-nearest neighbors, and a support vector machine ensemble-and long short-term memory classifiers. The results show that the accuracy of all of the models improves significantly after using the training data generated by weak supervision. Third, we develop an online real-time incident detection approach that leverages the model ensemble and the uncertainty quantification while detecting incidents. Finally, we show that our proposed weak supervised learning workflow achieves a high incident detection rate (0.90) and low false alarm rate (0.08).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Volumetric imaging of the 3D orientation of cellular structures with a polarized fluorescence light-sheet microscope

Polarized fluorescence microscopy is a valuable tool for measuring molecular orientations in biological samples, but techniques for recovering three-dimensional orientations and positions of fluorescent ensembles are limited. We report a polarized dual-view light-sheet system for determining the diffraction-limited three-dimensional distribution of the orientations and positions of ensembles of fluorescent dipoles that label biological structures. We share a set of visualization, histogram, and profiling tools for interpreting these positions and orientations. We model the distributions based on the polarization-dependent efficiency of excitation and detection of emitted fluorescence, using coarse-grained representations we call orientation distribution functions (ODFs). We apply ODFs to create physics-informed models of image formation with spatio-angular point-spread and transfer functions. We use theory and experiment to conclude that light-sheet tilting is a necessary part of our design for recovering all three-dimensional orientations. We use our system to extend known two-dimensional results to three dimensions in FM1-43-labeled giant unilamellar vesicles, fast-scarlet-labeled cellulose in xylem cells, and phalloidin-labeled actin in U2OS cells. Additionally, we observe phalloidin-labeled actin in mouse fibroblasts grown on grids of labeled nanowires and identify correlations between local actin alignment and global cell-scale orientation, indicating cellular coordination across length scales.

Science & Technology - Other Topics↗
DOE OSTI2023

Integrated deep learning framework for unstable event identification and disruption prediction of tokamak plasmas

Abstract The ability to identify underlying disruption precursors is key to disruption avoidance. In this paper, we present an integrated deep learning (DL) based model that combines disruption prediction with the identification of several disruption precursors like rotating modes, locked modes, H-to-L back transitions and radiative collapses. The first part of our study demonstrates that the DL-based unstable event identifier trained on 160 manually labeled DIII-D shots can achieve, on average, 84% event identification rate of various frequent unstable events (like H-L back transition, locked mode, radiative collapse, rotating MHD mode, large sawtooth crash), and the trained identifier can be adapted to label unseen discharges, thus expanding the original manually labeled database. Based on these results, the integrated DL-based framework is developed using a combined database of manually labeled and automatically labeled DIII-D data, and it shows state-of-the-art (AUC = 0.940) disruption prediction and event identification abilities on DIII-D. Through cross-machine numerical disruption prediction studies using this new integrated model and leveraging the C-Mod, DIII-D, and EAST disruption warning databases, we demonstrate the improved cross-machine disruption prediction ability and extended warning time of the new model compared with a baseline predictor. In addition, the trained integrated model shows qualitatively good cross-machine event identification ability. Given a labeled dataset, the strategy presented in this paper, i.e. one that combines a disruption predictor with an event identifier module, can be applied to upgrade any neural network based disruption predictor. The results presented here inform possible development strategies of machine learning based disruption avoidance algorithms for future tokamaks and highlight the importance of building comprehensive databases with unstable event information on current machines.

plasma instabilities↗
DOE OSTI2024

Boundary-Aware Adversarial Learning Domain Adaption and Active Learning for Cross-Sensor Building Extraction

The use of convolutional neural networks (CNNs) for building extraction from remote sensing images has been widely studied and many public datasets have been made available for accelerating development of these CNN models. Yet adapting pretrained models at scale in real-world scenarios remains a challenging task. The main barrier is that certain new labels are still needed to compensate for domain shifting between the labeled data and new images that potentially cover new geographic locations or that are from a different sensor. In this article, we propose to add informatively labeled samples from a new image pool under the paradigm of active learning. To select the most useful samples based on model uncertainty, we first tackle the problem of uncalibrated uncertainty estimation due to distribution shifting by adapting feature extractors with boundary-based adversarial learning. Calibrated uncertainty is used as the query criterion in the active learning process, where the most uncertain samples are selected for annotation and included for model retraining. The proposed workflow was tested with three data pairs in which each workflow represents a scenario often encountered in real-world applications, including adapting pretrained models to new images collected with different sensors or to new geographic areas where appearances and types of buildings are very different. Compared to several baselines, including random sampling, temperature scaling (a well-known uncertainty calibration technique), different query strategies, and active domain adaptation methods, the proposed workflow shows that strategically querying a smaller set of samples for labeling achieves comparable or better building extraction performance. The proposed method reduces the number of labeled samples required to achieve sufficient model accuracy, thus significantly reducing hundreds of person-hours for labeled data creation. In addition, we include a few considerations when deploying this workflow in a GPU cluster that can be easily adapted to achieve operational building extraction model retraining.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Automated Shift Detection in Sensor-Based PV Power and Irradiance Time Series: Preprint

PV power and irradiance sensor-based measurements are prone to error, resulting in issues such as abrupt time series data shifts. These shifts, which are usually unintentional, may be caused by software or hardware configuration changes on a PV system, and do not reflect an actual change in overall system performance. Locating these shifts and segmenting the associated time series aids in more accurate future PV analysis. In this research, an offline changepoint detection (CPD) algorithm that automatically detects these abrupt data shifts in sensor-based time series is introduced. Data shift periods in 101 daily PV power and irradiance time series were labeled manually by two solar experts. These data streams represent sensor-based measurements, and display a variety of data shift behaviors. A changepoint detection algorithm was tuned using the 101 labeled data streams, with each model configuration's ability to detect labeled changepoints benchmarked using metrics such as F1-score, recall, and Rand Index. Best performing models on seasonality-corrected data streams include the Pruned Exact Linear (PELT) method, the Binary Segmentation method, and the Bottom-Up method, all scoring an average F1-score of 0.76 or greater at detecting labeled changepoints within a 30-day window for the labeled data sets. To promote further research in this space, we are releasing the labeled data shift sets on U.S. Department of Energy's (DOE) DuraMAT Data Hub, and the associated algorithm in the Python PVAnalytics package.

changepoint detection↗