Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Anomaly Detection”

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 919 records · Page 51

Text and Structural Data Mining of Influenza Mentions in Web and Social Media

Text and structural data mining of Web and social media (WSM) provides a novel disease surveillance resource and can identify online communities for targeted public health communications (PHC) to assure wide dissemination of pertinent information. WSM that mention influenza are harvested over a 24-week period, 5-October-2008 to 21-March-2009. Link analysis reveals communities for targeted PHC. Text mining is shown to identify trends in flu posts that correlate to real-world influenza-like-illness patient report data. We also bring to bear a graph-based data mining technique to detect anomalies among flu blogs connected by publisher type, links, and user-tags.

60 APPLIED LIFE SCIENCES↗

Adaptive Machine Learning for Robust Diagnostics and Control of Time-Varying Particle Accelerator Components and Beams

Machine learning (ML) is growing in popularity for various particle accelerator applications including anomaly detection such as faulty beam position monitor or RF fault identification, for non-invasive diagnostics, and for creating surrogate models. ML methods such as neural networks (NN) are useful because they can learn input-output relationships in large complex systems based on large data sets. Once they are trained, methods such as NNs give instant predictions of complex phenomenon, which makes their use as surrogate models especially appealing for speeding up large parameter space searches which otherwise require computationally expensive simulations. However, quickly time varying systems are challenging for ML-based approaches because the actual system dynamics quickly drifts away from the description provided by any fixed data set, degrading the predictive power of any ML method, and limits their applicability for real time feedback control of quickly time-varying accelerator components and beams. In contrast to ML methods, adaptive model-independent feedback algorithms are by design robust to un-modeled changes and disturbances in dynamic systems, but are usually local in nature and susceptible to local extrema. In this work, we propose that the combination of adaptive feedback and machine learning, adaptive machine learning (AML), is a way to combine the global feature learning power of ML methods such as deep neural networks with the robustness of model-independent control. We present an overview of several ML and adaptive control methods, their strengths and limitations, and an overview of AML approaches.

97 MATHEMATICS AND COMPUTING↗

Machine Learning Applications to Maintain the NuMI Neutrino Beam Quality at Fermilab

The NuMI target facility at Fermilab produces an intense muon neutrino beam for NOvA (NuMI Off-axis νe Appearance) long baseline neutrino experiment. Three arrays of muon monitors located in the downstream of the hadron absorber in the NuMI beamline provide the measurements of the primary beam and horn current quality. We have studied the response of muon monitors with the proton beam profile changes and focusing horn current variations. The responses of muon monitors are used to develop Machine Learning (ML) algorithms to monitor the beam quality. We present the development of the machine learning applications and the future plans. This effort is important for future applications such as beam quality assurance, anomaly detections, neutrino beam systematics studies and neutrino beam quality assurance. Our results demonstrate the advantages of developing useful ML applications that can be leveraged for future beamlines such as LBNF.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

The Young Supernova Experiment Data Release 1 (YSE DR1): Light Curves and Photometric Classification of 1975 Supernovae

We present the Young Supernova Experiment Data Release 1 (YSE DR1), comprised of processed multicolor PanSTARRS1 griz and Zwicky Transient Facility (ZTF) gr photometry of 1975 transients with host–galaxy associations, redshifts, spectroscopic and/or photometric classifications, and additional data products from 2019 November 24 to 2021 December 20. YSE DR1 spans discoveries and observations from young and fast-rising supernovae (SNe) to transients that persist for over a year, with a redshift distribution reaching z ≈ 0.5. We present relative SN rates from YSE's magnitude- and volume-limited surveys, which are consistent with previously published values within estimated uncertainties for untargeted surveys. We combine YSE and ZTF data, and create multisurvey SN simulations to train the ParSNIP and SuperRAENN photometric classification algorithms; when validating our ParSNIP classifier on 472 spectroscopically classified YSE DR1 SNe, we achieve 82% accuracy across three SN classes (SNe Ia, II, Ib/Ic) and 90% accuracy across two SN classes (SNe Ia, core-collapse SNe). Our classifier performs particularly well on SNe Ia, with high (>90%) individual completeness and purity, which will help build an anchor photometric SNe Ia sample for cosmology. We then use our photometric classifier to characterize our photometric sample of 1483 SNe, labeling 1048 (~71%) SNe Ia, 339 (~23%) SNe II, and 96 (~6%) SNe Ib/Ic. YSE DR1 provides a training ground for building discovery, anomaly detection, and classification algorithms, performing cosmological analyses, understanding the nature of red and rare transients, exploring tidal disruption events and nuclear variability, and preparing for the forthcoming Vera C. Rubin Observatory Legacy Survey of Space and Time.

79 ASTRONOMY AND ASTROPHYSICS↗

Unsupervised Machine Learning for Exploratory Data Analysis of Exoplanet Transmission Spectra

Abstract Transit spectroscopy is a powerful tool for decoding the chemical compositions of the atmospheres of extrasolar planets. In this paper, we focus on unsupervised techniques for analyzing spectral data from transiting exoplanets. After cleaning and validating the data, we demonstrate methods for: (i) initial exploratory data analysis, based on summary statistics (estimates of location and variability); (ii) exploring and quantifying the existing correlations in the data; (iii) preprocessing and linearly transforming the data to its principal components; (iv) dimensionality reduction and manifold learning; (v) clustering and anomaly detection; and (vi) visualization and interpretation of the data. To illustrate the proposed unsupervised methodology, we use a well-known public benchmark data set of synthetic transit spectra. We show that there is a high degree of correlation in the spectral data, which calls for appropriate low-dimensional representations. We explore a number of different techniques for such dimensionality reduction and identify several suitable options in terms of summary statistics, principal components, etc. We uncover interesting structures in the principal component basis, namely well-defined branches corresponding to different chemical regimes of the underlying atmospheres. We demonstrate that those branches can be successfully recovered with a K-means clustering algorithm in a fully unsupervised fashion. We advocate for lower-dimensional representations of the spectroscopic data in terms of the main principal components, in order to reveal the existing structure in the data and quickly characterize the chemical class of a planet.

Matchev, Konstantin T. (ORCID:0000000341829096)↗

RODEM Jet Datasets

We present the RODEM Jet Datasets, a comprehensive collection of simulated large-radius jets designed to support the development and evaluation of machine-learning algorithms in particle physics. These datasets encompass a diverse range of jet sources, including quark/gluon jets, jets from the decay of W bosons, top quarks, and heavy new-physics particles. The datasets provide detailed substructure information, including jet kinematics, constituent kinematics, and track displacement details, enabling a wide range of applications in jet tagging, anomaly detection, and generative modelling.

FOS: Physical sciences↗

LandScan Global 2023: Silver Edition

For a quarter of a century, the LandScan Global (LSG) project has annually released a global, high-resolution gridded population dataset representing the ambient or unwarned population at a 30 arcsecond resolution. LSG supports a range of applications such as emergency management, disaster response, and human health and security for understanding populations at risk. The 2023 release of LSG, the LandScan Silver Edition, represents a major methodological leap forward while also leveraging previous knowledge—the previous year was the baseline for the current annual update carrying forward valuable knowledge of the built environment for the past quarter century—to train the machine learning models. Compared with annual releases over the past 24years, multiple advancements were made to different aspects of the methodology to achieve reproducibility, transparency, and consistent global propagation of solutions to modeling or population distribution issues identified during the review process. These novel changes include incorporation of the latest available geospatial inputs across the globe, machine learning models instead of manual modifications, population feature importance analysis, open-source solutions vs. proprietary software, generation of multiple global versions, analytic validations, and human-in-the-loop revisions to produce the final version. Additionally, algorithms—such as anomaly detection—were introduced to quickly identify areas of focus to develop a new and robust systematic review. Significant changes in modeled population distributions were observed between the 2022 and 2023 releases, largely attributable to improvements in data and methods and discussed thoroughly within this report. In summation, the LandScan Silver Edition leverages the best of the past quarter century of LSG legacy knowledge and continues a tradition of applying cutting-edge enhancements to serve as a new benchmark for accurate, actionable gridded population data

Lebakula, Viswadeep↗

NBIC Biofeeds: Deploying a New, Digital Tool for Open Source Biosurveillance across Federal Agencies

The National Biosurveillance Integration Center (NBIC) is deploying a scalable, flexible open source data collection, analysis, and dissemination tool to support biosurveillance operations by the U.S. Department of Homeland Security (DHS) and its federal interagency partners.IntroductionNBIC integrates, analyzes, and distributes key information about health and disease events to help ensure the nation’s responses are well-informed, save lives, and minimize economic impact. To meet its mission objectives, NBIC utilizes a variety of data sets, including open source information, to provide comprehensive coverage of biological events occurring across the globe. NBIC Biofeeds is a digital tool designed to improve the efficiency of analyzing large volumes of open source reporting and increase the number of relevant insights gleaned from this dataset. Moreover, the tool provides a mechanism to disseminate tailored, electronic message notifications in near-real time so that NBIC can share specific information of interest to its interagency partners in a timely manner. NBIC intends to implement operational use of the capability in FY 2018. The core components of the system are data collection, curation, and dissemination of information deemed important by NBIC subject matter experts. NBIC Biofeeds has captured information from more than 70,000 unique sources published from around the globe and presents, on average, 9,000 new biosurveillance-relevant articles to users each day. NBIC leverages a variety of data feeds, including third party aggregators like Google and subscription-based feeds such as HealthMap, as well as Really Simple Syndication (RSS) feeds and web-scraping of highly relevant sources. The NBIC biosurveillance taxonomy imbedded in the tool consists of more than 600 metadata targets that cover key information for understanding the significance of an active biological event, including etiologic agents, impact to humans and animals (e.g., infection severity, healthcare workers involved, type of host), social disruption, infrastructure strain, countermeasures engaged, and ‘red flag’ characteristics (e.g., pathogen appearance in a new geographic area, unusual clinical signs). This taxonomy serves as a foundation for data curation and can be tailored by NBIC partners to more directly meet their own mission objectives. At this time, metadata is predominately captured by NBIC analysts, who manually tag information, which triggers the population of three automatically-disseminated products from the tool: 1) the NBIC Daily Biosurveillance Review, 2) immediate and daily summary email notifications, and 3) custom-designed RSS feeds. These products are meant for individual recipients in the federal interagency and for consumption by other biosurveillance information technology systems, such as the Department of Defense, Defense Threat Reduction Agency (DTRA) Biosurveillance Ecosystem (BSVE). NBIC is working in partnership with DTRA to integrate NBIC Biofeeds as an application directly into the BSVE and further develop the BSVE as an all-in-one platform for biosurveillance data analytics. To improve the efficiency and effectiveness of gaining insights using NBIC Biofeeds, developers of the tool at the Pacific Northwest National Laboratory (PNNL) are researching and testing a variety of advanced analytics techniques focused on: 1) article relevancy ratings to improve the review of queried data, 2) significance ratings to elucidate the perceived severity of an event based on reported characteristics, 3) full-text article retrieval and storage for improved machine-tagging, and 4) anomaly detection for emerging threats. Testing and implementation of new analytic capabilities in NBIC Biofeeds is planned for this fiscal year. NBIC Biofeeds was developed to serve as a sophisticated and powerful open source biosurveillance technology of value to the federal government by providing information to stakeholders conducting open source biosurveillance as well as those consuming biosurveillance information. In FY 2018, NBIC Biofeeds will begin operational use by NBIC and an initial set of users in various federal agencies. User accounts for testing purposes will be available to other federal partners, and a broad scope of federal stakeholders can receive products directly from NBIC Biofeeds based on their interests. NBIC Biofeeds is expected to enable more rapid recognition and enhanced analysis of emerging biological events by NBIC analysts. NBIC anticipates other federal agencies with biosurveillance missions will find this technology of value and intends to offer use of the platform to those federal partners that can benefit from access to the tool and information generated from NBIC Biofeeds.

NBIC↗

Facilitating the Use of Epidemiological Models for Infectious Disease Surveillance

1. To develop a comprehensive model characterization framework to describe epidemiological models in an operational context. 2. To apply the framework to characterize “operational” models for specific infectious diseases and provide a web-based directory, the biosurveillance analytics resource directory (BARD) to the global infectious disease surveillance community. Epidemiological modeling for infectious disease is useful for disease management and routine implementation needs to be facilitated through better description of models in an operational context. A standardized model characterization process that allows selection or making manual comparisons of available models and their results is currently lacking. Los Alamos National Laboratory(LANL) has developed a comprehensive framework that can be used to characterize an infectious disease model in an operational context. We offer this framework and an associated database to stakeholders of the infectious disease modeling field as a tool for standardizing model description and facilitating the use of epidemiological models. Such a framework could help the understanding of diverse models by various stakeholders with different preconceptions, backgrounds, expertise ,and needs, and can foster greater use of epidemiological models as tools in infectious disease surveillance. We define, “operational” as the application of an epidemiological model to a real-world event for decision support and can be used by experts and non-experts alike. The term “model” covers three major types, risk mapping, disease dynamics and anomaly detection. To develop a framework for characterizing epidemiological models we collected information via a three-step process: a literature search of model characteristics, a review of current operational infectious disease epidemiological models, and subject matter expert (SME)panel consultation. We limited selection of operational models to five infectious diseases: influenza, malaria, dengue, cholera and foot-and-mouth disease (FMD). These diseases capture a variety of transmission modes, represent high or potentially high epidemic or endemic burden, and are well represented in the literature. We also developed working criteria for what attributes can be used to comprehensively describe an operational model including a model’s documentation, accessibility, and sustainability. To apply the model characterization framework, we built the BARD, which is publicly available (http://brd.bsvgateway.org).A document was also developed to describe the usability requirements for the BARD; potential users (and non-users) and use cases are formally described to explain the scope of use. 1. Framework for model characterization. The framework is divided into six major components (Figure 1):Model Purpose, Model Objective, Model Scope, Biosurveillance(BSV) goals, Conceptual Model and Model Utility; each of which has several sub-categories for characterizing each aspect of a model.2. Application to model characterization Models for five infectious diseases—cholera, malaria, influenza, FMD and dengue were characterized using the framework and are included in the BARD database. Our framework characterized disparate models in a streamlined fashion. Model information could be binned into the same categories, allowing easy manual comparison and understanding of the models.3. Development of the BARD. Our model characterization framework was implemented into an actionable tool which provides specific information about a model that has been systematically categorized. It allows manual category-to-category comparison of multiple models for a single disease and while the tool does not rank models it provides model information in a format that allows a user to make a ranking or an assessment of the utility of the model. With the model characterization framework we hope to encourage model developers to start describing the many features of their models using a common format. We illustrate the application of the framework through the development of the BARD which is a scientific and non-biased tool for selecting an appropriate epidemiological model for infectious disease surveillance. Epidemiological models are not necessarily being developed with decision makers in mind. This gap between model developers and decision makers needs to be narrowed before modeling becomes routinely implemented in decision making. The characterization framework and the tool developed (BARD) area first step towards addressing this gap.

97 MATHEMATICS AND COMPUTING↗

LCLS RF Station Phase Anomaly Candidate Dataset

A public anomaly detection dataset constructed from RF station faults for phase at SLAC's LCLS (Linac Coherent Light Source). We have compiled a dataset of the RF station diagnostic phase data and the beam-position monitor (BPM) signals, alongside the hand labels, for a labeled study period. The dataset consists of two HDF5 files (one for train and one for test) containing the raw data, two CSV files containing information about the candidates. The CSV file for the test dataset also contains the label.

Liang, Jia [Stanford Univ., CA (United States). In↗

Evaluation of ERTS-1 imagery for mapping Quaternary deposits and landforms in the Great Plains and Midwest

The author has identified the following significant results. The main landform associations and larger landforms are readily identifiable on the better images and commonly the gross associations of surficial Quaternary deposits also can be differentiated, primarily by information on landforms and soils. Maps showing the Quaternary geologic-terrain units that can be differentiated from the ERTS-1 images are being prepared for 20-odd potential study areas in Illinois, Iowa, Missouri, Kansas, Nebraska, and South Dakota. Among the more distinct features are the major moraines and outwash channels of the last (Wisconsin) glaciation. Analysis of dissection/drainage patterns from the synoptic imagery is proving useful for detecting anomalies that may be caused by stream diversions and moraines of pre-Wisconsin glaciations, by variable loess deposition, by tectonism, and other factors. Numerous abandoned river valleys have been mapped. Trend-lines of several known pre-Wisconsin moraine systems have been identified in Iowa, Nebraska, and Kansas, and also several similar trend-lines, that may indicate previously unknown moraine systems of middle and possibly early Pleistocene age, have been found in Iowa and Missouri. The area inundated by a major flood in southwestern Iowa also has been delineated from ERTS-1 imagery.

Morrison, R. B.↗

Lunar physical properties from analysis of magnetometer data

The electromagnetic properties of the lunar interior are discussed with emphasis on (1) bulk, crustal, and local anomalous conductivity; (2) bulk magnetic permeability measurements, iron abundance estimates, and core size limits; (3) lunar ionosphere and atmosphere; and (4) crustal magnetic remanence: scale size measurements and constraints on remanence origin. Appendices treat the phase relationship between the energetic particle flux modulation and current disc penetrations in the Jovian magnetosphere (Pioneer 10 inbound) theories for the origin of lunar magnetism; electrical conductivity anomalies associated with circular lunar maria; electromagnetic properties of the Moon; Mare Serenitatis conductivity anomaly detected by Apollo 16 and Lunokhod 2 magnetometers; and lunar properties from magnetometer data: effects of data errors.

Daily, W. D.↗

Supernovae and the origin of the solar system

This review concentrates on recent ideas involving a relationship between the early solar system and supernova explosions. It summarizes briefly the data that has helped inspire those ideas. Because the true relationship is still unknown and generates controversy, the distinct ideas are introduced singly in the historical context of their origins, and the active sense of surprise and controversy is visible. Quotations from pivotal papers are used as part of the exposition. The subject involves equally the isotopic anomalies detected in meteorites and the dynamic events of galactic evolution, nucleosynthesis, and protosolar collapse. Whatever the correct situation is, new connections have been found between the origin of the elements and the formation of the solar system. The objective of this review is to enable interested space scientists to quickly identify the competing points of view and the experiments and theories that have led to them.

Clayton, D. D.↗

Synthesis of Al-26 in explosive hydrogen burning

The possibility of Al-26 synthesis during the explosive processing of hydrogen-rich material in the outer layers of a supernova or in nova envelopes is investigated. It is found that in the peak temperature range of 1-3 x 10 to the 8th deg K and for expansion time scales of the order of 1-1000 s, values of (Al-26)/(Al-27) as high as 0.1-1 can be obtained for values of 0.001-100,000 for the product of the peak density in g/cu cm and the proton mass fraction. Such a level of Al-26 production is considerably higher than that of recent carbon/neon burning nucleosynthesis models, and is sufficient to account for the magnesium anomalies detected in certain inclusions from the Leoville and Allende meteorites. Al abundances resulting from a hydrodynamical calculation performed on the grounds of a 25 solar mass presupernova model are also presented, and the influence of uncertainties in the input physics is discussed; in particular, the rates of the (Mg-25)(p,gamma)(Al-26) and (Al-26)(p,gamma)(Si-27) key reactions are studied.

Arnould, M.↗

Space Shuttle main engine progress through the first flight

A system overview is presented for the Space Shuttle Main Engine (SSME), and extensive performance comparisons are made with the Saturn V booster powerplant showing (1) staged-combustion turbopump pressures of 7,500 vs 1,100 psi for the Saturn V first stage; (2) a specific impulse of 455 vs 300; (3) power density of 100 vs 20 hp/lb; and (4) chamber pressures of 3200 vs 1000 psia. An account is then given of the SSME testing program, engine certification process and flight aboard the Space Shuttle Columbia, with attention to anomalies detected during flight in turbopump operation and mixture ratio sensor function.

Johnson, J. R.↗

Discovery of s-process Nd in Allende residue

New interpretation is given to the isotopic anomalies detected by Lugmair et al. (1983) in an acid-resistant residue of the Allende meteorite. If the Nd-142 excess is due to Sm-146 decay, as the discoverers proposed, it is argued that the decay has occurred in interstellar grains, so that the conclusion that Sm-146 (1.03 x 10 to the 8th yr) was alive in the solar system is premature. It is shown by renormalizing their data that the discovery is likely to be s-process Nd, confirming the survival of red-giant stardust in carbonaceous interstellar dust.

Clayton, D. D.↗

Application of MAGSAT to Lithospheric Modeling in South America. Part 2: Synthesis of Geologic and Seismic Data for Development of Integrated Crustal Models

Research activities performed on MAGSAT scalar data over South America, Central America, and the adjacent marine areas are summarized. The geologic utility of magnetic anomalies detected by satellite is demonstrated by focusing on the spherical-Earth interpretation of scalar MAGSAT data in combination with ancillary geological and geophysical data to obtain lithospheric models for these regions related to their contemporary crustal dynamics processes, geologic history, current volcanism seismicity and natural resources.

Keller, G. R.↗

Analysis of LANDSAT-4 TM Data for Lithologic and Image Mapping Purpose

Lithologic mapping techniques using the near infrared bands of the Thematic Mapper onboard the LANDSAT 4 satellite are investigated. These methods are coupled with digital masking to test the capability of mapping geologic materials. Data are examined under medium to low Sun angle illumination conditions to determine the detection limits of materials with absorption features. Several detection anomalies are observed and explained.

Podwysocki, M. H.↗