Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Data processing and standardization”

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 1,207 records · Page 67

Pyrolysis_Molecular_Beam_Mass_Spectrometry_Analysis_of_hybrid_cross_of_Populus_tremula_x_P_alba_717-1B4_and_overexpression_of_a_lectin_receptor-like_kinase_(PtLecRLK1)

Stem tissues from the hybrid poplarPopulus tremula × P. albaclone 717-1B4 and from lectin receptor-like kinase overexpression lines PP7 and PP19 were individually colonized with the ectomycorrhizal fungiLaccaria bicolorstrain S238N,Hyaloscypha finlandicastrain PMI746, orUmbelopsis vinaceastrain PMI3018, as well as with a mixed fungal inoculum; non-inoculated plants served as controls. Plants were grown in a greenhouse at Oak Ridge National Laboratory and harvested in January 2025. Stem samples were analyzed using Pyrolysis–Molecular Beam Mass Spectrometry (Py-MBMS). Stems were harvested, debarked, dried, milled, destarched and ethanol extracted prior to analysis. Py-MBMS analysis was conducted using approximately 4 mg of wood from biomass and each sample was analyzed in duplicate. A Frontier PY2020 unit pyrolyzed samples at 500°C for 30 s in 80 µL deactivated stainless steel cups. An Extrel Super-Sonic MBMS Model Max 1000 was used to collect mass spectral data fromm/z30 to 450 at 17 eV and processed using Merlin Automation software (V3). Spectral ion intensities were normalized to the total ion chromatogram signal for each sample for analysis of spectral variance. Lignin content (wt %) was estimated based on relative responses from standards of known Klason lignin content using mean-normalized ion intensities ofm/z120, 124 (G), 137 (G), 138 (G), 150 (G), 152, 154 (S), 164 (G), 167 (S), 168 (S), 178 (G), 180, 181, 182 (S), 194 (S), 208 (S) and 210 (S) where G indicates guaiacyl-derived ions, S indicates syringyl-derived ions, and other ions either derive from other lignin monomers or multiple sources. Ratios of S and G lignin monomer units (S/G) were obtained by dividing the sum of S-based ions by the sum of G-based ions using mean-normalized ion intensities.

CBI↗

On the consistency of scale among experiments, theory, and simulation

As a tool for addressing problems of scale, we consider an evolving approach known as the thermodynamically constrained averaging theory (TCAT), which has broad applicability to hydrology. We consider the case of modeling of two-fluid-phase flow in porous media, and we focus on issues of scale as they relate to various measures of pressure, capillary pressure, and state equations needed to produce solvable models. We apply TCAT to perform physics-based data assimilation to understand how the internal behavior influences the macroscale state of two-fluid porous medium systems. A microfluidic experimental method and a lattice Boltzmann simulation method are used to examine a key deficiency associated with standard approaches. In a hydrologic process such as evaporation, the water content will ultimately be reduced below the irreducible wetting-phase saturation determined from experiments. This is problematic since the derived closure relationships cannot predict the associated capillary pressures for these states. We demonstrate that the irreducible wetting-phase saturation is an artifact of the experimental design, caused by the fact that the boundary pressure difference does not approximate the true capillary pressure. Using averaging methods, we compute the true capillary pressure for fluid configurations at and below the irreducible wetting-phase saturation. Results of our analysis include a state function for the capillary pressure expressed as a function of fluid saturation and interfacial area.

13 HYDRO ENERGY↗

An Overview of Project Planning for Hot-Isostatic Pressure Treatment of High-Level Waste Calcine for the Idaho Cleanup Project - 12289

The Calcine Disposition Project is responsible for retrieval, treatment by hot-isostatic pressure, packaging, and disposal of highly radioactive calcine stored at the Idaho Nuclear Technology and Engineering Center at the Idaho National Laboratory Site in southeast Idaho. In the 2009 Amended Record of Decision: Idaho High-Level Waste and Facilities Disposition Final Environmental Impact Statement the Department of Energy documented the selection of hot-isostatic pressure as the technology to treat the calcine. The Record of Decision specifies that the treatment results in a volume-reduced, monolithic waste form suitable for transport outside of Idaho by a target date of December 31, 2035. That target date is specified in the 1995 Idaho Settlement Agreement to treat and prepare the calcine for transport out of Idaho in exchange for allowing storage of Navy spent nuclear fuel at the INL Site. The project is completing the design of the calcine-treatment process and facility to comply with Record of Decision, Settlement Agreement, Idaho Department of Environmental Quality, and Department of Energy requirements. A systems engineering approach is being used to define the project mission and requirements, manage risks, and establish the safety basis for decision making in compliance with DOE O 413.3B, 'Program and Project Management for the Acquisition of Capital Assets'. The approach draws heavily on 'design-for-quality' tools to systematically add quality, predict design reliability, and manage variation in the earliest possible stages of design when it is most efficient. Use of these tools provides a standardized basis for interfacing systems to interact across system boundaries and promotes system integration on a facility-wide basis. A mass and energy model was developed to assist in the design of process equipment, determine material-flow parameters, and estimate process emissions. Data generated from failure modes and effects analysis and reliability, availability, maintainability, and inspectability analysis were incorporated into a time and motion model to validate and verify the capability to complete treatment of the calcine within the required schedule. The Calcine Disposition Project systems engineering approach, including use of industry-proven design-for-quality tools and quantitative assessment techniques, has strengthened the project's design capability to meet its intended mission in a safe, cost-effective, and timely manner. Use of these tools has been particularly helpful to the project in early design planning to manage variation; improve requirements and high-consequence risk management; and more effectively apply alternative, interface, failure mode, RAMI, and time and motion analyses at the earliest possible stages of design when their application is most efficient and cost effective. The project is using these tools to design and develop HIP treatment of highly radioactive calcine to produce a volume-reduced, monolithic waste form with immobilization of hazardous and radioactive constituents. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A Search for Neutral Supersymmetric Higgs Bosons at DØ

A search for Higgs bosons in multijet data from the DØ detector is reported in this thesis. The Higgs boson is the only remaining undiscovered particle in the Standard Model of particle physics, and plays an integral role in this model. It is known that this model is not a complete description of fundamental physics (it does not describe gravity, for example), and so searches for physics beyond the Standard Model are an important part of particle physics. One extension of the Standard Model, the Minimal Supersymmetric Standard Model (MSSM), predicts the existence of five Higgs bosons, two of which can show an enhanced coupling to bottom quarks. For this reason, a search in the bbb (multijet) channel is a sensitive test of Higgs boson physics. The analysis described in this thesis was conducted over 6.6 fb -1 of data. At the time of writing, the best limits on tan β (a key parameter of the MSSM) in the multijet channel were set by DØ. The new analysis described in this thesis included more data than the previous analysis in the channel, and made use of a new trigger and event-based analysis method. An improved Multivariate Analysis technique was used to separate signal and background events and produce a final discriminant for the limit setting process. These changes increased the expected sensitivity of this measurement by roughly 50% more than would be expected from the increase in the size of data sample alone.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

BEAM STOP DESIGN METHODOLOGY AND DESCRIPTION OF A NEW SNS BEAM STOP

The design of accelerator components such as magnets, accelerator cavities and beam instruments tends to be a fairly standardized and collective effort within the particle accelerator community with well established performance, reliability and, in some cases, even budgetary criteria. Beam stop design, by contrast, has been comparatively subjective historically with much more general goals. This lack of rigor has lead to a variety of facility implementations with limited standardization and minimal consensus on approach to development within the particle accelerator community. At the Spallation Neutron Source (SNS), for example, there are four high power beam stops in use, three of which have significantly different design solutions. This paper describes the design of a new off-momentum beam stop for the SNS. The technical description of the system will be complemented by a discussion of design methodology. This paper presented an overview of the new SNS HEBT off-momentum beam stop and outlined a methodology for beam stop system design. The new beam stop consists of aluminium and steel blocks cooled by a closed-loop forced-air system and is expected to be commissioned this summer. The design methodology outlined in the paper represents a basic description of the process, data, analyses and critical decisions involved in the development of a beam stop system.

36 MATERIALS SCIENCE↗

A novel coupling of noise reduction algorithms for particle flow simulations

Proper orthogonal decomposition (POD) and its extension based on time-windows have been shown to greatly improve the effectiveness of recovering smooth ensemble solutions from noisy particle data. However, to successfully de-noise any molecular system, a large number of measurements still need to be provided. In order to achieve a better efficiency in processing time-dependent fields, we have combined POD with a well-established signal processing technique, wavelet-based thresholding. In this novel hybrid procedure, the wavelet filtering is applied within the POD domain and referred to as WAVinPOD. The algorithm exhibits promising results when applied to both synthetically generated signals and particle data. In this work, the simulations compare the performance of our new approach with standard POD or wavelet analysis in extracting smooth profiles from noisy velocity and density fields. Numerical examples include molecular dynamics and dissipative particle dynamics simulations of unsteady force- and shear-driven liquid flows, as well as phase separation phenomenon. Simulation results confirm that WAVinPOD preserves the dimensionality reduction obtained using POD, while improving its filtering properties through the sparse representation of data in wavelet basis. This paper shows that WAVinPOD outperforms the other estimators for both synthetically generated signals and particle-based measurements, achieving a higher signal-to-noise ratio from a smaller number of samples. The new filtering methodology offers significant computational savings, particularly for multi-scale applications seeking to couple continuum informations with atomistic models. It is the first time that a rigorous analysis has compared de-noising techniques for particle-based fluid simulations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Constraining the destruction rate of K 40 in stellar nucleosynthesis through the study of the Ar 40 ( p , n ) K 40 reaction

Background: We present that K 40 plays a significant role in the radiogenic heating of Earth-like exoplanets, which can affect the development of a habitable environment on their surfaces. The initial amount of K 40 in the interior of these planets depends on the composition of the interstellar clouds from which they formed. Within this context, nuclear reactions that regulate the production of K 40 during stellar evolution can play a critical role. Purpose: In this study, we constrain for the first time the astrophysical reaction rate of K 40 ( n , p ) Ar 40 , which is responsible for the destruction of K 40 during stellar nucleosynthesis. We provide to the nuclear physics community high-resolution data on the cross section and angular distribution of the Ar 40 ( p , n ) K 40 reaction. These are important to various applications involving Ar 40 . The associated reaction rate of the Ar 40 ( p , n ) K 40 process addresses a reaction rate gap in the Joint Institute for Nuclear Astrophysics REACLIB database in the region of intermediate-mass isotopes. Methods: We performed differential cross-section measurements on the Ar 40 ( p , n ) K 40 reaction, for six energies in the center-of-mass system between 3.2 and 4.0 MeV and various angles between 0 ° and 135 ° . The experiment took place at the Edwards Accelerator Laboratory at Ohio University using the beam swinger target location and a standard neutron time-of-flight technique. We extracted total and partial cross sections by integrating the double differential cross sections we measured. Results: The total and partial cross sections varied with energy due to the contribution from isobaric analog states and Ericson type fluctuations. The energy-averaged neutron angular distributions were symmetrical relative to 90 ° . Based on the experimental data, local transmission coefficients were extracted and were used to calculate the astrophysical reaction rates of Ar 40 ( p , n ) K 40 and K 40 ( n , p ) Ar 40 reactions. The new rates were found to vary significantly from the theoretical rates in the REACLIB library. We implemented the new rates in network calculations to study nucleosynthesis via the slow neutron capture process, and we found that the produced abundance of K 40 is reduced by up to 10% compared to calculations with the library rates. At the same time, the above result removes a significant portion of the previous theoretical uncertainty on the K 40 yields from stellar evolution calculations. Conclusions: Our results support a destruction rate of K 40 in massive stars via the K 40 ( n , p ) Ar 40 reaction that is larger compared to previous estimates. The rate of K 40 destruction via the K 40 ( n , p ) Ar 40 reaction now has a dramatically reduced uncertainty based on our measurement. Lastly, this result directly affects the predicted stellar yields of K 40 from nucleosynthesis, which is a critical input parameter for the galactic chemical evolution models that are currently employed for the study of significant properties of exoplanets.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Search for resonant leptoquark production via lepton-jet signatures in pp collisions at $\sqrt{s}=13$ TeV and $\sqrt{s}=13.6$ TeV with the ATLAS detector

This paper presents a search for physics beyond the Standard Model targeting a heavy resonance visible in the invariant mass of the lepton-jet system. The analysis focuses on final states with a high-energy lepton and jet, and is optimised for the resonant production of leptoquarks — a novel production mode mediated by the lepton content of the proton originating from quantum fluctuations. Four distinct and orthogonal final states are considered: e+light jet, μ+light jet, e+b-jet, and μ+b-jet, constituting the first search at the Large Hadron Collider for resonantly produced leptoquarks with couplings to electrons and muons. Events with an additional same-flavour lepton, as expected from higher-order diagrams in the signal process, are also included in each channel. The search uses proton-proton collision data from the full Run 2, corresponding to an integrated luminosity of 140 fb −1 at a centre-of-mass energy of $\sqrt{s}=13$ TeV, and from a part of Run 3 (2022–2023), corresponding to 55 fb −1 at $\sqrt{s}=13.6$ TeV. No significant excess over Standard Model predictions is observed. The results are interpreted as exclusion limits on scalar leptoquark ($\tilde{S}$ 1 ) production, substantially improving upon previous ATLAS constraints from leptoquark pair production for large coupling values. The excluded $\tilde{S}$ 1 $\tilde{S}$ 1 mass ranges depend on the coupling strength, reaching up to 3.4 TeV for quark-lepton couplings y de = 1.0, and up to 4.3 TeV, 3.1 TeV, and 2.8 TeV for y sμ , y be , and y bμ couplings set to 3.5, respectively.

Hadron-Hadron Scattering↗

Laser Disdrometer Quantities (LDQUANTS) and Video Disdrometer Quantities (VDISQUANTS) Value-Added Products Report

Disdrometers are useful instruments that measure the distribution of raindrop sizes (drop size distribution: DSD) and the associated rainfall rates/accumulation as those raindrops fall to the ground. These DSDs are a quantity of interest to members of the modeler and observational communities. However, use of disdrometer data for model evaluation, radar monitoring, or other activities requires careful quality control and processing for key DSD properties of interest (e.g., the number concentration of drops) to ensure appropriate physical (scattering, fall speed) assumptions. The U.S. Department of Energy Atmospheric Radiation Measurement (ARM) Laser Disdrometer Quantities and Video Disdrometer Quantities Value-Added Product (LDQUANTS/VDISQUANTS VAP) uses standard methods from Tokay et al. (e.g., 2013, 2014) to filter drops with unrealistic fall speeds. Further, it estimates several microphysical/geophysical quantities of parameterized DSDs (gamma or exponential assumption type fitting methods) as in previous disdrometer studies and ARM long-term efforts (e.g., Testud et al. 2001, Giangrande et al. 2014, Thompson et al. 2015). Disdrometers are also beneficial for cross-checks with other instrumentation, including rain gauges and radars. To support research interests and related radar monitoring activities, this product calculates radar equivalent quantities, including dual polarization radar quantities (e.g., Z, Differential Reflectivity ZDR, etc.), using T-Matrix scattering and additional wavelength, temperature, and drop shape assumptions (e.g., Thurai et al. 2007). All of these efforts allow disdrometer data sets to become more useful and easily handled by modeling and observational studies, or routine instrumentation checks.

47 OTHER INSTRUMENTATION↗

Next generation models for storage and representation of microbial biological annotation

Background: Traditional genome annotation systems were developed in a very different computing era, one where the World Wide Web was just emerging. Consequently, these systems are built as centralized black boxes focused on generating high quality annotation submissions to GenBank/EMBL supported by expert manual curation. The exponential growth of sequence data drives a growing need for increasingly higher quality and automatically generated annotation. Typical annotation pipelines utilize traditional database technologies, clustered computing resources, Perl, C, and UNIX file systems to process raw sequence data, identify genes, and predict and categorize gene function. These technologies tightly couple the annotation software system to hardware and third party software (e.g. relational database systems and schemas). This makes annotation systems hard to reproduce, inflexible to modification over time, difficult to assess, difficult to partition across multiple geographic sites, and difficult to understand for those who are not domain experts. These systems are not readily open to scrutiny and therefore not scientifically tractable. The advent of Semantic Web standards such as Resource Description Framework (RDF) and OWL Web Ontology Language (OWL) enables us to construct systems that address these challenges in a new comprehensive way. Results: Here, we develop a framework for linking traditional data to OWL-based ontologies in genome annotation. We show how data standards can decouple hardware and third party software tools from annotation pipelines, thereby making annotation pipelines easier to reproduce and assess. An illustrative example shows how TURTLE (Terse RDF Triple Language) can be used as a human readable, but also semantically-aware, equivalent to GenBank/EMBL files. Conclusions: The power of this approach lies in its ability to assemble annotation data from multiple databases across multiple locations into a representation that is understandable to researchers. In this way, all researchers, experimental and computational, will more easily understand the informatics processes constructing genome annotation and ultimately be able to help improve the systems that produce them.

Biochemistry & Molecular Biology↗

Next Generation Models for Storage and Representation of Microbial Biological Annotation

Background Traditional genome annotation systems were developed in a very different computing era, one where the World Wide Web was just emerging. Consequently, these systems are built as centralized black boxes focused on generating high quality annotation submissions to GenBank/EMBL supported by expert manual curation. The exponential growth of sequence data drives a growing need for increasingly higher quality and automatically generated annotation. Typical annotation pipelines utilize traditional database technologies, clustered computing resources, Perl, C, and UNIX file systems to process raw sequence data, identify genes, and predict and categorize gene function. These technologies tightly couple the annotation software system to hardware and third party software (e.g. relational database systems and schemas). This makes annotation systems hard to reproduce, inflexible to modification over time, difficult to assess, difficult to partition across multiple geographic sites, and difficult to understand for those who are not domain experts. These systems are not readily open to scrutiny and therefore not scientifically tractable. The advent of Semantic Web standards such as Resource Description Framework (RDF) and OWL Web Ontology Language (OWL) enables us to construct systems that address these challenges in a new comprehensive way. Results Here, we develop a framework for linking traditional data to OWL-based ontologies in genome annotation. We show how data standards can decouple hardware and third party software tools from annotation pipelines, thereby making annotation pipelines easier to reproduce and assess. An illustrative example shows how TURTLE (Terse RDF Triple Language) can be used as a human readable, but also semantically-aware, equivalent to GenBank/EMBL files. Conclusions The power of this approach lies in its ability to assemble annotation data from multiple databases across multiple locations into a representation that is understandable to researchers. In this way, all researchers, experimental and computational, will more easily understand the informatics processes constructing genome annotation and ultimately be able to help improve the systems that produce them.

59 BASIC BIOLOGICAL SCIENCES↗

Dark matter and strong electroweak phase transition in a radiative neutrino mass model

We consider an extension of the standard model (SM) with charged singlet scalars and right handed (RH) neutrinos all at the electroweak scale. In this model, the neutrino masses are generated at three loops, which provide an explanation for their smallness, and the lightest RH neutrino, N{sub 1}, is a dark matter candidate. We find that for three generations of RH neutrinos, the model can be consistent with the neutrino oscillation data, lepton flavor violating processes, N{sub 1} can have a relic density in agreement with the recent Planck data, and the electroweak phase transition can be strongly first order. We also show that the charged scalars may enhance the branching ratio h→γγ, where as h→γZ get can get few percent suppression. We also discuss the phenomenological implications of the RH neutrinos at the collider.

79 ASTRONOMY AND ASTROPHYSICS↗

Optical modeling and electrical properties of cadmium oxide nanofilms: Developing a meta–heuristic calculation process model

Cadmium oxide thin films were deposited onto glass substrates by sol–gel dip-coating method and annealed in air. The normal incidence transmittance of the films was measured by a spectrophotometer. D.C electrical parameters such as carrier concentration and mobility were analyzed by Hall Effect measurements. A combination of Forouhi–Bloomer and standard Drude model was used to simulate the optical constants and thicknesses of the films from transmittance data. The transmittance spectra of the films in the visible domain of wavelengths were successfully fitted by using the result of a hybrid particle swarm optimization method and genetic algorithm. The simulated transmittance is in good accordance with the measured spectrum in the whole measurement wavelength range. The electrical parameters obtained from the optical simulation are well consistent with those measured electrically by Hall Effect measurements.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Search for heavy bottom-like quarks in 4.9 fb-1 of pp collisions at $ \sqrt {s} = 7 $ TeV

Results are presented from a search for heavy bottom-like quarks, pair-produced in pp collisions at sqrt(s) = 7 TeV, undertaken with the CMS experiment at the LHC. The b' quarks are assumed to decay exclusively to tW. The b' anti-b' to t W(+) anti-t W(-) process can be identified by its distinctive signatures of three leptons or two leptons of same charge, and at least one b-quark jet. Using a data sample corresponding to an integrated luminosity of 4.9 inverse femtobarns, observed events are compared to the standard model background predictions, and the existence of b' quarks having masses below 611 GeV is excluded at 95% confidence level.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Innovative Method for Enhancing Carbon Fibers Dispersion in Wet-Laid Nonwovens

The production of wet laid (WL) reinforcing fiber mats is adopted from the paper making industry to produce carbon fiber (CF) mats. The resulting mats are prone to defects induced by the mixing regime. Here, this study explores a new mixer design that achieved fully dispersed mats with 20 minutes processing time using 25.4 mm long CF. The proposed mixing method is compared against the traditional method and the density distribution of fibers is characterized using Back Light Scattering (BLS) technique. The mats showed high consistency with less than 8% standard deviation from the data collected from different regions within the mats. The innovative mixer resulted in mats with fiber distribution 70% closer to the theoretical distribution than that of the traditional mixer. This innovative method of preparing nonwoven CF mats will create new opportunities for CF nonwoven composite applications.

36 MATERIALS SCIENCE↗

Search for heavy, top-like quark pair production in the dilepton final state in pp collisions at s = 7 TeV

The results of a search for pair production of a heavy, top-like quark, t', in the decay mode (t' anti-t') to (b anti-W anti-b W) to (b anti-lepton neutrino anti-b lepton anti-neutrino) are presented. The search is performed with a data sample corresponding to an integrated luminosity of 5.0 inverse femtobarns in pp collisions at a center-of-mass energy of 7 TeV, collected by the CMS experiment at the LHC. The observed number of events agrees with the expectation from standard model processes, and no evidence of t' anti-t' production is found. Upper limits on the production cross section as a function of t' mass are presented, and t' masses below 557 GeV/c^2 are excluded at the 95% confidence level.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Generative adversarial networks for scintillation signal simulation in EXO-200

Generative Adversarial Networks trained on samples of simulated or actual events have been proposed as a way of generating large simulated datasets at a reduced computational cost. In this work, a novel approach to perform the simulation of photodetector signals from the time projection chamber of the EXO-200 experiment is demonstrated. The method is based on a Wasserstein Generative Adversarial Network — a deep learning technique allowing for implicit non-parametric estimation of the population distribution for a given set of objects. Our network is trained on real calibration data using raw scintillation waveforms as input. We find that it is able to produce high-quality simulated waveforms an order of magnitude faster than the traditional simulation approach and, importantly, generalize from the training sample and discern salient high-level features of the data. In particular, the network correctly deduces position dependency of scintillation light response in the detector and correctly recognizes dead photodetector channels. Furthermore, the network output is then integrated into the EXO-200 analysis framework to show that the standard EXO-200 reconstruction routine processes the simulated waveforms to produce energy distributions comparable to that of real waveforms. Finally, the remaining discrepancies and potential ways to improve the approach further are highlighted.

47 OTHER INSTRUMENTATION↗

DOC-DICAM: Domain Aware One Class Defect Identification in Composite Aerostructure Material

Fiber-reinforced composites are a common material used in the design of aircraft structures due to their good tensile strength and resistance to compression. During the manufacturing process, these structures are thoroughly inspected for flaws and defects to ensure structural integrity during commercial use. Non-destructive testing (NDT) is a collection of inspection methods that allow inspectors to evaluate material without altering it. Due to the high safety standards in aerospace manufacturing, the NDT process is done manually and can be a significant bottleneck in the development workflow. In this paper, we develop an AI-based assistance tool to drastically reduce inspection time. Typical AI workflows require large amounts of annotated data, but defects rarely occur resulting in strong class imbalance. To overcome this, we formulate the problem of defect identification as an anomaly detection task in which our primary focus is learning non-defect characteristics. To do this, we develop a multi-task self-supervised learning framework that embeds problem specific domain knowledge into the deep learning model. We verify our method using fuselage data generated in a production environment. As a result, we show that our method can effectively identify defects and requires minimal training and inference time.

anomaly detection↗