263134 research outputs found
Sort by
Automatische herkenning van golfmodes in signalen van akoestische emissie en toepassing op druktesten op sferische opslagtanks
No full textNon-destructive testing (NDT) aims at finding defects in equipment before the emergence of any safety risk. Most non-destructive techniques detect defects based on their interaction with actively generated magnetic fields, eddy currents, X-rays or ultrasonic waves. Acoustic emission stands out as non-destructive technique because it relies on ultrasound emitted by the defects themselves. The signals resulting from recording the ultrasound emitted during a pressure test or during use are at the basis of the integrity evaluation using acoustic emission. Linking the recorded signals to the sources of the acoustic emission and determining the source type and severity however remains challenging. When acoustic emissions are generated during a test, the result is often a recommendation to perform follow-up NDT for additional source characterization and to further clarify what was the source of the acoustic emissions.
Extracting information about the source from the acoustic emission signals has received and still receives considerable attention. For methods proposed on this topic, wave modes are often of importance. Certain methods rely on different wave modes to deduct information about the source. For others, wave modes are of importance because the change of appearance of ultrasonic waves as they propagate through the tested object depends on the wave modes' dispersion relation and attenuation.
Manually, wave modes can be distinguished on a time-frequency representation of the signals. Monitoring structures with the acoustic emission technique however can lead to several thousands of signals, rendering manual mode recognition impracticable. Existing methods for automatic mode recognition on the other hand rely on assumptions regarding the wave modes in incoming signals of which the validity may depend on the source. Certain sources risk being systematically assessed wrongly. The aim of the current thesis is to develop a method for automatic mode recognition while avoiding assumptions that may be invalid for certain sources. In the thesis, the development of a new method is described starting from basic approaches to distinguish wave modes based on differences in their physical properties. The thesis treats automatic mode recognition on a dataset of AE signals on a plate, proposes a method for determining the end of a mode and makes the transition from a plate in the laboratory to spherical storage tanks. Finally, it results in a method applied to a gas storage sphere with both controlled, artificial sources and uncontrolled, "natural" sources.
For recording a dataset on a plate, Hsu-Nielsen sources and sensor pulses are used as well as melting ice. For Hsu-Nielsen sources, the results of automatic recognition of the start of a wave mode are in agreement with manual mode recognition in 95% of cases. For sensor pulses and melting ice, this is only the case for 75%-80% of the compared signals. Recognition of the end of a wave mode is proposed for this dataset to enable excluding reflections from the recognized wave modes. For recognition of the end of a wave mode, agreement is reported between 98% and 69% depending on the source type and the wave mode.
To apply this automatic mode recognition method to gas storage spheres, several changes were made to the method. A frequency filter is applied to the reference wavelets. Furthermore, a limitation is imposed on the number of cross-correlation peaks can be allowed for a signal to be analysed. Finally, the loss function for peak selection is adapted to limit the impact of a single signal on automatic mode recognition on all other signals caused by the same source event. With those adaptations, the method is tested on a dataset consisting of signals due to Hsu-Nielsen sources, sensor pulses, impact by a metallic object and not further identified "natural" sources. The automatically recognized start of the mode matches with manual recognition in 97% of cases for Hsu-Nielsen sources while ranging between 85% and 73% for "natural" sources depending on the wave mode. The cases where automatic wave mode recognition yields results that are in conflict with manual mode recognition are analysed to allow proposing further improvements. Even though successful recognition rates vary for different source types, a method is proposed in this thesis that avoids assumptions of which the validity depends on the source. Thereby, the current work represents a step towards enabling automatic mode recognition for acoustic emission testing for spherical storage tanks.status: Publishe
Underlying working mechanisms of virtual reality exposure: Exploring the role of fearful expectancies and habituation
No full textWhile research has primarily focused on establishing the efficacy of virtual reality (VR) exposure, relatively scant attention has been paid to the underlying working mechanisms that drive the effects. The present study examined the role of fearful expectancies and fear reduction (habituation) in VR exposure. Fearful expectancies were measured before, during (retrospectively), and after a VR exposure session in 121 participants with elevated fear of spiders. In addition, skin-conductance and heart rate were measured throughout the exposure session to examine fear reduction within the exercises and across the session. Fearful expectancies decreased after VR exposure. Larger decreases were associated with better outcomes 1 week (in the verbal and behavioral measures) and 3 months (in one of the verbal measures) after exposure. Levels of expectancies during exposure were not associated with the outcome. We did not find evidence that expectancies about own reactions were better testable in VR exposure than expectancies about the spider. Fear reduction within the exercises or across the session did generally not predict VR exposure outcome. It is recommended to focus on various operationalizations and experimental manipulations of the mechanisms, as well as to compare these mechanisms between VR and in vivo exposure in future research.sponsorship: The authors would like to thank Eline Camerman, Lotte Goossens and Tinne Vanempten for their help in collecting the data, and Mathijs Franssen (BePhysLab) for his technical assistance. We are grateful to Dirk Hermans and Yannick Boddez for earlier discussions about this topic. This work was supported by a KU Leuven C1 project (C16/19/02) and by an FWO red Noses project (G0D6322N) . The funding sources had no involvement in the study design, data collection or analysis. (KU Leuven C1 project|KU Leuven C1, KU Leuven C1 project|C16/19/02, KU Leuven C1 project|G0D6322N, FWO red Noses project|C16/19/02)status: Publishe
ESM-Q: A consensus-based quality assessment tool for experience sampling method items
No full textThe experience sampling method (ESM) is increasingly used by researchers from various disciplines to answer novel questions about individuals' daily lives. Measurement best practices have long been overlooked in ESM research, and recent reviews show that item quality is often not reported in ESM studies. The absence of information about item quality may be partly explained by the lack of consensus on how ESM item quality should be evaluated. As part of the ESM Item Repository project (esmitemrepository.com)-an international open science initiative that collects ESM items in an open item bank and evaluates their quality-we brought together 42 international ESM experts to develop an ESM item quality assessment tool. In four Delphi phases, experts suggested 57 item quality criteria, rated the criteria, provided arguments for and against the criteria, and rated the criteria again, considering reflections from other experts. The result of the Delphi process is ESM-Q: a quality assessment tool consisting of 10 core criteria, as well as an additional 15 supplementary criteria, to be used depending on the type of items being rated and the availability of supplementary information. The criteria cover topics ranging from construct validity to the optimal wording of items. ESM-Q can aid ESM researchers in selecting existing ESM items, developing new high-quality ESM items, and evaluating the quality of ESM items in systematic reviews. Expert reflections also highlight open research questions surrounding ESM item design that form a research agenda for ESM measurement.sponsorship: GE and AH were supported by Research Foundation Flanders (G0F8416N; G049023N; 1223725N), OJK by Research Foundation Flanders (FWO 1257821N) and KU Leuven (CPLUS/24/009), IMG and OJK by KU Leuven (C16/23/011), RWC by the National Institute on Alcohol Abuse and Alcoholism (NIAAA, K23 AA029729), SJW by National Institutes of Health grant R01DA055774, PK by IBOF grant iBOF/21/090 and KU Leuven C14/23/062, KN-G by a grant from the National Institute of Mental Health (R01MH118218), JAB by Stichting tot Steun VCVGZ (grant no. 239), VEH by the Dutch Research Council (NWO; 406.XS.04.051), LMB by the National Institute of Mental Health (R01MH118218, P50MH130957), KM by the Western Australian Future Health Research and Innovation Fund DF2022-23/1 and a Partnership Award #0059 from the Stan Perron Charitable Foundation, and ABN by the German Research Foundation (NE 2480/1-1). (IBOF|iBOF/21/090, Stichting tot Steun VCVGZ|grant no. 239, National Institute on Alcohol Abuse and Alcoholism|NIAAA, National Institute on Alcohol Abuse and Alcoholism|K23 AA029729)., Deutsche Forschungsgemeinschaft|NE 2480/1-1, Dutch Research Council|NWO, Dutch Research Council|406.XS.04.051)., KU Leuven|C14/23/062, KU Leuven|C16/23/011, Western Australian Future Health Research and Innovation Fund|DF2022-23/1, Fonds Wetenschappelijk Onderzoek|1223725N, National Institute of Mental Health|P50MH130957, National Institute of Mental Health|R01MH118218, National Institutes of Health|R01DA055774, Stan Perron Charitable Foundation|Partnership Award #0059, Research Foundation Flanders|1257821N), Research Foundation Flanders|G049023N, Research Foundation Flanders|G0F8416N)status: Publishe
New developments in experience sampling methodology
No full textExperience Sampling Methodology (ESM) has been widely used over the past decades to study feelings, behaviour and thoughts as they occur in daily life. Typically, participants complete several assessments per day via a smartphone for multiple days. The growing adoption of ESM has spurred a number of methodological advancements. In this paper, we provide an overview of recent developments in ESM design, statistical analysis and implementation. In terms of design, we discuss considerations around what to measure-including the reliability and validity of self-report measures as well as mobile sensing-as well as when to measure, where we focus on the pros and cons of burst designs and advances in sample size planning methodology. Regarding statistical analysis, we highlight non-linear models, survival analysis for understanding time-to-event data and real-time monitoring of ESM time series. At the implementation level, we address open science practices and advances in data preprocessing. Although most of the topics discussed in this paper are generic, many of the examples are focused on the study of affect in daily life.sponsorship: The writing of this article was supported by a KU Leuven Research Council grant (C14/23/062) awarded to Francis Tuerlinckx, Eva Ceulemans and Peter Kuppens, IBOF grant (IBOF/21/090) awarded to Peter Kuppens and Eva Ceulemans, and the EOS excellence of science program (EOS 4000528/G0I2422N) awarded to Eva Ceulemans. (Fonds Wetenschappelijk Onderzoek|C14/23/062, Fonds Wetenschappelijk Onderzoek|IBOF/21/090, KU Leuven Research Council|EOS 4000528/G0I2422N, EOS excellence of science program)status: Published onlin
Invloed van dynamische grond-structuurinteractie op de respons van periodische bruggen met meerdere overspanningen
No full textDynamic soil-structure interaction (SSI) affects the modal characteristics and response of civil engineering structures such as buildings, bridges, tracks and tunnels. The flexibility of a structure's foundation lowers its natural frequencies and increases its modal damping ratios, resulting in a possible positive or negative effect on its response. For railway bridges, analyzing the influence of dynamic SSI is important for three main reasons: (1) to ensure a safe and cost-effective design, (2) to improve the estimation of modal characteristics as commonly used in structural health monitoring applications and (3) to predict ground-borne vibration in urban environments.
The effect of dynamic SSI on a structure's response is usually assessed with numerical models. From a computational point of view, 3D element-based models are only feasible for short bridges with a few spans. Longer bridges, however, often consist of repeated identical spans, enabling the use of periodic structure theory to reduce the computational effort by modeling only a reference cell. This effectively reduces computation time and memory usage.
This thesis develops numerical models based on periodic structure theory for long, continuous multi-span bridges supported by piled foundations, incorporating dynamic SSI and foundation-soil-foundation interaction (FSFI), i.e. the interaction between neighboring foundations through the soil. A sub-structure approach is employed, where the dynamic stiffness of a piled foundation is precomputed using finite element-boundary element (FE-BE) models or finite element-perfectly matched layer (FE-PML) models.
First, a model based on the Floquet transform is developed. The structure's response is computed in the wavenumber-frequency domain using an FE-BE or FE-PML model of the reference cell. The inverse Floquet transform is then evaluated to determine the response in any cell. As integral transforms are used, the model implicitly assumes that the structure is infinitely long. It is demonstrated that dynamic SSI mainly affects the torsional and vertical bending modes with in-phase motion in neighboring spans. The effect of FSFI on the bridge response is generally negligible, especially when compared to the overall effect of dynamic SSI which is much more important.
To model periodic bridges of finite length, the wave finite element method (WFEM) is used. This model allows for any boundary conditions at the bridge ends and can also be employed to compute the response of quasi-periodic bridges. Additionally, the method directly provides the dispersion curves of the structure from the free wave characteristics of the reference cell. It is demonstrated that the natural frequencies decrease significantly due to dynamic SSI, especially for the first torsional mode with in-phase motion in neighboring spans. The modal damping ratios increase, particularly for the vertical bending modes. It is also shown that, in case of fixed footings, the response is higher for short bridges. By accounting for dynamic SSI, however, short and long bridges show a similar response, indicating that infinitely long bridge models could be used to predict the response of short bridges in this case.status: Publishe
Gecombineerde modellering van spin-kwantumbits en hun ladingsruisomgeving
No full textThis PhD research work aims at gaining a deeper understanding of the decoherence effects on Quantum Dots Spin-based Qubits. The work would involve studying the fundamentals of the system in order to build an experimentally relevant model for Qubit systems. The model would be used to explore ways for improvement of the decoherence with the goal of large scale integration of qubit systems. Experimental work would also be performed in order to test the theoretical studies and models in realistic implementations.status: Publishe
The Values of Fame: Exploring The Visual and Textual Representations of Basic Values in Influencers’ Instagram Content
No full textDespite the popularity of social media influencers (SMIs), little is known about how their content reflects and conveys certain values, leaving a gap in understanding their role as value intermediaries. This content analysis examined the representation of Schwartz values in the Instagram profiles of 59 most followed Western SMIs, celebrities, and athletes. Relying on 1,256 posts and 2,936 stories, the study documented the prevalence of values, modalities of representations (multimodal complexity and post-caption congruence), and differences between SMIs, athletes, and celebrities. Results revealed that 60.3% of the content portrayed at least one value, with achievement, benevolence, and hedonism being the most frequent. Multilevel analyses indicated that SMIs and athletes were more likely to post hedonism and benevolence, while celebrities were more likely to share universalism than SMIs. Most values were represented through low to medium levels of multimodal complexity, and only 15.3% showed post-caption congruence. These findings underscore the need to document how global digital platforms and actors mediate value representation, as they have the potential to shape audience values and cultural norms.sponsorship: European Research Council|852317status: Published onlin
The effect of high-pressure homogenization on physicochemical properties, flow behavior and thermal gelation of commercial pea protein isolate
No full textsponsorship: Y. Li is a Doctoral Researcher funded by KU Leuven Internal Funds. Q. Masijn is holder of a Flanders Research Foundation Strategic Basic Research grant (1S29922N, 2021-2025). (KU Leuven Internal Funds, Flanders Research Foundation Strategic Basic Research grant|1S29922N)status: Published onlin
Isidore of Pelusium
No full textAn exhaustive study of all the borrowings of Philo of Alexandria by Isidore of Pelusium.status: Publishe
AI-gedreven Oplossingen voor Neuromusculaire MRI: Van Spiersegmentatie tot Klinische Besluitvorming
No full textNeuromuscular diseases impact a significant number of people worldwide. Accurate and efficient evaluation of these conditions is essential for both clinical trials and everyday medical practice. Traditional methods for analyzing MRI scans, such as manual segmentation, are time-consuming and prone to inter-rater variability.
Recent advancements in artificial intelligence (AI) and deep learning, particularly convolutional neural networks like U-Net, have revolutionized medical image segmentation. These AI-driven techniques offer automated, precise, and rapid segmentation of muscles from MRI scans, significantly reducing the time and potential biases associated with manual methods. By integrating AI-based segmentation with quantitative MRI (qMRI), researchers can develop sensitive biomarkers, such as for instance proton density fat fraction percentage, to monitor disease progression and treatment efficacy.
In this project, advanced segmentation algorithms will be explored to precisely segment individual muscles in patients with diverse levels of fat replacement. A classification method will be designed to identify patients by analysing muscle patterns visible in MRI images, and XAI will be used to explain the decision-making process of the black box model.status: Publishe