Title Promoter Affiliations Abstract "Structural Health Monitoring Based on Operational Modal Analysis from Long Gauge Dynamic Strain Measurements" "Edwin Reynders" "Structural Mechanics Section" "Vibration-Based Monitoring (VBM) is a non-destructive method for structural damage identification and condition assessment. The main challenge for VBM is to identify structural characteristics that are as sensitive as possible to damage and at the same time as insensitive as possible to measurement noise, loading, and environmental factors such as temperature. Damage alters the stiffness, mass or energy dissipation of a structure, which are properties that are directly related to its modal characteristics. VBM aims to identify changes in modal characteristics of a structure that are directly related to damage. Natural frequencies and displacement mode shapes are probably the most commonly used modal characteristics for damage identification. However, they can exhibit a low sensitivity to certain types of damage while the sensitivity of natural frequencies to environmental influences may be sufficiently high to completely mask the effect of even severe damage. Modal strains are a promising alternative for damage identification as they are more sensitive to local damage. The use of modal strains for VBM applications is explored in this thesis. Two methods for accurate dynamic sub-microstrain measurements are  developed, to deal with the difficulties that current measurement techniques encounter in capturing the low strain levels that occur on civil structures during ambient or operational excitation. Fiber-optic Bragg gratings (FBG) are used to measure the dynamic strains, which subsequently are used in modal analysis to obtain the modal strains of the tested structures. Both methods exhibited very high accuracy in modal strain identification of laboratory and full-size structures. The damage detection and localization capabilities of modal strains and of related quantities that are directly obtained from them, such as neutral axis positions, are investigated through operational modal analyses on prestressed concrete beams that are subjected to progressive damage tests. It is demonstrated that the cracks that are induced in the beams by the progressive damage tests do have a clear and local influence on the strain mode shapes and neutral axis positions, even at low damage levels. Thorough experimental investigations are conducted in laboratory and on two railway bridges to identify the influence of temperature on modal strains. It is found that the temperature changes do not have a measurable influence on the modal strains and neutral axis positions." "Operational Modal Analysis Challenges in Case of Rotating Machines" "Francesco Marulo, Wim Desmet" "Production Engineering, Machine Design and Automation (PMA) Section" "Operational Modal Analysis (OMA) is widely employed and became an industrial standard technique for identifying the modal parameters (i.e. resonance frequencies, damping ratios and mode shapes) of mechanical structures. Its main applications are in the automotive, aerospace, civil engineering domains and many others. The advantage, if compared with Experimental Modal Analysis (EMA), is that it is not necessary to stop the machine, but its modal characteristics can be estimated during its operating cycles. In other words, OMA does not rely on known and deterministic excitation, but it uses exclusively the natural vibrations of the structure. It is very useful in cases in which the forces cannot be measured or when it is very difficult to excite a structure and it is more convenient to exploit the natural ambient excitation. The input forces are unknown and the main concept behind OMA is that the structure can be identified from a dynamic perspective by analyzing only the measured output signals.Several hypothesis need to be verified in order to apply OMA. First of all the structure must be Linear Time Invariant (LTI), but this is not the case if several parts are moving with respect to each other. Secondly, the forces acting on the structure must be represented by white noise in the frequency range of interest. This means that all the frequencies must be uniformly excited. This is often the case for wind excitation, but it is not valid anymore if periodical loads due to rotating elements are introduced in the system. The main scope of this dissertation is to fully understand and propose solutions to the challenges and the limitations occurring when applying classical OMA techniques in case of rotating machines. Several simulation and test cases will be discussed in order to validate the proposed solutions." "Operational modal analysis and structural health monitoring using transmissibility measurements and fiber sensor technology" "Christof Devriendt" "Applied Mechanics" "Operational Modal Analysis allows identifying the dynamic behavior of structures in operational conditions. In my previous and ongoing research a new OMA approach based on transmissibility measurements was proposed that increases the reliability and applicability of OMA techniques. The goal of structural health monitoring (SHM) is to detect damage before it becomes critical for the structure's integrity. In my previous research a promising approach based on transmissibility measurements for detecting and locating damage in changing operational conditions was proposed. The first promising steps have been made and transmissibility functions have proven to be very useful in the field of OMA and SHM. However, there are definitely a number of aspects that require more study and suggest a natural direction for future work. I also believe that, both in the field of OMA and SHM, the biggest leap forward will be made by investigating the combination of the above techniques with enhanced and innovative measurement systems, such as the Laser Doppler Vibrometer and fibre optic sensors. Finally the testing and monitoring of wind turbines are considered as one of the most relevant industrial applications of the proposed techniques and definitely require further investigation." "Baekeland mandate: Population-based resonance frequency monitoring for wind farms through augmented operational modal analysis" "Christof Devriendt" "Acoustics & Vibration Research Group, Applied Mechanics" "Monitoring the resonance frequencies of wind turbines (at sea) is a core product of 24SEA. The proposed project aspires to drastic improvements in the current algorithms, improvements that will make it possible to run this service bid on each turbine of a wind farm (compared to 10% now). In addition, the targeted use of machine learning will sensitivity and scope of the service are significantly improved." "Single-step Modal Parameter Identification and Vehicle Component Load Prediction using Modal Testing" "Wim Desmet" "Mecha(tro)nic System Dynamics (LMSD)" "Vibration analysis has always been a ubiquitous part of engineering, especially so since the 1960s, when experimental vibration analysis was essentially seeded with help from computing advances. Since then, modal analysis has been a key focus in key industries such as the automotive sector to not only characterise mechanical components, but also modify structures to avoid running into sometimes catastrophic issues due to structural dynamics. Over the past few decades, several procedures have been proposed for experimentally identifying modal parameters to characterise structures. These methods are matured and robust and have been even gathered under a common framework -- the Unified Matrix Polynomial Approach (UMPA), out of which some (like the poly-Reference Least Squares Complex Frequency(pLSCF) or PolyMax), have gained sustained commercial acceptance. However, it is observed that all state-of-the-art modal parameter estimation (MPE) algorithms have a common requirement, namely an ``expert'' who is aware of the intricacies involved in experimental modal analysis.This dissertation introduces a new MPE method called Direct Estimation of Residues from Rational-fraction Polynomials (DERRP), that computes all modal parameters in a single global least squares step. The complex poles are calculated as roots of the monic polynomial characteristic equation, which are the same as the eigen-values of the traditional companion matrix. A recursive algorithm is used to compute a closed-form solution for the residues (matrix product of mode shapes, participation vectors, and modal scaling). Numerical conditioning of the least-squares solution is handled via the discrete Z-transform. The DERRP algorithm is developed for generalised Multiple Input Multiple Output (MIMO) FRF matrices, within the UMPA framework. It results in accurate modal parameters ni comparison to those that are theoretically derived and experimentally obtained (using the widely accepted industrial standard pLSCF method). The trade-off is a quantified increase in operational complexity, which is also derived. With the currently advanced processors, this increase in computational time and memory is deemed acceptable.Furthermore, a new method to unify the spatial aspect of the modal problem is introduced as the ``longest vector'', wherein a consolidated set of measurement degrees-of-freedom (DoFs) is formed from all the reference and response DoFs. Modal deflections are estimated at DoFs where responses are not measured, but references are (and vice-versa) by using the reciprocal property of the residue matrix. This method is seen to work best with real modes as it utilises mean phases between modal vectors and participation vectors for projections. Scaling between the two vectors is identified using driving point residues in a least squares sense. The ``longest vector'' is shown to be useful for spatial enhancement (in terms of mode shape animations) as well as prediction of unmeasured frequency response functions (FRFs) through a reduction and expansion process.A modified stabilisation chart is also proposed, which compares not only the poles and participation vectors, but also the modal and ``longest vectors'' over increasing model order of the underlying rational fraction polynomial. The benefit purported is that the modal model is estimated in its entirety. A quantitative comparison can be made between the measured and modally reconstructed FRF matrices to estimate the best model order that describes the given structural FRF dataset. The Auto-Moment Frequency Response Function Similarity Metric (AMFRFSM) and Pole in FRF-Automoment Confidence (PiFAC) are introduced as a strategy to automate the complete MPE process and are shown to result in accurate modal parameter estimates. The DERRP algorithm allows an interactive selection of modes as well, which is instantaneously comparable through the AMFRFSM against the measured data for accuracy.Finally, application of modal testing to legacy automotive products is discussed with a focus on prediction of internal loads using modal test data. Free-free FRF matrices are used to compute transmissibility functions between boundary locations and DoFs of interest, and the vibrations are predicted against standard profiles applied based on component mounting strategy and driving conditions. These predictions are useful in determining the loads that the internal electric and electronic components undergo based on the assembly boundary conditions, such that sub-component testing can be carried out at an early stage of the development cycle." "German modal particles as multimodal constructions of intersubjectivity." "Kurt Feyaerts" "Multimodality, Interaction and Discourse, Kulak Kortrijk Campus, Multimodality, Interaction and Discourse, Leuven, Multimodality, Interaction and Discourse, Antwerp Campus" "Core of this project is a multimodal description of three clusters of German modal particles (MPs) which are defined onomasiologically: 1° particles which link a question to its context (typically denn and eigentlich), 2° particles which indicate that the content of the utterance should already be known by the hearer (typically ja and doch), and3° particles which mark the content of the utterance as obvious (typically eben, einfach, and halt). At the multimodal level, particular attention is paid to the co-occurrence of these MPs with gestures, the focus being on gestures which themselves convey MP meanings. Starting from a number of case studies, a detailed analysis of these co-occurrence patterns is offered, including a discussion of the factors which may influence their distribution. Finally, the question is raised whether these co-occurrence patterns can be analyzed as multimodal constructions inthe sense of Construction Grammar. Although this cannot be taken for granted, it is shown in the dissertation that a number of potential objections cannot be considered as true counterarguments, and that therefore, an analysis in terms of Construction Grammar is justified." "Processes of change in modal collocation." "Grammar and Pragmatics" "To express 'modality' (roughly the expression of necessity and possibility), English has lexical – e.g. 'modal adverbs' (probably, possibly, well, ...) –, and grammatical strategies – a paradigm of 'modal verbs' (can, will, should, may, must etc.). The combination of these strategies – 'modal collocation' (e.g. may well, could possibly) – has been under-researched and lacks a solid framework. The following facts are the background of my research question:(1) A decline in frequency of the modal verbs;(2) 'Emerging' constructions taking over traditional modal functions (e.g. havta, gonna);(3) A seeming resistance of decline for those modal verbs which co-occur more often with modal adverbs. The proposed research takes a cognitive-functional approach with reference to Construction Grammar. It investigates whether modal collocation should be understood as an emerging construction with strong internal dependencies and a level of schematicity. If this is found, a point can be made that we are dealing with constructional emergence at increasingly more schematic levels – 'constructionalisation'. This, in turn, could explain why the modal verbs that often collocate with modal adverbs have resisted a decline in use. I will additionally investigate whether there is competition between the modal verbs and modal collocation. I hypothesise that several processes of change are at work in modal collocation (e.g. lexicalisation and grammaticalisation), resulting in a continuum of utterances displaying degrees of constructionalisation and different functional uses. I also assume that modal collocation has played a role in the current-day frequencies of modal verbs in English. I will conduct corpus studies on data from Old English until Present Day English, looking into co-occurrence patterns of modal verbs and modal adverbs within the same sentence as well as the individual development of modal (ad)verbs. I will use three parameters for the analysis of constructionalisation (generality, schematicity and productivity) next to parameters traditionally found in grammaticalisation literature. I will also conduct statistical tests, such as collostructional analysis, to back up and elaborate on the findings. Confirming the hypothesis will not only offer insights into the workings of English modality, but it will strengthen the plausibility of constructional approaches to language change, which are still in need of empirical evidence." "Advanced performance assessment of existing concrete buildings by coupling 3D analysis of load arrangement, structure analysis including membrane action and test-based model updating" "Tom Molkens" "Materials and Constructions" "To meet climate objectives there is a clear need to deal differently with our building patrimony. In the future, buildings structures must be re-used possibly in combination with changes of their intended function. This research work will develop an assessment procedure for existing structures. After collecting the necessary historical data, the procedure focuses on the modelling of load arrangements through linear FEA. Despite being the starting point of further (advanced) analysis, disproportionally little research attention had been given to this. Often, simplified methods are used which results in conservative outcome. Starting from the results of the linear load model, a coupling will be proposed with a non-linear FEA to perform a structural analysis. In this non-linear FEA (NLFEA), the possibility to use membrane actions in ambient conditions and system behaviour is researched. To validate the proposed mathematical models a real test structure will be build on a scale of 1/3, which allows for model updating and perform test till the level of robustness checks. The use of NLFEA in structural analysis requires also the study of reliability levels and model uncertainties to obtain a credible working procedure." "Data analysis in the framework of the development of a shoulder model and analysis of the Biomechanics of the sternoclavicular joint." "Toon Huysmans" "Vision lab" "This project represents a formal service agreement between UA and on the other hand BVOT. UA provides BVOT research results mentioned in the title of the project under the conditions as stipulated in this contract." "Analysis of the market areas of intermodal terminals for transport to and from the sea ports of Zeebrugge and Ghent and coupling and benchmarking of foreign terminals by extending the existing Location Analysis Model of Belgian Intermodal Terminals (LAMBI" "Cathy Macharis" "Mathematics, Operational Research, Statistics and Information Systems for Management, Electromobility research centre" "This study analyses two components of the intermodal transport market in Flanders. First, it is discussed what the potential is of existing and additional intermodal links between the sea ports of Zeebrugge and Ghent and the Flemish hinterland. Second, the possible competition between Belgian intermodal terminals and terminals in France, the Netherlands, Germany and Luxemburg is researched."