Search

© Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics. All rights reserved. When conducting uncertainty quantification, probability methods are widely used to represent quantities of a variable nature. First, a type of probability density function has to be chosen, often based on initial assumptions and physical boundary conditions, defining a set of stochastically parameters. Because the available data is physically limited to a finite number, there will always be uncertainty on the exact value of these parameters. At some point, the amount of experiments will be too low for the estimated parameters to be of practical use. In the case of very low data availability, one can attempt to use intervals instead to quantify the variability. Because no probability density function is defined, intervals can be used in the presence of low data availability, leaving only an upper and lower bound to be defined. This paper will present a method to estimate interval bounds based on a limited set of experiments, trying to optimally use the information they contain and determine practically usable interval bounds. ispartof: pages:5207-5214 ispartof: Proceedings of ISMA2018-USD2018 pages:5207-5214 ispartof: USD2018 location:Leuven, Belgium date:17 Sep - 19 Sep 2018 status: published

Dynamic fluid loading is of major concern to the off shore industry. Impacted structures may exhibit accel- erated fatigue and even failure. Load monitoring has therefore become an integral part of structural health monitoring. Obtaining direct load measurements can be cumbersome, e.g. obtaining strain gauge or pres- sure taps measurements. In this work, an alternative indirect method is proposed. The method revolves around building a mapping between more easily accessible wake velocity measurements and the induced loads. Classical machine learning techniques are adopted and relatively compact models, able to be run in real-time, are obtained. The method is illustrated on vortex-induced loads on a submerged cylinder.

Global non-destructive testing methods use changes in modal data as damage indicators. Depending on their geometry, damages influence the modal parameters in different way. While breathing cracks influence the natural frequencies due to only stiffness change, open cracks affect supplementary these frequencies due to loss of mass. This paper describes a study devoted to clarify the influence of the damage geometry on the natural frequencies of beams considering the two above mentioned aspects. First, analytical and numerical analyses are performed, in order to highlight the effect of mass changes on the frequency changes. Afterwards, the study focuses on breathing cracks in open and closing stages as well as on open cracks and emphasizes the influence of the geometrical discontinuity. It was demonstrated that, for beams with a breathing crack, a two degrees of freedom model is necessary to characterize its dynamic behavior. Finally, a relation between frequency changes and damage geometry and location is contrived.

ispartof: pages:783-798 ispartof: PROCEEDINGS OF INTERNATIONAL CONFERENCE ON NOISE AND VIBRATION ENGINEERING (ISMA2014) AND INTERNATIONAL CONFERENCE ON UNCERTAINTY IN STRUCTURAL DYNAMICS (USD2014) pages:783-798 ispartof: ISMA 2014 location:Leuven, Belgium date:Sep - Sep 2014 status: published

This paper presents a novel method for the assessment of power radiated by a (treated) test panel. Conventional measurement techniques for panel radiation typically use either a diffuse field or a free-field as receiving environment. However, such an environment is not trivial to create, especially in a compact set-up. Therefore, a method is proposed to retrieve radiated power from a set of pressure measurements that can be applied in a small, reverberant receiving cavity. A base of modal shapes is used to describe the radiating panel velocity, complemented with a direct acoustic simulation of the receiving cavity. In this model, panel shape contribution factors are estimated from the pressure measurements. Finally, radiated power can be calculated from this model. For validation purpose, a bare plate case is considered, showing a good agreement between the original and reconstructed radiated power. ispartof: pages:4037-4046 ispartof: PROCEEDINGS OF INTERNATIONAL CONFERENCE ON NOISE AND VIBRATION ENGINEERING (ISMA2014) AND INTERNATIONAL CONFERENCE ON UNCERTAINTY IN STRUCTURAL DYNAMICS (USD2014) pages:4037-4046 ispartof: International Conference on Noise and Vibration Engineering location:Leuven date:15 Sep - 17 Sep 2014 status: published

An accurate stress or strain history at fatigue critical locations is often needed for a fatigue assessment. Unfortunately it is not feasible to install strain gauges as these fatigue hotspots. This contribution compares two techniques to obtain a reliable stress history at any location of the turbine structure, one is based on modal decomposition and expansion, the other is based on a Kalman filter. Both techniques will be validated and compared using data from an offshore wind turbine monitored by OWI-lab. The monitored turbine is instrumented with strain gauges at the interface between transition piece and tower and accelerometers at multiple levels. The installed strain gauges allow to validate the proposed techniques with respect to the reality.

DTU Wind Energy continues the experimental investigation of the wind turbine blades to assess innovative designs of long and slender blades. This paper presents an experimental structural dynamics identification and structural model validation of the 14.3m long research blade. Unique feature of the blades is that its internal layup design has been highly optimized w.r.t. stretching the rotor and substantial mass reduction at the same time. As the result, the blade is more flexible than the traditional one. The results of the modal tests following analyses were performed: (i) Uncertainty Quantification of the experimental modal parameters for the blades, (ii) non-linearity assessment, (iii) numerical model correlation – frequencies and mode shapes of the experimental model comparison with those from Finite Element (FE). Finally, the outlook for the future experimental blade research activity is outlined.

Latest trends in the industry of mechanical transmissions demands for a new set of design tools, able to work beyond the assumption of the traditional analytical formulation. Recent progress in the field of model order reduction schemes finally made feasible to deal with dynamic 3D simulations of contact problems for both gears and bearing components, thus posing the base for a new frontier of simulation capabilities where different operational conditions can be analyzed. This paper aims at integrating some of these techniques into a commercial multi-domain 1D simulation software package, moving the first steps towards efficient system level analysis of typical 3D effects like micro/macro geometry modification, misaligned mounting, and perturbed loading conditions. A first proof-of-concept implementation is discussed along the paper. Furthermore, a broad set of case studies is presented, in order to showcase the main advantages of the proposed methodology, with respect to available commercial solution. ispartof: pages:1357-1370 ispartof: ISMA2016 including USD2016 pages:1357-1370 ispartof: Leuven Conference on Noise and Vibration Engineering (ISMA 2016) location:Leuven date:19 Sep - 21 Sep 2016 status: published

Direct Field Environmental Acoustic testing consists in reproducing the sound field inside a spacecraft vehicle's fairing using an array of loudspeakers. In order to ensure the accurate reproduction of the sound pressure field, the actuators need to be actively controlled. The performance of state-of-the-art Multiple-Input Multiple-Output (MIMO) random controllers is strongly dependent on the control strategy. The system identification, methodology applied at low power level excitation before the test, is one of the most sensitive aspects of such strategy. This is because it provides a linear nonparametric model of the electroacoustic system, which is used to design the signals to drive the loudspeakers. In this work novel system identification strategy is explored. The proposed method is a MIMO Best Linear Approximation (BLA). In this technique, instead of using classical excitation signals and the H1 estimate, a BLA model is estimated, and the coherence function is split into noise level and nonlinear contribution estimates.

© Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics. All rights reserved. Recent developments indicate the possibility to obtain compact reduced order models for systems with many inputs by Krylov subspace techniques. Therefore, this paper looks into the possibility of applying these reduced order models for the numerical assembly of a structure consisting of unit cells with dynamic substructuring methods. The proposed strategy consists of two steps: First the unit cell is reduced by using the aforementioned Krylov subspace technique, where the inputs are defined as the interface degrees of freedom. Then, these reduced unit cells are combined by a dual assembly technique to arrive at the numerical model of the full structure. This model is then reduced for a second time with Krylov based model order reduction. Several strategies for the assembly of the reduced basis on unit cell level are shown. The performance of the several reduction strategies are compared, and show that the resulting reduced order models are very accurate and compact. no ISSN ispartof: pages:445-456 ispartof: Proceedings of the International Conference on Noise and Vibration Engineering (ISMA) / International Conference on Uncertainty in Structural Dynamics (USD) pages:445-456 ispartof: International Conference on Noise and Vibration Engineering (ISMA 2018) location:Leuven, Belgium date:17 Sep - 19 Sep 2018 status: Published online

The assessment of the contact and friction forces acting on guiding rails is of high importance in the design and maintenance of mechanical systems. The direct measurement of these forces is however difficult as the use of a dedicated sensor is impractical and in some cases unfeasible. The use of model-based virtual sensing approaches is an attractive option to estimate these forces using off-the-shelf sensors. In this paper a nonlinear model that includes the lateral dynamics of a vertical transportation system is developed. The model is then combined with simulated measurements by means of an Extended Kalman Filter to estimate the forces in the guiding system. Special attention is given to reducing the modeling effort while keeping a highly accurate model. With that purpose the model is developed in Modelica and combined with the estimation filters by means of the Functional Mock-up Interface. The effect of the uncertainties in the characteristics of the contact model is also assessed. The estimated forces will be analyzed to monitor the condition of the guiding rails. ispartof: pages:1203-1215 ispartof: PROCEEDINGS OF ISMA2016 INTERNATIONAL CONFERENCE ON NOISE AND VIBRATION ENGINEERING AND USD2016 INTERNATIONAL CONFERENCE ON UNCERTAINTY IN STRUCTURAL DYNAMICS pages:1203-1215 ispartof: International Conference on Noise and Vibration Engineering (ISMA) / International Conference on Uncertainty in Structural Dynamics (USD) location:Leuven date:19 Sep - 21 Sep 2016 status: published

Requirements for the dynamic properties of a mechanical structure are commonly expressed in terms of eigenfrequencies. Therefore, during the design phase of a mechanical structure much attention is paid to natural frequencies and mode shapes. However, these modal properties tend to vary strongly if small variations are applied on the design model. Therefore, the presence of open design decisions or other uncertain parameters complicates strongly a meaningful dynamic design validation in an early design stage. A fuzzy finite element method for the eigenvalue analysis has been developed to cope with this problem. This method results in fuzzily defined modal properties indicating the degree of uncertainty on the eigenfrequencies and eigenmodes resulting from the uncertain model. However, rather than eigenfrequencies, the specification of responses at frequencies that are critical in the operating conditions of the structure would be a more realistic concept for dynamic design requirements. In order to facilitate the use of such design requirements, a tool is needed to analyze the influence of design decisions on responses at these specific frequencies. Therefore a method for the calculation of fuzzy frequency response functions of uncertain structures is presented. The procedure calculates guaranteed bounds on the envelope on the response function starting from interval bounded uncertainties on the input model. It necessitates the calculation of guaranteed boundaries on the contribution of each individual mode to the total frequency response function. ispartof: pages:417-424 ispartof: Proceedings of the 25th International Conference on Noise and Vibration Engineering, ISMA pages:417-424 ispartof: International Conference on Noise and Vibration Engineering location:Leuven, Belgium date:13 Sep - 15 Sep 2000 status: published

This paper presents an efficient nonparametric time-varying (TV) system identification method for the Operational Modal Analysis (OMA) framework. OMA tackles industrial measurements of vibrating structures in real-life operating conditions without the exact knowledge of the excitation signal. The main issue is that the dynamics of underlying systems may vary significantly when operating in real-life conditions. In this case it is desired to carefully verify and track the vibration behavior, since during flutter appearance, system destruction can occur. In this work the first results of a regularized time domain based TV OMA method are presented to estimate the linear TV output autocorrelation function of the observed system. The method is illustrated on the measurement of wind tunnel test of an airplane model. Using the proposed methodology, the estimated TV 2D output autocorrelation model provides a good data-fit with respect to tracking of varying resonances.

In this article, we develop a nonlinear data-driven model of the aerodynamic force on an aerofoil made to oscillate in pitch in a wind tunnel under constant flow conditions. The proposed model structure is a polynomial nonlinear state-space model (PNLSS) which is an extension of the classical linear state-space model with nonlinear functions. In this work, we use experimental data of an aerofoil onto which a pitching motion is imposed, using a dedicated cyber-physical fluid dynamic (CPFD) setup. The pitch angle is considered as the input to the system, while the aerodynamic force is considered as the output of the system. Different input signals (single sines, multisines) are considered at different amplitudes and offset angles for the pitching aerofoil. Results show that the PNLSS model is three times more accurate than the linear model in minimising the least squares cost function. To the author’s knowledge, this constitutes the first instance of a fully nonlinear state space model of an oscillating aerofoil developed from experimental measurements.

Periodical loads, such as waves and rotating machinery, are a problem for operational modal analysis (OMA). In OMA only the vibrations of a structure of interest are measured and little to nothing is known about the loads causing these vibrations. Therefore, it is often assumed that all dynamics in the measured data are linked to the system of interest. A novel approach to this problem is the research into time-varying transmissibilities for OMA, recently pub- lished in MSSP. In that contribution it was shown how to use time-varying single-reference transmissibilities to estimate the modal parameters from a single dataset without being influenced by the periodical loads. However, transmissibilities are only independent of the input spectra, and therefor the periodical loads, when the number of references equals the number of uncorrelated loads. This property limits the applicability of a single-reference approach to systems excited by a single distributed load. The solution is to use multi-reference transmissibilities for OMA a.k.a. pTOMA. With pTOMA the estimated modal parameters can again be independent of the input spectra. The main goal of this paper is to expand the estimation of time-varying transmissibilities to multi-reference transmissibilities. And to use these time- varying multi-reference transmissibilities for pTOMA. This paper will recapitulate the time-varying version of single reference TOMA. We will introduce the time-varying version of multi-reference TOMA (pTOMA). The newly introduced algorithm will be demonstrated and discussed trough a lab experiment.

© Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics. All rights reserved. We propose a novel phase bound magnetic vibration absorber based on parallel combinations of magnetic springs. The corresponding design methodology is based on magnetic spring toolchain consisting of FE simulations of magnetic spring and 1D models together with system dynamics model. This approach is applied to an industrially relevant vibration reduction problem of gear meshing, specifically for wind turbine applications. In order to capture the relevant gearbox dynamic a semi-analytical Cai-Hayashi formulation of the gear stiffness is used in combination with ISO 6336 norm for spur gear capacity. We compare the modeling results to a state-of-the-art tuned mass damper approach, addressing the issues of tonalities in wind turbines. As opposed to the frequency bound vibration absorbers, the novel magnetic spring vibration absorber is order-selective, similarly to centrifugal pendulum absorbers, with the added value of a virtually infinite lifetime, no de-tuning issues, and feasible high-order vibration absorbers. ispartof: pages:4833-4845 ispartof: Proceedings of ISMA2018 International Conference on Noise and Vibration Engineering pages:4833-4845 ispartof: International Conference on Noise and Vibration Engineering location:Leuven, Belgium date:17 Sep - 19 Sep 2018 status: published

© Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics. All rights reserved. The challenge addressed in this paper is the detection of changed boundary conditions at boundaries, which are either partially or entirely not accessible and possibly also not visible. One possible field in which such a strategy could be applied is nondestructive testing and structural health monitoring. In contrast to conventional methods targeting the same challenge, it is proposed to test the structure in a mid-frequency range, where both the required sensitivity is reached and the desired robustness to geometrical features or changes in the environment is still present. Furthermore it is proposed to exploit a numerical model of the healthy test object in order to predict damage in areas, which are not accessible for manual inspection. Therefore, a two-step algorithm is proposed, where the first step consists of a Time-Reversal Multiple Signal Classification to narrow down the defect location. In the second step, the damage location is further concretised and the defect extent is estimated based on a model updating scheme using a parametrically reduced order model. no ISSN ispartof: pages:473-487 ispartof: Proceedings of the International Conference on Noise and Vibration Engineering (ISMA) / International Conference on Uncertainty in Structural Dynamics (USD) pages:473-487 ispartof: International Conference on Noise and Vibration Engineering (ISMA 2018) location:Leuven, Belgium date:17 Sep - 19 Sep 2018 status: Published online

Vibration signals of machinery typically consist of various types of signal content. Two such types can be classified as stochastic and deterministic. In the past, several methods have been developed to tackle the separation of these two categories. One such solution is the Discrete/Random Separation method which estimates a frequency-domain transfer function from the signal to separate the predictable periodic content from the random content. Two of the downsides of the existing Discrete/Random Separation method in its current form are the variance of the estimated filter and its dependency on the time delay choice. This paper investigates the potential to alleviate these downsides by extending the definition of the filter from a single-delay to a multi-delay filter estimation procedure. This extension improves the reliability of the filter for practical usage and reduces the dependency of the result on the chosen filter delay. The improved DRS filter is examined on both simulated and experimental wind turbine gearbox vibration data.

In this study, it is investigated how the random decrement (RD) technique can be used to linearize the nonlinear stiffness of a single degree of freedom system with a hardening spring, known as a Duffing oscillator. The linearization is implemented by using a set of trigger levels in the response signal to define RD signatures. The modal parameters are estimated at the same response level for various excitation force levels, from the RD signatures describing the system at that specific level, and extracting the modal parameters using time domain parameter extraction methods. The system described by the RD signature is assumed to behave as a linear system with an equivalent stiffness depending on the trigger level. Various triggering conditions for the RD technique are investigated, to study their applicability in linearizing the nonlinear system. The results of the study show that, with the Local Extremum, and Envelope triggering condition, the RD technique can be successfully used to linearize the modal parameters of the Duffing oscillator.

In the present paper a study of test and material uncertainty in modal analysis of certain wooden junctions is presented. The main structure considered here is a T-junction made from a particleboard plate connected to a spruce beam of rectangular cross section. The size of the plate is 1.2 m by 0.6 m. The T-junctions represent cut-outs of actual full size floor assemblies. The aim of the experiments is to investigate the underlying uncertainties of both the test method as well as variation in material and craftmanship. For this purpose, ten nominally identical junctions are tested and compared to each other in terms of modal parameters such as natural frequencies, modeshapes and damping. Considerations regarding the measurement procedure and test setup are discussed. The results indicate a large variation of the response at modes where the coupling of torsion in the beam to bending of the plate is significant.

Recent trend in the field of mechanical engineering designs promote the use of flexible multibody dynamics simulations as a crucial tool, in order to perform the required design analysis, especially when dealing with contact problems. Although parametric model order reduction techniques demonstrated great potential to deal with contact simulations, a true hyper-reduction of the multi-body contact is still not feasible. Moreover, non-smooth discretization used into the canonical Finite Element method leads to convergence issues in the analysis of sliding contact problems (e.g. gears and bearing) which is commonly compensated by usage of highly detailed meshes, whose level of detail is pushed far beyond the convergence requirements. In order to solve this fundamental conflict, this paper proposes to locally enrich the description of the contact boundaries, by means of the Hermite interpolants, in order to evaluate the contact forces on a finer detail. Preliminary results shows that the proposed approach is able to reduce the impact of the above mentioned conflict, allowing to reach the same quality of results with a coarser mesh, and a reduced computational impact ispartof: pages:1347-1356 ispartof: ISMA proceedings vol:2016 pages:1347-1356 ispartof: 2016 Leuven Conference on Noise and Vibration Engineering (ISMA 2016) location:Leuven date:19 Sep - 21 Sep 2016 status: published

This paper starts with the presentation of the shift cell technique, which allows the description of the propagation of all existing waves starting from the unit cell through a quadratic eigenvalue problem. Its major advantage is that it allows the implementation of any frequency dependence and damping in the problem: this is a fundamental advantage when computing the dispersion curves of a porous material modeled as an equivalent fluid. The second part of this work concerns the investigation of the link between the dispersion curves and the acoustic properties of the material. Deriving the equivalent acoustic properties of the unit cell from its dispersion characteristics, indeed, could be a very efficient approach for designing the sound packages with a simple a preliminary eigenvalue analysis. © Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics. All rights reserved.

© Proceedings of ISMA 2018 - International Conference on Noise and Vibration Engineering and USD 2018 - International Conference on Uncertainty in Structural Dynamics. All rights reserved. Classical (independent) interval analysis considers a hyper-cubic input space consisting of independent intervals. This stems from the inability of intervals to model dependence and might result in a serious over-conservatism as no physical guarantee of independence of these parameters exists. In a spatial context, dependence of one model parameter over the model domain is usually modelled using a series expansion over a set of basis functions that interpolate a set of globally defined intervals to local (coupled) uncertainty. However, such approach might prove to be too inflexible to model the dependence between a large number of model parameters. This paper therefore presents a flexible approach for the modelling of dependent intervals that is also applicable to high-dimensional problems. Specifically, it is proposed to construct the dependence structure in a similar approach to copula pair constructions, yielding a limited set of 2-dimensional dependence functions. The applied case studies indicate the flexibility and performance of the method. ispartof: pages:5071-5080 ispartof: Proceedings of the International Conference on Uncertainty in Structural Dynamics pages:5071-5080 ispartof: International Conference on Uncertainty in Structural Dynamics location:Leuven, Belgium date:17 Sep - 19 Sep 2018 status: Published online

Over the past decades models of structural systems have moved from single-query applications in uniform domain towards a design process where multiple queries of a multi-domain model are required for different design parameters. In order to limit the computational burden associated with these multi-query applications, parametric model order reduction (pMOR) becomes an indispensable tool. These techniques allow a low(er) cost evaluation of the original model (e.g. a finite element model) for a range of design parameters. This paper investigates the possibility of applying a greedy algorithm to generate a reduced order basis (ROB) from sampled eigenvectors for structural dynamics applications. The resulting ROB can be used in projection-based pMOR techniques and in the numerical example it is used to reduce an affine representation of a finite element (FE). ispartof: pages:3563-3571 ispartof: Proceedings of the 27th international conference on noise and vibration engineering. pages:3563-3571 ispartof: ISMA 2016 location:Leuven date:19 Sep - 21 Sep 2016 status: published

ispartof: pages:2321-2331 ispartof: PROCEEDINGS OF INTERNATIONAL CONFERENCE ON NOISE AND VIBRATION ENGINEERING (ISMA2020) / INTERNATIONAL CONFERENCE ON UNCERTAINTY IN STRUCTURAL DYNAMICS (USD2020) pages:2321-2331 ispartof: International Conference on Noise and Vibration Engineering (ISMA) / International Conference on Uncertainty in Structural Dynamics (USD) location:BELGIUM, Leuven date:7 Sep - 9 Sep 2020 status: published

ispartof: pages:3563-3577 ispartof: PROCEEDINGS OF INTERNATIONAL CONFERENCE ON NOISE AND VIBRATION ENGINEERING (ISMA2020) / INTERNATIONAL CONFERENCE ON UNCERTAINTY IN STRUCTURAL DYNAMICS (USD2020) pages:3563-3577 ispartof: ISMA2020 location:Leuven, Belgium date:7 Sep - 9 Sep 2020 status: published

In this paper a simplified model of a rotating tire is developed and free vibrations of the tire are analysed on an example case. The tire belt is modelled as a rotating cylindrical shell. The shell ends are assumed to be with free boundary conditions. A flexible foundation is considered under the belt surface in order to take into account the stiffness of the air inside the tire as well as the tire sidewall. Moreover, the tire inflation pressure is accounted for. This is necessary since it generates an important tangential pre-stress in the tire belt. Finally, the effects of rotation onto the resonance frequencies and the mode shapes of the tire are considered. This is done by introducing Coriolis and centrifugal terms into the equations of motion plus the tangential centrifugal pre-stress. The boundary conditions of the elastically supported shell are satisfied exactly and the corresponding mode shapes and the resonance frequencies are calculated for an example case. Unlike the shells with simply supported ends, the shells with free ends have the functional form of the mode shapes changing with frequency. In particular, the dependence of a mode shape on the axial coordinate exhibits strong generic variation with frequency. Consequently, different combinations of transcendental functions (hyperbolic and circular) occur at different frequency ranges. The Coriolis accelerations and the centrifugal tension are shown to play an important role in how the resonance frequencies of the tire veer with rotation speed. In fact, the modes rotating backward and forward along the tire circumference tend to slow down or speed up with the increase in rotation speed. Thus they no longer superimpose into standing waves. In contrast, they keep propagating as travelling modes.

A validation and improvement of a recently developed time domain impedance formulation based on recursive convolution has been carried out. This formulation has been implemented using an in-house time-domain Runge-Kutta Discontinous Galerkin (RKDG) Linearized Euler Equations (LEE) solver. At first, a validation has been carried out on a 2D lined rectangular duct both without flow and with uniform mean flow. The RKDG results have been compared with the analytical reference solutions for the problem under consideration. In a second stage the case of the 2D rectangular lined duct with a non-uniform mean flow is validated to include the effect of a finite boundary layer thickness on the acoustic damping properties. ispartof: pages:1281-1295 ispartof: Proceedings of the 26th Conference on Noise and Vibration Engineering (ISMA2014) vol:26 pages:1281-1295 ispartof: Conference on Noise and Vibration Engineering (ISMA2014) location:Leuven, Belgium date:15 Sep - 17 Sep 2014 status: published

This paper presents an application of a Kalman filtering algorithm for strain estimation in the tower of an offshore monopile wind turbine in the Belgian North Sea. Real measured data obtained from in situ measurements are used in the validation study. The system model used in the strain estimation is constructed based on the mode shapes and natural frequencies obtained from a finite element model of the structure and the damping characteristics that are obtained from a prior output-only operational modal analysis (OMA) of the tower. A recently developed approach for quantification of the estimation uncertainty introduced by sensor noise and unknown excitation acting on the structure is applied. The approach is used to determine the noise statistics for Kalman filtering that minimize the estimation uncertainty. Finally, the quality of the strain estimates is assessed by comparison with the actual, measured strains. Very accurate strain estimates are obtained.

It is commonly accepted that bolted structures can present nonlinearities which can be caused by the contact conditions and that these are severely exercised for large vibration amplitudes. Several researches have targeted these joint's nonlinear behaviours and experimental model validations have demonstrated to be successful. The scope of this paper is to demonstrate that nonlinearities which are quantified in the modal space can be correlated with the same ones which are identified in the physical space by using reverse path identification method. A dumbbell test rig was designed and manufactured so as to have a lap joint connecting the two weights placed at either ends of the bolted beam. The test structure will be modelled by both one degree of freedom lump parameter model and Finite Element model. The nonlinearities will be identified by the frequency and damping nonlinear functions in the modal space and then correlated with the equivalent laws identified by reverse path method in the physical space.