No abstract is provided for this article.
No abstract is provided for this article.
No abstract is provided for this article.
In structural engineering, model updating is often used for non-destructive damage assessment: by calibrating stiffness parameters of finite element models based on experimentally obtained (modal) data, structural damage can be identified, quantified and located. However, the model updating problem is an inverse problem prone to ill-posedness and ill-conditioning. This means the problem is extremely sensitive to small errors, which may potentially detract from the method׳s robustness and reliability. As many errors or uncertainties are present in model updating, both regarding the measurements as well as the employed numerical model, it is important to take these uncertainties suitably into account. This paper aims to provide an overview of the available approaches to this end, where two methods are treated in detail: a non-probabilistic fuzzy approach and a probabilistic Bayesian approach. These methods are both elaborated for the specific case of vibration-based finite element model updating for damage assessment purposes.
Vibration testing is a well-known practice for damage identification of civil engineering structures. The real modal parameters of a structure can be determined from the data obtained by tests using system identification methods. By comparing these measured modal parameters with the modal parameters of a numerical model of the same structure in undamaged condition, damage detection, localization, and quantification is possible. This paper presents a real-life application of this technique to assess the structural health of the 50-year old bridge of Tilff, a prestressed three-cell box-girder concrete bridge with variable height. A complete ambient vibration survey comprising both vertical accelerations and axial strains has been carried out. The in situ use of optical fiber strain sensors for the direct measurement of modal strains is an original contribution of this work. It is a big step forward in the exploration of modal curvatures for damage identification because the accuracy in calculating the modal curvatures is substantially improved by directly measuring modal strains rather than deriving the modal curvatures from acceleration measurements. From the ambient vibrations, natural frequencies, damping factors, modal displacements and modal curvatures are extracted by the stochastic subspace identification method. These modal param- eters are used for damage identification which is performed by the updating of a finite element model of the intact structure. The obtained results are then compared to the inspections performed on the bridge.
Vibro-acoustic analysis of complex systems at higher frequencies faces two challenges: how to compute the response without using an excessive number of degrees of freedom (DOFs), and how to quantify the uncertainty of the response due to small spatial variations in geometry, material properties, and boundary conditions, which have a wave scattering effect? In this study, a general method of analysis is presented that provides an answer to both questions while overcoming most limitations of statistical energy analysis. The fundamental idea is to numerically compute an artificial ensemble of realizations for the components of the built-up system that are highly sensitive to small random wave scatterers. This can be efficiently performed because their eigenvalue spacings and mode shapes conform to Gaussian orthogonal ensemble spacings and Gaussian random fields, respectively. The DOFs of the overall system are therefore limited to those of the deterministic components and the interface DOFs of the random components. The method is extensively validated by application to plate structures. Good agreement between the predicted response probability distributions and the results of detailed parametric probabilistic models is obtained, also for cases of low modal overlap, single point loading, and strong subsystem coupling.
No abstract is provided for this article.
The substructuring technique has been employed in structural analysis for 25 years. Until now, the use of this technique in nonlinear analysis has been limited to localized situations. Moreover, the benefit in calculation time one can obtain is restricted to the amount of 50%. However, as will be described in this paper, a new scheme of multi-level substructuring technique in nonlinear analysis has been developed. With this scheme an efficiency of 75–85% has been reached and a connection level higher than three can easily be expanded. The Newton-Raphson method and its modified version are the most well known algorithms in nonlinear numerical analysis. The former has, nevertheless, the severe disadvantage of reforming the stiffness matrix at every iteration and, in fact, some of the stages are not necessary. The latter has also the disadvantage of a slow convergence, as the recalculation of the stiffness matrix is executed in certain fixed iterations. Thus, to overcome the difficulties mentioned above, an experimental self-adaptive Newton-Raphson algorithm, which offers an automatic decision as to whether the stiffness matrix is reformed or not, is introduced. In applying this algorithm a saving of 30–50% in execution time can be achieved. Combining these two new schemes in 2-D nonlinear analysis, a dramatic effort of 85–90% is yielded. A typical flow chart and two numerical examples are also presented.
In this paper, Srneulders' modification of Biot's saturated poroelastic theory is used to account for the presence of a small amount of air in the pores of an unsaturated medium. The influence of the gas phase on the compressibility of the generalized pore fluid is accounted for by a complex effective bulk modulus of the gas phase. This modification is introduced in a spectral element formulation, enabling to model transient wave propagation in layered dry, saturated arid unsaturated poroelastic media. The extended application area of the method is illustrated by a numerical example.
Glass fibre-reinforced pipes are used in many applications, especially where corrosion resistance and the light weight have to be considered. This paper outlines a fracture mechanics approach for optimizing the geometry of an adhesive joint for glass fibre-reinforced pipes. The main tasks of this project are to optimize the coupler properties and the joint geometry. Cracks can originate at the coupler edges and at the interface between the two pipes. Fracture mechanics parameters are extracted from a finite element model and are used to optimize the joint length, coupler thickness, taper slope, coupler stacking sequence, and material properties.
Recently a new technique for Operational Modal Analysis was proposed and validated. This technique makes use of transmissibility measurements only. In general, the poles that are identified from transmissibility measurements do not correspond with the system’s poles. However, by combining transmissibility measurements under different loading conditions, it has been shown that the model parameters still can be identified. For comparison and validation, this new method and an existing output-only technique using the power spectra will be applied to operational data of a bridge. The advantage of the recently proposed technique is that the operational forces are no longer assumed to be white noise. They can be arbitrary (colored noise, swept sine, impact ...) as long as they are persistently exciting in the frequency band of interest.
Stochastic subspace identification has become an industrial standard for operational modal analysis because of its computational efficiency and statistical optimality. For the time-domain version of the algorithm, a computationally efficient method exists for the estimation of (co)variances of the identified system matrices and the related modal characteristics. In the present paper, a computationally efficient uncertainty quantification method is developed for a frequency-domain subspace algorithm that starts from nonparametric positive power spectral density estimates. A connection with the time-domain method is made, and the performance is verified against Monte Carlo simulations in a numerical experiment.
