• Smart Structures and Systems
• /
• v.23 no.5
• /
• pp.507-520
• /
• 2019

In the structural health monitoring field, damage detection has been commonly carried out based on the structural model and the engineering features related to the model. However, the extracted features are often subjected to various errors, which makes the pattern recognition for damage detection still challenging. In this study, an automated damage identification method is presented for hanger cables in a tied-arch bridge using a convolutional neural network (CNN). Raw measurement data for Fourier amplitude spectra (FAS) of acceleration responses are used without a complex data pre-processing for modal identification. A CNN is a kind of deep neural network that typically consists of convolution, pooling, and fully-connected layers. A numerical simulation study was performed for multiple damage detection in the hangers using ambient wind vibration data on the bridge deck. The results show that the current CNN using FAS data performs better under various damage states than the CNN using time-history data and the traditional neural network using FAS. Robustness of the present CNN has been proven under various observational noise levels and wind speeds.

• Computational Structural Engineering
• /
• v.3 no.4
• /
• pp.133-141
• /
• 1990

The nature of fluctuating air pressure inside building was studied by testing a building model in a wind tunnel. The model has a single room and a window opening. Various opening conditions were tested in both laminar uniform wind and turbulent boundary-layer wind. The RMS and the spectra of the fluctuating internal pressure were measured. The test results support a recent theory which predicts the behavior of internal pressure under high wind based on aerodynamic analysis.

• Wind and Structures
• /
• v.30 no.1
• /
• pp.37-54
• /
• 2020

To investigate the motion-induced aeroelastic effects (or aerodynamic feedback effects) on a square cylinder in uniform flow, a series of wind tunnel tests involving the pressure measurement of a rigid model (RM) and simultaneous measurement of the pressure and vibration of an aeroelastic model (AM) have been systematically carried out. More specifically, the aerodynamic feedback effects on the structural responses, on the mean and root-mean-square wind pressures, on the power spectra and coherence functions of wind pressures at selected locations, and on the aerodynamic forces were investigated. The results indicated the vibration in the lock-in range made the shedding vortex more coherent and better organized, and hence presented unfavorable wind-induced effects on the structure. Whereas the vibration in the non-lock-in range generally showed insignificant effects on the flow structures surrounding the square cylinder.

• Wind and Structures
• /
• v.20 no.4
• /
• pp.527-548
• /
• 2015

Aeroelastic wind tunnel experiments were conducted for conventional and tapered super-tall building models to investigate the effect of taper on fundamental aeroelastic behaviors in various incident flows. Three incident flows were simulated: a turbulent boundary-layer flow representing urban area; a low-turbulent flow; and a grid-generated flow. Results were summarized focusing on the effect of taper and the effect of incident flows. The suppression of responses by introducing taper was profound in the low-turbulence flow and boundary-layer flow, but in the grid-generated flow, the response becomes larger than that of the square model when the wind is applied normal to the surface. The effects of taper and incident flows were clearly shown on the normalized responses, power spectra, stability diagrams and probability functions.

• Wind and Structures
• /
• v.8 no.5
• /
• pp.357-372
• /
• 2005

Mean and extreme pressure distributions on a large cantilevered flat roof model are measured in a boundary layer wind tunnel. The largest peak suction values are observed from pressure taps beneath conical "delta-wing type" corner vortices that occur for oblique winds, then the characteristics and causes of the local peak suctions are discussed in detail. Power spectra of fluctuating wind pressures measured from some typical taps located at the roof edges under different wind directions are presented, and coherence functions of fluctuating pressures are also obtained. Based on these results, it is verified that the peak suctions are highly correlated with the conical vortices. Furthermore, according to the characteristics of wind loads on the roof, an aerodynamic solution to minimize the peak suctions by venting the leading edges and the corners of the roof is recommended. The experimental results show that the suggested strategy can effectively control the generation of the conical vortices and make a reduction of 50% in mean pressures and 25% in extreme local pressures at wind sensitive locations on the roof.

• Journal of Korean Association for Spatial Structures
• /
• v.17 no.1
• /
• pp.77-84
• /
• 2017

Vortex-induced vibration and instability vibration of tall buildings are very important fluid-structure interaction phenomenon, and many fundamental questions concerning the influence of body movement on the unsteady aerodynamic force remain unanswered. For tall buildings, there are two experimental methods to investigate the characteristics of unsteady aerodynamic forces, one is forced vibration method and the other is free vibration method. In the present paper, a free vibration method was used to investigate the unsteady aerodynamic force on tall building whose aspect ratio is 9 under boundary layer simulating city area. Wind pressures on surfaces and tip displacements were measured simultaneously, and the characteristics of tip displacements and generalized forces were discussed. It was found that variation of across-wind displacements showed different trend between the case when wind speed increases and wind speed decreases, and the fluctuating generalize forces in across-wind direction of vibrating model are larger than that of static model near the resonant wind speed and approach to the static value. And for higher wind speed range, there were two peaks in across-wind power spectra of generalize forces of vibrating model, which means that two frequency components are predominant in unsteady aerodynamic forces.

• Wind and Structures
• /
• v.23 no.1
• /
• pp.1-18
• /
• 2016

This work experimentally and numerically analyzes the flow configurations and the dynamic wind loads on panels of rectangular L/h 5:1 cross section mounted on a structural frame of rectangular bars of L/h 0.5:1, corresponding to a radar structure. The fluid dynamic interaction between panels and frame wakes imposes dynamic loads on the panels, with particular frequencies and Strouhal numbers, different from those of isolated elements. The numerical scheme is validated by comparison with mean forces and velocity spectra of a panel wake obtained by wind tunnel tests. The flow configuration is analyzed through images of the numerical simulations. For a large number of panels, as in the radar array, their wakes couple in either phase or counter-phase configurations, changing the resultant forces on each panel. Instantaneous normal and tangential force coefficients are reported; their spectra show two distinct peaks, caused by the interaction of the wakes. Finally, a scaled model of a rectangular structure comprised of panels and frame elements is tested in the boundary layer wind tunnel in order to determine the influence of the velocity variation with height and the three-dimensionality of the bulk flow around the structure. Results show that the unsteady aerodynamic loads, being strongly influenced by the vortex shedding of the supporting elements and by the global 3-D geometry of the array, differ considerably on a panel in this array from loads acting on an isolated panel, not only in magnitude, but also in frequency.

• Wind and Structures
• /
• v.2 no.4
• /
• pp.253-266
• /
• 1999

This paper presents the effects of free-stream turbulence on streamwise surface pressure fluctuations on two-dimensional rectangular cylinders. Particular attention is given to possible effects of turbulence integral scale on fluctuation and peak pressures. The mean, standard deviation, peak pressure coefficients, spectra and cross-correlation of fluctuating pressures were measured to investigate the nature of the separation and reattachment phenomenon in turbulent flows over a wide range of turbulence intensity and integral scale.

• Wind and Structures
• /
• v.18 no.4
• /
• pp.457-471
• /
• 2014

A summary of the main results from an international comparative study for the high-frequency base balance is given. Two buildings were specified - a 'basic' and an 'advanced' building. The latter had more complex dynamic response with coupled modes of vibration. The predicted base moments generally showed good agreement amongst the participating groups, but less good agreement was found for the roof accelerations which are dominated by the resonant response, and subject to measurement errors for the generalized force spectra, to varying mode shape correction techniques, and different methods used for combining acceleration components.

• Wind and Structures
• /
• v.23 no.3
• /
• pp.231-251
• /
• 2016

This paper presents the experimental results of the wind tunnel tests for three symmetric, rectangular, tall building models on a typical open terrain considering the torsional motion-induced vibrations. The time histories of the wind pressure on these models under different reduced wind speeds and torsional amplitudes are obtained through the multiple point synchronous scanning pressure technique. Thereafter, the characteristics of both the Root Mean Square (RMS) coefficients and the spectra of the base shear/torque in the along-wind, across-wind, and torsional directions, respectively, are discussed. The results show that the RMS coefficients of the base shear/torque vary in the three directions with both the reduced wind speeds and the torsional vibration amplitudes. The variation of the RMS coefficients in the along-wind direction results mainly from the change of the aerodynamic forces, but sometimes from aeroelastic effects induced by torsional vibration. However, the variations of the RMS coefficients in the across-wind and torsional directions are caused by more equal weights of both the aerodynamic forces and the aeroelastic effects. As such, for the typical tall buildings, the modification of the aerodynamic forces in the along-wind, across-wind, and torsional directions, respectively, and the aeroelastic effects in the across-wind and torsional directions should be considered. It is identified that the torsional vibration amplitudes and the reduced wind speeds are two significant parameters for the aerodynamic forces on the structures in the three directions.