Abstract
Artificial Neural Networks (ANN) have the capability to develop functional relationshipsrnbetween input-output patterns obtained from any source. Thus ANN can be conveniently used to developrna generalised relationship from limited and sometimes inconsistent data, and can therefore also be appliedrnto tackle the data obtained from wind tunnel tests on building models with large number of variables. Inrnthis paper ANN model has been developed for predicting wind induced pressures in various zones of arnGable Building from limited test data. The procedure is also extended to a case wherein interference effectsrnon a gable roof building by a similar building are studied. It is found that the Artificial Neural Networkrnmodelling is seen to predict successfully, the pressure coefficients for any roof slope that has not beenrncovered by the experimental study. It is seen that ANN modelling can lead to a reduction of the windrntunnel testing effort for interference studies to almost half.
Key Words
wind pressure coefficients; artificial neural network; interference factors and training data
Address
Naveen Kwatra, Department of Civil Engineering, Thapar Institute of Engineering & Technology, Patiala, IndiarnP. N. Godbole and Prem Krishna, Department of Civil Engineering, University of Roorkee, Roorkee, India
Abstract
This paper reports the numerical calculations of uniform turbulent shear flow around a squarerncylinder. The predictions are obtained by solving the two-dimensional unsteady Navier-Stokes equations inrna finite volume technique. The turbulent fluctuations are simulated by the standard k- e model and one ofrnits variant which takes care of the realizability constraint in order to suppress the excessive generation ofrnturbulence in a stagnation condition. It has been found that the Strouhal number and the mean dragrncoefficient are almost unaffected by the shear parameter but the mean lift coefficient is increased. Thernpresent predictions are compared with available experimental data.
Key Words
vortex shedding; shear flow; k- e variant; square cylinder.
Address
A.K.M. Sadrul Islam, Department of Mechanical Engineering, Bangladesh University of Engineering & Technology, Dhaka-1000, BangladeshrnR.G.M. Hasan, Computational Modelling Section, Health & Safety Laboratory, Buxton SK17 9JN, U.K.
Abstract
Wind flow and pressure on the roof of the Texas Tech Experimental Building are studied alongrnwith the incident wind in an effort to understand the wind-structure interaction and the mechanisms of roofrnpressure generation. Two distinct flow phenomena, cornering vortices and separation bubble, are investigated. Itrnis found for the cornering vortices that the incident wind angle that favors formation of strong vortices isrnbounded in a range of approximately 50 degrees symmetrical about the roof-corner bisector. Peak pressures onrnthe roof corner are produced by wind gusts approaching at wind angles conducive to strong vortex formation. Arnsimple analytical model is established to predict fluctuating pressure coefficients on the leading roof corner fromrnthe knowledge of the mean pressure coefficients and the incident wind. For the separation bubble situation, thernmean structure of the separation bubble is established. The role of incident wind turbulence in pressure-generationrnmechanisms for the two flow phenomena is better understood.
Address
Zhongshan Zhao, Mustang Engineering Inc., 16001 Park Ten Place, Houston, TX 77084, USArnPartha P. Sarkar, Department of Aerospace Engineering and Engineering Mechanics, Iowa State University, Ames, IA 50011-2271, USArnKishor C. Mehta, Wind Engineering Research Center (WERC), Department of Civil Engineering, Texas Tech University, Lubbock, TX 79409-1023, USArnFuqiang Wu, ABS Consulting, 16850 Diana Lane, Houston, TX 77058-2527, USA
Abstract
This paper describes a simple method for evaluating the design wind loads for the structuralrnframes of circular flat roofs with long spans. The dynamic response of several roof models werernnumerically analyzed in the time domain as well as in the frequency domain by using wind pressure datarnobtained from a wind tunnel experiment. The instantaneous displacement and bending moment of the roofrnwere computed, and the maximum load effects were evaluated. The results indicate that the wind-inducedrnoscillation of the roof is generally dominated by the first mode and the gust effect factor approach can bernapplied to the evaluation of the maximum load effects. That is, the design wind load can be representedrnby the time-averaged wind pressure multiplied by the gust effect factor for the first mode. Based on thernexperimental results for the first modal force, an empirical formula for the gust effect factor is provided asrna function of the geometric and structural parameters of the roof and the turbulence intensity of thernapproach flow. The equivalent design pressure coefficients, which reproduce the maximum load effects,rnare also discussed. A simplified model of the pressure coefficient distribution is presented.
Address
Yasushi Uematsu, Department of Architecture and Building Science, Tohoku University, Sendai, JapanrnMotohiko Yamada, New Industry Creation Hatchery Center, Tohoku University, Sendai, Japan
Abstract
Wind tunnel aeroelastic model tests of the Commonwealth Advisory Aeronautical Research Councilrn(CAARC) standard tall building were conducted using a three-degree-of-freedom base hinged aeroelastic(BHA)rnmodel. Experimental investigation into the effects of coupled translational-torsional motion, cross-wind/torsionalrnfrequency ratio and eccentricity between centre of mass and centre of stiffness on the wind-induced responserncharacteristics and wind excitation mechanisms was carried out. The wind tunnel test results highlight thernsignificant effects of coupled translational-torsional motion, and eccentricity between centre of mass andrncentre of stiffness, on both the normalised along-wind and cross-wind acceleration responses for reduced windrnvelocities ranging from 4 to 20. Coupled translational-torsional motion and eccentricity between centre of massrnand centre of stiffness also have significant impacts on the amplitude-dependent effect caused by the vortexrnresonant process, and the transfer of vibrational energy between the along-wind and cross-wind directions.rnThese resulted in either an increase or decrease of each response component, in particular at reduced windrnvelocities close to a critical value of 10. In addition, the contribution of vibrational energy from the torsionalrnmotion to the cross-wind response of the building model can be greatly amplified by the effect of resonancernbetween the vortex shedding frequency and the torsional natural frequency of the building model.
Address
S. Thepmongkorn, Construction Engineering and Technology Division, Mahanakorn University of Technology, ThailandrnK.C.S. Kwok, Department of Civil Engineering, Hong Kong University of Science and Technology, Hong Kong and Department of Civil Engineering, University of Sydney, NSW 2600, Australia
