Font Size: a A A

Investigation On Wind Load Characteristics And Wind-resistant Aerodynamic Measures Of Large Span Roof

Posted on:2003-03-30Degree:MasterType:Thesis
Country:ChinaCandidate:J Y FuFull Text:PDF
GTID:2132360062486524Subject:Structural engineering
Abstract/Summary:PDF Full Text Request
With the development of architectural technique and its construction, a number of roof structures with new types, large spans, low-rise, low-frequency and high flexibility have been emerged. These roof structures are generally in high turbulent regions, then wind loads become important to these structures. However, wind-induced response for most of roof structures has not been considered in the present load code of building structures because of the complexity of the structures and wind loads. So the investigation for characteristics of wind load of large span roofing becomes a critical subject.Wind tunnel investigation of large span flat roof was carried out. Mean local shape factor and extreme local shape factor distributed on the roof of the structure with two exposures are measured. Effects of terrain exposures on the mean local shape factor and extreme local shape factor are discussed. The fluctuating Omnidirectional wind pressure power spectrum density (PSD) of some typical locations and the frequency domain coherence property of the fluctuating wind pressure between different tapping positions at different wind directions were measured and discussed.In respect of method to analyze wind load characteristics, focusing the complex factors that influence the wind load distributions of large span roof and with the wind tunnel test of the Shenzhen Convention and Exhibition Center. A new fuzzy neural network model (FNNM) is applied to forecast the wind load distribution of large span roof. The result given by this paper shows that the FNNM is easy to synthesize different factors that are difficult to deal with a normal numeric algorithm. The results show that the well-trained FNNM has a good behavior to forecast the wind load distribution of the large span roof. The successful application of the FNNM to forecast the wind load distribution of roof structures supplies a new method for dealing with this problem.And finally, according the wind load distribution characteristics of the roof, an approach was presented to reduce the wind load on the roof's wind load sensitive location.
Keywords/Search Tags:long span roof, shape factor, wind pressures distribution, wind loads, power spectrum density, coherence function, wind-resistant measures, fuzzy neural network, forecasting
PDF Full Text Request
Related items