01-08-2012, 02:48 PM
Wind loading and its effects on single-layer reticulated cylindrical shells
1effects on single-layer.pdf (Size: 1.69 MB / Downloads: 44)
Abstract
In this paper, the wind load distribution as well as its effects on single-layer reticulated cylindrical
shells, which are widely used in practice, is investigated. At first, wind pressure distributions on three
rigid cylindrical shell models considering different shape ratios are measured simultaneously in a
wind tunnel. Some special characteristics of the measured wind pressure, especially its fluctuating
component, are discussed. Then, considering the special structural mechanical behaviors, the effects
of wind load, especially the fluctuating component, on the limit load-carrying capacity and the
stability of the shells are investigated by both equivalent static analysis and dynamic analysis.
Suitable methods to get a reasonable estimation of wind load effects in estimating the equivalent
static wind load distribution for such shells are investigated. Finally, with comparison analyses using
different methods mentioned in this paper.
Introduction
Reticulated shell structures are a kind of space-latticed system with the features of bar
system structures and thin shells. They have been widely used due to attractive
architectural shapes and reasonable mechanical behaviors [1]. Single-layer reticulated
shells are economical and can be conveniently constructed, and are a suitable choice for
middle or large span roofs. For such a system, stability is an important, even dominant
problem, and geometrically nonlinear behaviors should be considered in structural
analysis. In addition, such a structural system is very sensitive to the initial imperfections,
such as the initial geometrical imperfection (e.g., the lack of fitting), the initial residual
stress distribution resulting from welding and/or forced fitting, etc. On the other hand, the
difference between actual external loads that the structures will be subjected to at use stage
and the estimated values at design stage, which can be taken as ‘‘load imperfections’’, also
has important effects on the stability of shells.
Wind tunnel tests on rigid cylindrical shell models
Wind tunnel and test conditions
In order to determine the characteristics of wind load distribution on cylindrical shells,
wind tunnel tests on scaled models of cylindrical shells considering different length to span
aspect ratios, L=S ¼ 1:0, 2.0 and 3.0, respectively, were conducted in the boundary layer
wind tunnel (BLWT) of the Wind Engineering Research Center, Tokyo Polytechnic
University (WERC, TPU). With the spire-roughness technique, three types of wind
profiles, Terrain type II, III and IV, were simulated successfully, the same as in Ref. [4].
The size of three cylindrical shell models and the distribution of the measurement taps on
their surface are shown in Fig. 1. For the model with the aspect ratio equal to 3.0, because
of the limited channels of the pressure data acquisition system for simultaneous sampling,
one complementary part without pressure tap was used, as shown in Fig. 1(d). As usual,
wind speed and corresponding velocity pressure at the height of the apex of the models are
taken as the reference wind speed and reference pressure in analyzing the wind pressure
coefficients. Fig.
Summary
In equivalent static wind-resistant analysis for single-layer reticulated cylindrical shells,
estimating the equivalent static wind load distribution should consider the effects of wind
load distribution on structural stability. This paper focuses on the effects of fluctuating
component of wind load on the stability of single-layer reticulated cylindrical shells. Based
on above work, the following results were obtained:
(1) In comparison with the results observed for the spherical model, the first mode
obtained from POD for the cylindrical model gives a smaller contribution to the whole
distribution of wind pressure due to sharp edges at both ends in the longitudinal
direction. Especially when the attack angle is equal to 901, the contribution of the first
mode, as well as the cumulated contribution of the first several modes tends to be
significantly smaller than for the attack angle equal to 01 or 451. For an expected
cumulated contribution near to 90%, more than 20 modes should be considered.
Especially when the attack angle is equal to 901, more than 50 modes should be
considered.