A study on the effects of turbulence scale on flutter of long span bridges

大跨度桥梁和高层建筑等工程结构暴露在高度湍流的风中。由于大跨度桥梁对颤振的敏感性，G.驰振、扭振、涡激等，紊流对颤振的影响是一个长期存在的问题。湍流积分尺度L_x与物体尺寸h之比L_x/h是识别湍流效应的一个尺度。本文研究了矩形棱柱体的驰振问题。本文的主要目的是在较宽的L_x/h范围内全面了解湍流尺度对驰振的影响。为了获得较大的L_x/h值，不仅采用了二维矩形棱柱，而且还采用了一端固定在墙上的三维棱柱（三维模型）。湍流流场由不同的网格生成，强度为u '/U约10%，其中u'为r。M. S.二维模型：研究了矩形截面上的驰振现象 ...更多信息 d/h为0.6-3.0的较大棱镜，其中d为棱镜的深度，h为棱镜的高度。(a)d/h=0.6-1.0：均匀流和湍流都发生了驰振，β a（振荡的空气动力学增长率）略为正值，其中尺度效应不显著。抖振引起的振动很弱。(b)d/h=2.0：对于湍流流动，由于分离剪切层在侧面的再附着，驰振消失，而对于均匀流动，驰振发生。随着L_x/h的增大，湍流的β a逐渐趋近于光滑流的β a。这清楚地表明，驰振取决于湍流尺度。(c)d/h=3.0：均匀流和湍流的驰振均消失，因此β a为负值。随着L_x/h的增加，β a减小，并趋于均匀流的β a。三维模型：研究了d/h为2.0和3.0的矩形棱柱体上的驰振。随着L_x/h的增大，湍流的β a值逐渐趋近于光滑流的β a值，与二维模型相似。抖振：在大L_x/h的湍流中，d/h=2.0和3.0的模型除发生驰振外，还发生抖振。当L_x/h为2.0-3.0时，抖振振幅迅速增大，但当L_x/h超过10.0时，抖振振幅接近一个上限。少

Engineering structures such as long span bridges and tall buildings are exposed to highly turbulent wind. Since long span bridges are susceptible to flutter, e. g. galloping, torsional flutter, vortex excitation, etc., the effects of turbulence on flutter have been a longstanding problem. The ratio, L_x/h, of turbulence integral scale L_x to the body size h is one of the measures which identify the turbulence effect. In the present research, galloping of rectangular prisms is considered. The principal purpose is to comprehensively understand the effects of turbulence scale on galloping over a wide range of L_x/h. Not only 2-D rectangular prisms but also 3-D prisms with one end fixed to the wall (3-D model) were used in order to attain larger values of L_x/h. Turbulent flows were generated by different grids with intensity u'/U about 10%, where u'is the r. m. s. of the u-component of velocity fluctuation and U is the free stream velocity.2-D model : Galloping is investigated on rectangu … More lar prisms with d/h of 0.6-3.0, where d is the depth and h is the height of the prism. (a) d/h=0.6-1.0 : Galloping occurs for both the uniform and turbulent flows with a slightly positive betaa, the aerodynamic growth rate of oscillation, where the scale effect is not significant. The oscillation due to buffeting is very weak. (b) d/h=2.0 : Galloping vanishes for the turbulent flows because of the promotion of reattachment of the separated-shear-layer to a side face, whereas galloping occurs for the uniform flow. With increasing L_x/h, betaa for the turbulent flows asymptotes to that for the smooth flow. This clearly shows that galloping is dependent on turbulence scale. (c) d/h=3.0 : Galloping vanishes for both the uniform and turbulent flows so that betaa negative. With increasing L_x/h, betaa decreases and tends to that for the uniform flow.3-D model : Galloping is investigated on rectangular prisms with d/h of 2.0 and 3.0. With increasing L_x/h, betaa for turbulent flows asymptotes to that for the smooth flow in a similar way to the 2-D model.Buffeting : In addition to galloping, models with d/h=2.0 and 3.0 at high reduced flow velocities are suffered from buffeting in turbulent flows with large Lx/h. The amplitude of buffeting increases rapidly at L_x/h of 2.0-3.0, but beyond around 10.0, the amplitude approaches an upper hmit. Less