基于极值统计理论的屋盖围护结构设计风荷载

发布时间：2018-01-02 11:48

本文关键词：基于极值统计理论的屋盖围护结构设计风荷载　出处：《北京交通大学》2013年硕士论文　论文类型：学位论文

【摘要】：多起大跨屋盖围护结构的风致灾害表明,现有荷载规范不能为大跨屋盖围护结构的设计风荷载提供足够可靠信息,因此需要对其进行更深入的研究。本文基于平屋盖风洞试验数据,分析了屋盖表面的风压分布特性,在划分风压区的基础上使用各种极值估计方法得到平屋盖围护结构的设计风荷载。主要内容如下： (1)介绍了目前使用的大跨屋盖围护结构风压极值的计算方法,重点阐述了区组最大值模型和超阈值模型的基础理论。 (2)分析了平屋盖表面的平均风压、脉动风压、极值风压的分布特性。结果表明,平均风压与脉动风压的分布模式不相同,而极值风压与平均风压的分布模式相似。此外还探讨了风速对风压系数的影响。结果表明,提高风速对平均风压系数的影响很小,对风压系数脉动均方根的影响较大。 (3)探讨了样本量对区组最大值模型参数估计结果的影响。结果表明,样本量对形状参数的估计值影响最大,对位置参数的估计值影响最小,样本量达到100时估计结果已基本趋于稳定。并通过各个风向角下区组最大值模型和超阈值模型的形状参数估计值在屋盖表面的分布确定各个极值类型所占的区域。结果验证了两种极值方法所得的极值类型分布模式十分相似。 (4)使用各种极值估计方法得到标准时距下风压系数极小值的期望及其具有78%保证率的分位数。通过对比分析,表明区组最大值模型和超阈值模型可以得到较精确的极值风压估计,而Gumbel法的估计值普遍偏低。最后在划分风压区的基础上给出了各区的设计极值风压系数。
[Abstract]:Many wind-induced disasters of large-span roof structures show that the existing load codes can not provide enough reliable information for the design wind loads of long-span roof enclosures. Based on the wind tunnel test data of flat roof, the distribution characteristics of wind pressure on roof surface are analyzed in this paper. On the basis of dividing the wind pressure area, the design wind load of the flat roof enclosure structure is obtained by using various extreme value estimation methods. The main contents are as follows: This paper introduces the calculation method of wind pressure extremum of large-span roof enclosure structure, and expounds the basic theory of block maximum model and over-threshold model. The distribution characteristics of mean wind pressure, pulsating wind pressure and extreme wind pressure on the surface of flat roof are analyzed. The results show that the distribution pattern of mean wind pressure is different from that of fluctuating wind pressure. The distribution pattern of extreme wind pressure is similar to that of average wind pressure. In addition, the influence of wind speed on wind pressure coefficient is also discussed. The results show that increasing wind speed has little effect on mean wind pressure coefficient. It has a great influence on the RMS root of wind pressure coefficient pulsation. 3) the effect of sample size on the parameter estimation results of block maximum model is discussed. The results show that the sample size has the greatest influence on the shape parameter estimation, and the least effect on the location parameter estimation value. When the sample size reaches 100, the estimated results have been basically stable, and the distribution of the shape parameters of the maximum value model and the super-threshold model under the wind direction angle on the roof surface is determined by the distribution of each extreme value type. The results show that the distribution patterns of the two extremum types are very similar. 4) the expectation of minimum value of wind pressure coefficient under standard time interval and the quantile with 78% guarantee rate are obtained by using various extreme value estimation methods. It shows that the maximum value model and the over-threshold model can get accurate extreme wind pressure estimation. The estimation of Gumbel method is generally low. Finally, the design extreme wind pressure coefficient of each region is given on the basis of dividing the wind pressure region.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TU352.2

【参考文献】