节点文献

大跨屋盖风致响应分析

Wind Induced Response of Long Span Roofs

【作者】 孙逊

【导师】 倪振华;

【作者基本信息】 汕头大学 , 减灾防灾与防护工程, 2006, 硕士

【摘要】 大跨屋盖结构通常具有质量轻、柔性大、阻尼小、自振频率低等特点,且往往比较低矮,处于大气边界层中风速变化大、湍流度高的近地区域,对风荷载十分敏感。在实际工程中,一般采用截断的振型叠加法来解决结构的振动问题。通常把大跨空间结构视为空间桁架体系,在动力有限元方法中选取前几阶或几十阶振型(或称振动模态)进行响应分析。而对于大跨度空间结构,其自振频率分布密集,往往从最低阶算起前面数十个振型都对其响应有贡献。若采用传统的模态叠加法,会造成计算量很大,甚至无法计算的结果。 本文提出以Lanczos向量直接叠加法确定大跨屋盖结构的风致响应。传统的模态叠加法中无法保证精确计算的特征向量一定能够参与动态响应。与传统模态叠加法不同,在生成正交的Lanczos向量算法中只产生在荷载展开式中有较大参与系数的向量,并用于后续的模态叠加法之中。产生第一个Lanczos向量的空间向量来自于屋盖风压场的本征正交分解(POD)。利用同步多点压力扫描技术对一个圆拱顶屋盖进行了风洞试验。利用所提出的方法分析了屋盖风致响应,并与使用快速CQC法改进的传统方法以及采用平均风荷载作为空间向量的Lanczos向量直接叠加法进行了比较,说明了本文方法的有效性。

【Abstract】 Roof structures with light weight, large spans, low-rise, low-frequency and high flexibility are generally in high turbulent regions, then wind loads become important to these structures, In the practicality engineering design, wind-induced response is analyzed mode suspense method. How to choose proper modes is an important job. And using this method would result in large account.A direct superposition of Lanczos vectors to determine the wind-induced response of large roofs is proposed. In contrast to the traditional mode superposition where is no assurance that the exact eigenvectors will participate the dynamic response, in the algorithm generating orthogonal Lanczos vectors only those vectors with large participation factor of the loading expansion are generated and adopted in the subsequence mode superposition. The space vector used to generate the first Lanczos vector or block lanczos is obtained from the proper orthogonal decomposition (POD) of the wind pressure field of roof. Wind tunnel tests of a dome-like roof were carried out by means of the synchronous multi-pressure scanning technique. The wind-induced responses of roof are analyzed with the proposed method, and the results are compared with the fast Complete-Quadratic-Combination method and the superposition of Lanczos vectors with mean wind loads being the space vector, which illustrate the effectiveness of the proposed method.

  • 【网络出版投稿人】 汕头大学
  • 【网络出版年期】2006年 12期
  • 【分类号】TU312.1
  • 【被引频次】1
  • 【下载频次】149
节点文献中: