大跨屋盖脉动风压的非高斯峰值因子计算方法  

Peak factor method for non-Gaussian fluctuating pressures on long-span roofs

作　　者：林巍[1] 黄铭枫[2] 楼文娟[2] 

机构地区：[1]浙江大学建筑设计研究院,杭州310028 [2]浙江大学结构工程研究所,杭州310058

出　　处：《建筑结构》2013年第15期83-87,14,共6页Building Structure

基　　金：国家自然科学基金项目(51008275;50578144)

摘　　要：以杭州新火车东站为例进行风洞试验,研究大跨度屋盖结构表面风压的非高斯分布特性及峰值因子计算方法。分别通过基于零穿越率的峰值因子法、改进峰值因子法、Sadek-Simiu法和基于经典极值理论的改进Gumbel法对各风向角下风压数据的峰值因子进行了估计,并在改进峰值因子法的基础上提出了偏度非高斯峰值因子法。对各种方法的适用性及计算结果进行对比分析表明,改进峰值因子法并不适用于峰度小于3的过程。提出的偏度非高斯峰值因子法与Sadek-Simiu法相比,计算结果十分吻合,而且对高斯过程同样适用,无需对屋盖测点进行高斯与非高斯分区。在应用时由于无需编程、计算量小、适用性好,更方便工程设计使用。The non-Gaussian characteristics and peak factor methods of wind pressures on a long-span roof were systematically analyzed by taking wind tunnel test of Hangzhou New Train Station for example. The traditional peak factor method based on a zero-crossing rate, the improved peak factor method, the Sadek-Simiu method and the improved Gumbel method were employed to estimate the peak factors of wind-induced pressures. Then a new skewness-parameter non- Gaussian peak factor method was developed based on the improved peak factor method. The applicability of various methods was discussed and compared to each other. Results show that the improved peak factor method is not applicable for some pressure processes with kurtosis smaller than three. Results of the proposed method agree well with that of the Sadek-Simiu method. Moreover, the proposed method is also applicable for Gaussian processes, without the Gaussian and non-Gaussian partition for the taps on the long-span roof. For no programming, small calculation and well application, it is more convenient for engineering design.

关 键 词：大跨屋盖 非高斯特性 偏度 峰度 峰值因子 偏度非高斯峰值因子法 

分 类 号：TU312.1[建筑科学—结构工程]