【作者】 张明亮；

【导师】 李秋胜；

【作者基本信息】 湖南大学 ， 结构工程， 2013， 博士

【摘要】 随着社会经济的发展和科技的进步，各种造型独特、结构新颖的空间结构大量涌现。大跨屋盖结构因其具有优美的造型和良好的性能，被广泛应用于各种大型公共建筑中，由于这些建筑物大多具有质量轻、柔性大、阻力小等特点，对风荷载敏感，风荷载往往成为其结构设计的主要控制荷载。此外，由于建筑物部分门窗的开敞或突然开孔导致的内压增大对屋盖结构的安全会产生较大的影响。大量风灾调查表明，屋盖结构的风致破坏在很大程度上都是由于内外压的联合作用所引起的。然而目前人们对内压变化的产生机理及评估尚处于研究阶段，在进行结构的抗风设计时，国内外现行荷载规范仅是提出名义封闭或者开敞时的内压系数建议值，因此开展屋盖结构风致内压的研究具有极其重要的学术意义和工程应用价值，也是现代结构风工程领域研究的热点之一。本文通过对复杂体型的鱼形屋盖在台风风场与A类常规风场下进行了风洞对比试验研究，发现迎风屋檐的脉动风压明显大于屋面其它区域。不同湍流度下屋盖结构的风荷载存在较大的差异，屋盖在高湍流度下（台风风场）的风荷载比常规A类风场明显要大。建议在台风多发地区的建筑，特别是体型复杂的重要建筑，对结构风荷载的评估除了按照规范常规要求进行设计外，通过台风风场的风洞试验加以验证很有必要。基于吉林火车站的风洞试验，研究了屋盖的风荷载分布特性，发现较大的吸力均分布在迎风屋檐、屋盖角区和主站楼凸起的天窗附近。由于气流在高度较低处的湍流特性较大，更容易受周边地貌的影响，因而站台雨棚的升力系数受周边建筑的干扰效应较为明显。在屋盖的气流分离区域，因存在较大的负风压，风压的概率分布向负压段延伸，建议对于屋盖不同的紊流强度区域应取不同的峰值因子，在高紊流区域峰值因子可取3.0~4.0，在低紊流区域峰值因子可取2.5~3.0。对比昆明南站风洞试验结果与荷载规范规定值，发现当立面入口封闭时主站房屋面大部分区域的体型系数与荷载规范规定值大小相当，但在悬挑屋檐及屋面棱角处，体型系数明显大于规范规定值；当考虑有立面入口门洞时，屋面风载体型系数均大于规范规定的名义封闭下的体型系数。在主站房与站台雨棚边缘区域，脉动风压概率密度分布函数体现出较为明显的非高斯分布特性，在尾部负压区域更为显著。对有双面幕墙开孔结构的内压传递方程进行了推导，并从理论上阐述了背立面开孔时对内压附加阻尼效应的影响。分析了孔口等效阻尼比eq随幕墙开孔面积、建筑内部容积、参考风速及孔口处风压高度变化系数等参数的变化趋势。同时对开孔结构发生Helmholtz共振与立面幕墙开孔率的关系进行了探讨。通过刚性模型风洞试验对不同建筑内部容积、幕墙开孔面积、开孔位置及屋盖矢跨比的大跨屋盖结构的风荷载进行了系统的分析研究，并探讨了这些参数对屋盖等围护结构风荷载特性的影响。通过对内压测点的脉动风压功率谱分析，发现开孔结构内部风压谱不仅包含了大气湍流中的能量成分，同时也包含了由于孔口阻尼特性引起的特征湍流的成分。对考虑有多面幕墙开孔的内压理论估算公式进行了推导，并将试验值与理论估算值进行了对比。通过对屋盖结构的风致响应分析发现对于本身刚度较大的屋盖结构，风致响应的大小主要取决于屋盖的自振频率，而幕墙开孔率大小的变化对风致响应的影响并不是很明显。将传统的风振系数计算方法与采用目标概率法得到的位移风振系数进行对比发现对于脉动风影响较为显著的屋盖区域，利用传统的风振系数计算方法可能会过于低估脉动响应的影响，对结构设计是偏于不安全的。本文以风洞试验为主，结合理论分析详细研究了屋盖结构的风荷载特性和考虑幕墙开孔时建筑内部风效应及屋盖风致响应。研究成果可为屋盖结构的抗风设计及荷载规范相关条文的修订提供参考依据。更多还原

【Abstract】 With the development of economy and progress of science and technology, lots ofbeautiful and innovative spatial structures have been built. Such long-span roofstructures are widely used in various large public buildings due to their exquisiteshapes and favorable performance. However, these structures, with the characteristicsof light mass, high flexibility and slight damping, are sensitive to wind loads, and thewind loads generally govern the design of these structures. In addition, suddenopening of doors and windows would result in an increase of internal pressure, whichwould be a major threaten to the roof structures. According to wind disaster surveys,the wind damages to roof structures, in a large degree, are caused by the combinedresults of internal and external pressures. At present, the mechanism of generating theinternal pressure changes is still not fully clear, and only the internal pressurecoefficients in nominal sealing or opening condition are provided in current domesticand international load codes for the wind-resistant design. Hence, further research onthe effect of wind-induced internal pressures on the wind loads on roof structures isrequired, which is also a hotspot in wind engineering.Based on the investigation of wind tunnel tests for fish-shaped roof structures interrain category A specified in the Chinese load code and typhoon wind field, it wasfound that the fluctuating pressure coefficients on the eaves and cantilevered roofswere larger than those on other areas of the roof. Obvious wind load differences forthe roof structures existed under different turbulence intensity conditions, the windloads under the typhoon wind field were greater than those in terrain category A.Besides the conventional norm for the assessment of design wind loads, it wassuggested that wind tunnel test in typhoon wind field was necessary when designingstructures are located in typhoon-prone areas, especially those with complicatedshapes.Based on the wind tunnel test of rigid model of Jilin Railway Station, the wind loaddistribution characteristics were presented and discussed. It was found that negativepressures (suctions) occurred on the eaves, cantilevered roof and the bulge part of theroof on the main station building. The turbulence characteristics of flow in lower partwere more obvious and the flow was easily affected by the surrounding topography.Therefore, the interference effect of the surrounding buildings on the lift coefficients of the platform awning of the station was more significant. Since stronger negativewind pressures were observed in the burbling zone of the roof, and the probabilitydistribution of wind pressure stretched away in the negative pressure area, so it wassuggested that different peak factors should be taken in the different areas of the roofwith different turbulence intensity levels, such as3.0-4.0would be appropriate for theintensive turbulence intensity area while2.5-3.0for the lower one.On comparison of the data of the wind tunnel test for the Kunming Railway (South)Station and those specified by the local load code, it was discovered that the shapecoefficients of the most areas of the main station building matched those specified bythe code when the fa ade entrance was enclosed, while the coefficients on the eavesand the cantilevered roof were distinctly greater than the specified ones. The shapecoefficients on the roof when there were openings were larger than those specified bythe load code with enclosed condition. The probability density function of fluctuatingwind pressure in the marginal area of the main station buildings and the platformawnings of the station shows obvious characteristics of Non-Gaussian, especially inits negative tail.The transfer equation of the internal pressure of roof structure with wall-openingconditions was deduced, and the additive damping effect of the internal pressure bythe opening in the leeward wall was also presented and discussed. Moreover, severalrelated factors were investigated, such as the variation tendency of the orificedamping with different opening areas, internal volume of structure, wind speed andwind pressure height coefficient of the opening. The relations between the openingrate of facade wall and the Helmholtz resonance of the opening structure were alsodiscussed.The assessment of internal and external pressure on long-span roof structuresinvolves several factors such as internal volume, opening area of wall, openinglocations and height-to-span were investigated by the wind tunnel tests of rigidmodels. It was observed from the Power Spectral Density (PSD) of internalfluctuating wind pressure that the spectra included not only the energy component ofatmospheric turbulence, but also the turbulence caused by the opening features. Thetheoretical estimation formula of the internal pressure under multiple wall openingsconditions was deduced, and a comparison between the experimental data and thetheoretical estimated values was also made.The analysis of wind-induced responses of long-span roofs indicated that thenatural frequencies of the roof structures play a significant role on the wind-induced responses, but little influence by the opening ratio. The comparison between the gustresponse coefficients calculated from traditional method and displacement gust factorcalculated by objective-probability method indicated that the traditional methodwould underestimate the response caused by fluctuating pressure where influencedmore by fluctuating wind, and it is not safe enough for the structural design.In this study, the combination of wind tunnel tests and theoretical analysis wereadopted to investigate the wind effects and wind-induced internal pressures of roofstructures with wall-openings. The outputs of this study are expected to providevaluable information and reference for the wind-resistant design of roof structures andthe revision of wind load codes in the future.更多还原