中文摘要：

板片空间结构是一种轻质、高强的新型空间结构体系，其整体稳定性是研究、设计过程中的一个关键问题。板片空间结构属于缺陷敏感结构，缺陷稳定分析也是迫切需要而尚未完全得以解决的问题。设计两个板片结构试验模型进行稳定性试验。采用自动协调加载控制系统和电测相结合的方法，连续跟踪记录结构失稳全过程的荷载位移曲线以及应变变化隋况。并以结构实测缺陷为依据，分别采用实测缺陷法、一致缺陷模态法和改进随机缺陷法考虑节点位置偏差缺陷的影响。结果表明，自动协调加载控制系统可保证对失稳全过程的跟踪，连续数据采集方式采样率高而且数据测量精度高。实测缺陷法可比较精确地反映板片空间结构实际的稳定性能。一致缺陷模态法分析的临界荷载并不是所有缺陷结构中的最小值，其概率可靠度随着模型的变化而变化。通过改进随机缺陷法，对板片空间结构的稳定性能有了比较全面的认识，也定量分析了设计临界荷载和一致缺陷模态法分析的临界荷载的可靠性。

英文摘要：

Sheet space structure is a new type of light weight and high strength structure system. The stability of such structure is a key problem in the research and design. Since the sheet space structure is an imperfection-sensitive structure, the stability analysis and nonlinear analysis of the structure with random imperfections are urgently needed. Two models were designed, and destructive loading tests were conducted to study the stability of sheet space structure. With a coordinate automatic control loading system and electrical measurement, the load-displacement curves and load-strain curves were obtained. Three methods, including measured imperfection method, consistent mode imperfection method and advanced stochastic imperfection method, were used to consider the influence of the initial imperfections on the critical load of the structures. The results show that the coordinate automatic control loading system can ensure the exact trace of the whole buckling process, and that high precision and sampling frequency can be achieved through the successive data acquisition technology. It is disclosed that the results of the dual nonlinear analysis by using the measured imperfection method can reflect the stability behavior of the sheet space structure. The critical load by using consistent mode imperfection method is not the minimum of all the imperfect structures and the reliability of the critical loads with consistent mode imperfections is variable for different models. The reliability of both the critical load by using the consistent mode imperfection method and the critical design load can be evaluated accurately with the use of advanced stochastic imperfection method.