论文部分内容阅读
为了全面分析轻型可移动折叠网壳结构的风压分布特性,基于BP神经网络的基本原理,运用Matlab软件对野营折叠网壳结构风压分布特性进行了神经网络预测;并结合折叠网壳结构风洞试验,分析了其在不同风速和不同风向角下的风压系数分布规律和表面风压分布特性,将结果与风洞试验进行对比,两者吻合较好。结果表明:将风洞试验技术与神经网络方法相结合,基于有限的风洞试验数据可以预测结构未知点的风压系数;通过对测点进行插值计算可以全面掌握结构的风压分布特征,为结构抗风的精细化分析和设计提供一个有效方法。这为轻型可移动折叠网壳结构安全合理设计和改进建筑气动外形提供了理论依据。
In order to comprehensively analyze the wind pressure distribution characteristics of lightweight movable reticulated shell structure, based on the basic principle of BP neural network, the neural network prediction of the wind pressure distribution characteristics of the folded reticulated shell structure is carried out by using Matlab software. Combined with the folded reticulated structure wind Hole test, the wind pressure coefficient distribution and surface wind pressure distribution under different wind speed and different wind direction angles were analyzed. The results were compared with the wind tunnel test, and the two agree well. The results show that wind tunnel test technique and neural network method are combined to predict the wind pressure coefficient of unknown structure point based on the limited wind tunnel test data. The wind pressure distribution characteristics of the structure can be fully grasped by interpolation calculation, Structural wind resistance refinement analysis and design provides an effective method. This provides a theoretical basis for the safe and reasonable design of lightweight movable folding reticulated shell structure and improving the aerodynamic shape of a building.