• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Science and Engineering of Composite Materials >Safety factor determining for space trusses by non-linear analysis and artificial neural network method
【24h】

Safety factor determining for space trusses by non-linear analysis and artificial neural network method

机译：非线性分析和人工神经网络方法确定空间桁架的安全系数

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Determining a feasible safety factor for space trusses is an important phase in structural analysis that could have economic benefits. We know there are many kinds of imperfections in structural elements, which include both material and geometric flaws. Predicting factual behavior of structures is very difficult and occasionally impossible. Elements with initial geometric imperfections in space trusses are a common phenomenon, in addition, equivalent initial geometric imperfections can be applied for modeling of residual stresses or eccentric loading effect. The number of members in the space structures is usually high as is the diversity in the kind of initial imperfection. Therefore, there is a high likelihood that models must be analyzed. The structure must be analyzed with non-linear methods, making these approaches time consuming, and potentially uneconomical. In this study, we selected 30 cases for random analysis based on Monte Carlo methods to find the bearing capacity of the space truss. We attained results from the LUSAS program LUSAS Modeller, Version 13, UK program and these were then exported as input data to the Artificial Neural Network (ANN) program. A reasonable neural network has been found of predicting another 30 cases for load bearing capacity without any analysis and only based on the neural network program. Finally, a new approach for determining the load capacity of the space trusses was extracted and we predicted the occurrence possibility of the convenience load bearing capacity in 60 cases.
机译：确定可行的空间桁架安全系数是结构分析中可能具有经济效益的重要阶段。我们知道结构元素中存在多种缺陷，包括材料和几何缺陷。预测结构的实际行为非常困难，有时甚至是不可能的。在空间桁架中具有初始几何缺陷的元素是一种常见现象，此外，可以将等效的初始几何缺陷应用于残余应力或偏心载荷效应的建模。空间结构中的成员数量通常很高，初始缺陷类型的多样性也很高。因此，很有可能必须分析模型。必须使用非线性方法分析结构，这使这些方法既费时，又可能不经济。在这项研究中，我们选择了30个基于蒙特卡洛方法的随机分析案例，以求得空间桁架的承载力。我们从LUSAS程序LUSAS Modeller，Version 13，UK程序中获得了结果，然后将这些结果作为输入数据导出到了人工神经网络（ANN）程序中。已经找到了一个合理的神经网络，无需进行任何分析，仅根据神经网络程序即可预测另外30个承载能力的情况。最后，提出了一种确定空间桁架承载力的新方法，并预测了60例方便承载力的发生可能性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号