摘要: |
针对运载火箭结构价格昂贵,通过试验获取强度变差系数困难问题,提出了基于产品实测数据仿真预估强度变差系数方法。以轴压载荷作用下的1 m直径薄壁正交网格加筋圆柱壳试验件为对象,开展了实现技术途径及试验验证研究。测量并统计试验产品材料性能、结构尺寸及几何形貌不确定性;应用双重傅立叶级数对实测几何形貌表征,将其引入到结构理想有限元分析模型,实现极限强度精确仿真分析;采用混沌多项式展开(Polynomial Chaos Expansion,PCE)代理模型,实现考虑关键参数不确定性的极限强度高效模拟打靶和强度总体变差系数预估;并开展了3个试验件的轴压强度破坏试验验证。结果表明,预估的强度变差系数可很好包络试验样本变差系数,具有较高可靠性的同时,相对历史统计数据有较大降低,对实现结构可靠性安全系数的精确量化设计,挖掘现代先进结构承载潜力有重要意义。 |
关键词: 薄壁加筋壳 强度变差系数 几何形貌 仿真 |
DOI: |
|
基金项目:民用航天预研项目(305060504) |
|
Prediction of Strength Variation Coefficient of Thin-Walled Stiffened Shell by Simulation |
XU Weixiu,YANG Fan,JIANG Liangliang,WU Huiqiang,HAO Peng |
(Beijing Institute of Astronautical Systems Engineering, Beijing 100076, China;State Key Laboratory of Structural Analysis for Industrial Equipment,Department of Engineering Mechanics,Dalian University of Technology,Dalian 116024, China) |
Abstract: |
In view of the difficulty in obtaining the strength variation coefficient through test due to the high price of the launch vehicle structure, a simulation method of predicting the strength variation coefficient based on the measured product datum is proposed.Taking the test articles of an 1 m diameter orthogrid stiffened cylindrical shell under axial compression load as the object, the technical approaches and test verification research are carried out.The uncertainty of material properties, geometric dimensions and geometric morphology of the test articles were measured and counted. The measured geometric imperfections are characterized by double Fourier series and introduced into the perfect finite element model to realize the accurate simulation of the structural ultimate strength .The polynomial chaos expansion surrogate model is used to realize the efficient simulation of ultimate strength and predict the strength population coefficient of variation considering the uncertainly of key parameters. The axial compression failure tests of three test articles were carried out.The results show that the predicted strength variation coefficient can well envelope the variation coefficient of test samples,has high reliability,and is greatly reduced compared with historical statistical data.The research results are of great significance to realize the accurate quantitative design of structural reliability safety factor and tap the bearing potential of modern advanced structures. |
Key words: Thin-walled stiffened shell Strength variation coefficient Geometric morphology Simulation |