• 更多栏目

    王博

    • 教授     博士生导师   硕士生导师
    • 主要任职:党委常委、副校长
    • 其他任职:工业装备结构分析国家重点实验室副主任
    • 性别:男
    • 毕业院校:大连理工大学
    • 学位:博士
    • 所在单位:力学与航空航天学院
    • 学科:工程力学. 计算力学
    • 办公地点:工程力学系系楼304房间
    • 联系方式:办公电话: 0411-84706608; 手机: 壹叁玖肆贰捌伍玖捌伍伍
    • 电子邮箱:wangbo@dlut.edu.cn

    访问量:

    开通时间:..

    最后更新时间:..

    Extended multiscale finite element method for large deflection analysis of thin-walled composite structures with complicated microstructure characteristics

    点击次数:

    论文类型:期刊论文

    发表时间:2018-09-01

    发表刊物:THIN-WALLED STRUCTURES

    收录刊物:SCIE

    卷号:130

    页面范围:273-285

    ISSN号:0263-8231

    关键字:Extended multiscale finite element method; Thin-walled composite structures; Large deflection analysis; Multiscale base functions; Displacement boundary conditions; Microstructure characteristics

    摘要:An efficient multiscale finite element method is developed for large deflection analysis of thin-walled composite structures with complicated microstructure characteristics. The multiscale base functions are reconstructed to consider the coupling effects of thin-walled composite structures by introducing some additional coupling terms among translations and rotations. For the construction of multiscale base functions, two kinds of displacement boundary conditions are proposed for in-plane and out-plane degrees of freedom. Moreover, two kinds of relaxed decoupled displacement boundary conditions are constructed by adopting the oversampling technique to further improve the accuracy of the method. Then, the equivalent incremental/iterative equilibrium equations for each load step can be constructed and solved directly on the macro scale which will improve the computing efficiency significantly. The microscopic results can be obtained by downscale computation in which the incremental/ iterative equilibrium equations on the micro scale are solved under the incremental boundary conditions updated by incremental macroscopic displacements. Several numerical examples demonstrate that the developed method possesses high computing accuracy and efficiency compared with the conventional finite element method.