阎军

个人信息Personal Information

教授

博士生导师

硕士生导师

主要任职:力学与航空航天学院院长、党委副书记

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. 固体力学. 航空航天力学与工程. 船舶与海洋结构物设计制造

办公地点:工程力学系系楼305房间

联系方式:0411-84706832

电子邮箱:yanjun@dlut.edu.cn

扫描关注

论文成果

当前位置: 中文主页 >> 科学研究 >> 论文成果

Integrated optimization of the material and structure of composites based on the Heaviside penalization of discrete material model

点击次数:

论文类型:期刊论文

发表时间:2015-03-01

发表刊物:STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION

收录刊物:SCIE、EI、Scopus

卷号:51

期号:3

页面范围:721-732

ISSN号:1615-147X

关键字:Composite material; Discrete material optimization; Heaviside Penalization; Integrated optimization of material and structure

摘要:Based on discrete material optimization and topology optimization technologies, this paper discusses the problem of integrated optimization design of the material and structure of fiber-reinforced composites by considering the characteristics of the discrete variable of fiber ply angle because of the manufacture requirements. An optimization model based on the minimum structural compliance with a specified composite volume constraint is established. The ply angle and the distribution of the composite material are introduced as independent variables in two geometric scales (material and structural scales). The void material is added into the optional discrete material set to realize the topology change of the structure. This paper proposes an improved HPDMO (Heaviside Penalization of Discrete Material Optimization) model to obtain a better convergent result, and an explicit sensitivity analysis is performed. The effects of the HPDMO model on the convergence rate of the optimization results, the objective function value and the iteration history are studied and compared with those from the classical Discrete Material Optimization model and the Continuous Discrete Material Optimization model in this paper. Numerical examples in this paper show that the HPDMO model can effectively achieve the integrated optimization of the fiber ply angle and its distribution in the structural domain, and can also considerably improve the convergence rate of the optimal results compared with other DMO models. This model will help to reduce the manufacture cost of the optimal design.