摘要
对具有随机参数的连续体结构进行了拓扑优化设计,并考虑了随机参数满足正态分布的可靠性约束,获得优化结构的同时保证了结构的可靠度.根据可靠性理论的一次二阶矩方法,利用随机有限元求解了连续体结构的可靠度,将隐式的可靠度约束转化为等价的常规约束.将模式识别技术的K邻近(KNN)方法引入到连续体结构拓扑优化设计领域,对连续体结构进行了拓扑优化设计.根据该设计方法给出了计算示例,结果表明该设计计算方法是可行的,为工程设计提供了理论依据.
The continuum structure with stochastic parameters was designed by topological optimization.The reliability constraint of stochastic parameters under normal distribution was also considered to guarantee the reliability.Based on the first-order second-moment method of the reliability theory,the stochastic finite element method is used to analyze the continuum structural reliability.The implicit reliability constraint is transformed into the equal conventional constraint.The K-nearest neighbor method of the pattern recognition technique is introduced to the design domain of the continuum structure to carry on the topology optimization design.Finally,several calculation examples prove that the design method is feasible,and proves theory basis for engineering design.
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2011年第9期1304-1307,共4页
Journal of Northeastern University(Natural Science)
基金
长江学者和创新团队发展计划项目(IRT0816)
"高档数控机床与基础制造装备"科技重大专项(2010ZX04014-014)
国家自然科学基金资助项目(50875039)
"十一五"国家科技支撑计划项目(2009BAG12A02-A07-2)
关键词
连续体
结构优化
拓扑优化
K邻近
模式识别
可靠度
continuum
structural optimization
topological optimization
K-nearest neighbor
pattern recognition
reliability
