Optimum design of composite laminates

A new procedure for the optimum layup design of composite laminates is described in this paper. In this method the global optimum search and local relaxing constraints are adequately combined together. The design variables are divided and applied in separate processes. A 16-node hybrid semi-Loof element model is used in the global structure analysis. The iterative complex method is adopted in the optimization. The method is suitable for the minimum weight design of a laminate subjected to the strength and stiffness constraints under multiple loading. The numerical results demonstrate the high efficiency and reliability of the present method.The project was supported by National Natural Science Foundation of China     

复合材料加筋壁板的优化设计

本文用简化的方法,考虑了受压及压剪联合作用的复合材料加筋壁板的总体屈曲和局部屈曲、材料强度、刚度、轴向应变等因素以及其它工程要求的影响,用数学规划法提出壁板的优化设计问题,用乘子罚函数法求解。以石墨-环氧帽型加筋板为例,所得优化设计结果与NASA得出的最优曲线一致,证明了所用方法的可靠性,并可进一步推广应用于工程实践。     

应力约束全局化策略下的连续体结构拓扑优化

利用Mises强度理论,提出了应力约束全局化策略,将局 部的应力约束问题转化为结构整体的应变能约束问题. 基于ICM(独立、连续、映射)方法, 引入了独立、连续的拓扑变量,对单元重量、单元刚度和单元许用应力的过滤函数进行了选 择,建立了以重量为目标,以结构应变能代替应力约束的多工况下连续体结构拓扑优化模型, 寻找到了多工况下的最佳传力路径. 运用对偶二次规划方法对上述优化模型进行了求解. 另 外,利用PCL语言,在MSC/PATRAN的开发平台上,实现了应用应力约束全局化策略进行连 续体结构拓扑优化的模块化处理. 数值算例表明了该方法的可行性和有效性.     

用ICM法拓扑优化静位移及频率约束下连续体结构

用ICM方法建立了静位移及频率约束下、重量最小为目标的连续体结构拓扑优化模型。采用独立于截面及形状参数的连续拓扑变量,借助于过滤函数,位移约束用莫尔定理显式化,频率约束用瑞利商求导数借助模态动能及模态应变能近似显式化。用图形过滤处理的方法解决了棋盘格及网格依赖问题。通过构造适当的过滤函数有效地防止了局部模态问题。动态引入防止模态交换的频率约束条件,使迭代过程不发生振荡。算例表明:用ICM方法建立的模型在处理多工况静位移约束、多频率约束及解决局部模态及模态交换等问题上有优势。     

SIMULTANEOUS SHAPE AND TOPOLOGY OPTIMIZATION OF TRUSS UNDER LOCAL AND GLOBAL STABILITY CONSTRAINTS

A new approach for the solution of truss shape and topology optimization problem sunder local and global stability constraints is proposed. By employing the cross sectional areas of each bar and some shape parameters as topology design variables, the difficulty arising from the jumping of buckling length phenomenon can be easily overcome without the necessity of introducing the overlapping bars into the initial ground structure. Therefore computational efforts can be saved for the solution of this kind of problem. By modifying the elements of the stiffness matrix using Sigmoid function, the continuity of the objective and constraint functions with respect to shape design parameters can be restored to some extent. Some numerical examples demonstrate the effectiveness of the proposed method.     

基于空间连续尺寸场的动态均匀化方法

摘 要：提出了基于空间连续尺寸场的动态均匀化方法。方法以均匀化理论为基础,将各向同性蜂窝状微孔结构作为材料描述方式,合理的物理描述自然克服了动态优化中的局部模态现象。不同于将微孔结构尺寸变量依附于单元或节点,采用物质点对应的微孔尺寸为设计变量,基于修正过滤公式的形函数,构造了具有空间连续性的尺寸场,从而克服了棋盘格等数值不稳定性问题。基于复合函数求导法则,推导了总刚度阵、总质量阵等敏度表达式。以动态结构响应量最小化或最大化为目标,体积比为约束,建立了动态结构拓扑优化模型,通过二维结构数值算例对理论方法进行验证。结果表明,方法在连续体结构动态拓扑优化设计中具有可行性和有效。     

应力约束下三维连续体结构拓扑优化分析

本文基于ICM(独立、连续、映射)方法,建立了以连续体结构重量为目标的应力约束下多工况的三维连续体结构拓扑优化模型。利用von Mises强度理论,提出了应力约束全局化方法,从而将局部的应力约束转化为全局的应变能约束,由此减少了约束数目,降低了问题的规模,避免了敏度分析。另外,利用对偶理论,将建立的优化模型转化为对偶模型,用序列二次规划法进行了求解,从而减少了设计变量的数目,提高了求解效率。借助MSC．Patran和MSC．Nastran提供的软件平台,利用PCL语言实现了本文的优化算法。数值算例验证了方法的可行性与有效性。     

Formulation of Three-Dimensional Joint Constraints Using the Absolute Nodal Coordinates

A wide variety of mechanical and structural multibody systems consist ofvery flexible components subject to kinematic constraints. The widelyused floating frame of reference formulation that employs linear modelsto describe the local deformation leads to a highly nonlinear expressionfor the inertia forces and can be applied to only small deformationproblems. This paper is concerned with the formulation and computerimplementation of spatial joint constraints and forces using the largedeformation absolute nodal coordinate formulation. Unlike the floatingframe of reference formulation that employs a mixed set of absolutereference and local elastic coordinates, in the absolute nodalcoordinate formulation, global displacement and slope coordinates areused. The nonlinear kinematic constraint equations and generalized forceexpressions are expressed in terms of the absolute global displacementsand slopes. In particular, a new formulation for the sliding jointbetween two very flexible bodies is developed. A surface parameter isintroduced as an additional new variable in order to facilitate theformulation of this sliding joint. The constraint and force expressionsdeveloped in this paper are also expressed in terms of generalizedCholesky coordinates that lead to an identity inertia matrix. Severalexamples are presented in order to demonstrate the use of theformulations developed in the paper.     

基于双向演化的局部水平集组合算法用于拓扑优化

本文基于双向渐进结构优化(BESO)方法和局部水平集方法(LLSM),提出局部水平集组合算法,在拓扑优化中实现双向演化。为改进LLSM的孔洞成核能力,新算法以所提离散水平集函数为节点设计变量,拓扑导数为灵敏度,按照BESO优化准则进行双向演化得到稳定拓扑解。然后通过迭代求解距离正则化方程(DRE)来组合LLSM获得最终拓扑。在LLSM中,DRE用来代替重生成步骤,并构造条件稳定差分格式求解DRE。最终给出典型实例验证所提算法的收敛性和数值稳定性。     

Performances of Unknown Input Observers for Chaotic LPV Maps in a Stochastic Context

Jointly modeling chaotic maps as Linear Parameter Varying (LPV) systems and using Unknown Input Observers (UIOs) for retrieving the information in a secure communication scheme has previously been motivated in a deterministic context [Millerioux, G. and Daafouz, J. International Journal of Bifurcation and Chaos 14(4), 2004]. In this paper, some new theoretical results from a control theory point of view, concerning the design in a stochastic and so more realistic context of UIOs for chaotic LPV systems is provided. The design of such observers is expressed in terms of the resolution of a finite set of Matrix Inequalities constraints and guarantees some prescribed performances on the state reconstruction error.     

一种改进的结构拓扑优化水平集方法

针对有限的结构设计区域,界面力的大小和作用的位置以及部分给定位移的结构边界始终保持不变的要求,为了解决基于水平集方法的结构边界演化停滞等问题,导出了基于水平集演化的结构拓扑优化所需的法向速度场.研究了由该法向速度场导出的序列解的收敛特性,并建立了一套新的结构柔顺度为目标函数,体积为约束条件的水平集演化算法,给出的算例验证了该方法的正确性和有效性.     

Euler-Lagrange Equations for Nonlinearly Elastic Rods with Self-Contact

 We derive the Euler-Lagrange equations for nonlinearly elastic rods with self-contact. The excluded-volume constraint is formulated in terms of an upper bound on the global curvature of the centre line. This condition is shown to guarantee the global injectivity of the deformation of the elastic rod. Topological constraints such as a prescribed knot and link class to model knotting and supercoiling phenomena as observed, e.g., in DNA-molecules, are included by using the notion of isotopy and Gaussian linking number. The bound on the global curvature as a nonsmooth side condition requires the use of Clarke's generalized gradients to obtain the explicit structure of the contact forces, which appear naturally as Lagrange multipliers in the Euler-Lagrange equations. Transversality conditions are discussed and higher regularity for the strains, moments, the centre line and the directors is shown. (Accepted December 20, 2002) Published online April 8, 2003 Communicated by S. S. Antman     

A heterogeneous modeling method for porous media flows

This paper presents a new heterogeneous multiscale modeling method for porous media flows. Physics at the global level is governed by one set of PDEs, while features in the solution that are beyond the resolution capacity of the global model are accounted for by the next refined set of governing equations. In this method, the global or coarse model is given by the Darcy equation, while the local or refined model is given by the Darcy–Stokes equation. Concurrent domain decomposition where global and local models are applied to adjacent subdomains, as well as overlapping domain decomposition where global and local models coexist on overlapping domains, is considered. An interface operator is developed for the case where global and local models commute along the common interface. For the overlapping decomposition, a residual‐based coupling technique is developed that consistently facilitates bottom‐up embedding of scale effects from the local Darcy–Stokes model into the global Darcy model. Numerical results are presented for nonoverlapping and overlapping domain decompositions for various benchmark problems. Computed results show that the hierarchically coupled models accurately account for the heterogeneity of the medium and efficiently incorporate local features into the global response. Copyright © 2014 John Wiley & Sons, Ltd.     

基于拟满应力设计和遗传算法的网架截面优化方法

通过在遗传算法中嵌入拟满应力算子,提出了一种以网架结构杆件截面作为离散变量的优化设计方法,即基于拟满应力设计和遗传算法的网架截面优化方法.分析结果表明,该法能够提高遗传算法的搜索效率和获得全局最优解的可靠性,对于同时有应力和位移约束的网架等空间结构截面优化问题,这种混合算法有较高的效率.     

Design of functionally graded composite structures in the presence of stress constraints

In this work new homogenization results are used to introduce a methodology for the design of structural components made from composite materials in the presence of stress constraints. A numerical method is developed for finding functionally graded materials that provide the maximum torsional rigidity while keeping the mean square stress inside the composite structure below a prescribed level.     

考虑破损-安全的连续体结构拓扑优化ICM方法

本文瞄准连续体在破损-安全考虑下的结构拓扑优化问题,旨在克服传统模型求解所得最终构型存在的弊病,避免结构因缺乏合理的冗余结构而敏感于局部破坏,实现破损-安全的目标. 首先,梳理了以往虽然用到却并不明晰的4个概念:结构局部破损模式、结构局部破损区域、结构破损状况、结构破损状况的预估分布. 之后,基于独立连续映射(ICM)方法,对该问题建立了力学性能约束下结构体积极小化的模型. 建立目标函数时,利用Minimax的概念将可能出现的结构破损状况对应的所有结构体积目标转化为原结构的唯一结构体积目标,克服了多目标问题的困难. 建立近似约束函数时,将可能出现的所有结构破损状况对应的力学性能的约束皆考虑进去,既能处理载荷单工况也能处理载荷多工况. 最后,以位移约束为例,建立了优化模型并求解. 单工况及多工况位移约束拓扑优化算例验证了算法的有效性. 结果表明:本方法相比于不考虑破损-安全的拓扑优化设计,得到的最优拓扑更复杂,体积比更大即所用材料更多,亦即最优结构具有更多的冗余,此正是考虑破损-安全设计原则的结果. 本文的研究对于航空、航天、其他水、陆等领域运载工具以及其他工程结构在意外破坏、战争创伤或恐怖袭击下的结构设计,乃是非常重要的进展.      

Optimum Design of Multistory Multispan Frames for Prescribed Elastic Compliance∗

This paper presents an analytical method for minimum cost design of regular rectangular building frames for constrained elastic compliance. The method, consisting of a semi-inverse method and a design region extension procedure, is illustrated by two classes of exact solutions. It is shown that the relative story displacements are almost uniformly distributed in an optimally designed frame with almost uniform story heights and that the solutions enable one to calculate all maximum member-end stresses from member-end curvatures. It is suggested that the proposed design formulas may be utilized for design problems, subject to relative story displacement and stress constraints. Since these solutions are shown to be dependent upon the nature of the prescribed minimum stiffnesses, a method of finding a practically reasonable set of minimum stiffnesses is also presented.     

桁架结构拓扑优化设计可行域研究

本文提出了桁架结构拓扑优化设计的k维可行子域、相邻可行子域、k维连通可行子域、k维奇异可行子域的定义，在这些定义的基础上，采用集合描述的方法，对Rozvany关于结构拓扑优化设计奇异最优解的定义进行了重新描述。基于以上定义，本文研究了桁架结构拓扑优化设计的可行域，证明了对于截面尺寸下限为零，且无尺寸上限的桁架结构受应力约束的拓扑优化设计问题，其设计空间不同拓扑的可行子域总是连通的，同时也给出了对于具有尺寸下限约束、具有局部稳定性约束的桁架结构拓扑优化设计的可行子域不连通的实例。     

考虑稳定性约束的五零能模式材料结构设计方法研究

五零能模式材料是一种声学超材料,虽然它一般采用固体材料设计,并且单胞常应用微桁架构型,但理论上这种材料设计的结构是不稳定的。本文从分析设计结构整体出发,探讨一种通过施加预载荷,确保五零能模式材料结构稳定性的方法,并给出了保证结构稳定性的充分条件,且应用典型的五零能模式材料单胞构型加以验证。同时,为方便提出的稳定性条件在设计过程中的实施,将五零能模式材料结构的设计过程描述为五零能模式材料的变换过程,采用结构尺寸优化模型来描述。当材料单胞为微桁架构型时,这一结构尺寸优化模型简单,结构稳定性约束和五零能模式材料性质约束均为线性约束。