共查询到18条相似文献,搜索用时 62 毫秒
1.
2.
连续体结掏多目标拓扑优化是结构优化领域中的一个较难的研究课题.本文提出了一种改进的SPEA2多目标优化算法.该算法中采用数学形态学中的四方向链码的编码方式进行结构拓扑表达,使产生的拓扑结构清晰且无异议,完全消除了模糊的拓扑边界和棋盘格现象.提出了"质量向量"的概念,用以量化两个结构拓扑之间的相似性,从而使得不同结构拓扑之间的相似性比较成为可能.同时还将机器学习中的范例学习的思想融入到新算法中,利用前面的有限元分析结果对后来的结构分析进行指导,剔除了很多不必要的重复的计算,从而使算法的总计算量大幅度地减少.将新算法应用于悬臂和简支两类深梁的多目标拓扑优化,获得了高质量的Pareto最优解,且具有很好的分布. 相似文献
3.
4.
5.
6.
7.
提出了一种改进的遗传算法,用于优化具有离散尺寸、连续形状和0-1拓扑变量的桁架问题。考虑到离散和连续变量的本质,文中提出了混合编码方法,其中包括二进制和实数编码,整数和实数编码。本文采用了凝聚选择法-基于约束和适应度值双重标准,完全适应约束问题的本质。在优化过程中,初始种群和算子具有不确定性,因此有必要检验结构拓扑的合理性。为了增强算法的可靠性,采用了改进的重新开始算子,引入新基因并且探索新空间。求解了典型的算例,证明改进的遗传算法是可行且有效的。 相似文献
8.
针对柔性伸缩蒙皮支撑结构的多目标拓扑优化问题,分析了其实际应用中的优化对象及目标函数,提出使用六边形对设计域进行离散的方法,并建立离散模型与位矩阵的映射关系。采用基于位矩阵的NSGA-Ⅱ(Non-dominated Sorting Genetic Algorithm Ⅱ)进行求解。优化过程中引入个体连通性分析以提高计算效率。通过对可行个体进行有限元分析,获得其目标值;使用罚函数法对不可行个体加以惩罚,最终得到一组互不支配的支撑结构。结果表明,本方法可为柔性伸缩蒙皮支撑结构的多目标拓扑优化问题提供可行、有效的解,也可用于求解其它二维多目标拓扑优化问题。 相似文献
9.
连续体结构拓扑优化的高精度逼近ICM方法 总被引:2,自引:0,他引:2
采用指数类函数为快滤函数的高精度逼近ICM
(independent continuous and mapping)方法, 建立了以结构重量为目
标, 应力和位移共同约束下的连续体结构拓扑优化模型. 利用结构畸变比能的方法全局化应
力约束, 单位虚载荷法显式化位移约束, 归一化约束以解决约束限数量级不一致的问题. 针
对不同性态的过滤函数, 给出了指数类快滤函数参数的取值方法. 单工况和多工况的算例表
明了高精度逼近的ICM方法处理多种约束下连续体结构拓扑优化的可行性与有效性. 相似文献
10.
考虑性态约束时多工况桁架结构拓扑优化设计 总被引:9,自引:0,他引:9
本文提出了一种适用于桁架结构的拓扑优化设计方法。它以杆内力为设计变量,以结构重量为目标函数。该方法的主要特点是:第一,通过引入杆内力为设计变量,既克服了已有方法要求预定位移场这一主要困难,又为在拓扑优化过程中考虑应力、位移等性态约束创造了条件;第二,将多工况的拓扑优化问题描述为一个非光滑的数学规划问题,再通过一个变量代换将其转化为一般的规划为题,进而将原问题的求解又转化为几个线性规划问题的求解;第三,基于结构力学的三个基本方程,将位移与应力约束提成为线性不等式约束,这些约束同重量的目标函数一起构成了拓扑优化设计的线性规划模型。最后,将本方法应用于几个工程算例,得到了满意的数值计算结果。 相似文献
11.
This paper presents an approach to solving truss topology optimization problem with small uncertainty in the locations of the structural nodes. The nodal locations in the truss are assumed to be random, and the probabilistic method is used here to deal with the uncertainty. The objective of the optimization problem is to minimize the mean compliance of the truss structure under nodal location uncertainty. It is a well-acknowledged barrier to compute the inverse of the structural stiffness matrix which involves variations in the optimization problem. In this paper, based on Neumann series expansion, this optimization problem can be recast into a simpler deterministic structural optimization problem. In order to avoid the sensitivity calculations for the objective function, the proportional topology optimization method which shows comparable e?ciency and accuracy with gradient-based method is used. The numerical examples demonstrate the effectiveness and high e?ciency of the proposed approach, and further illustrate that the optimal truss topology can be dramatically impacted by nodal location uncertainties. 相似文献
12.
阶跃函数高精度逼近的结构拓扑优化方法 总被引:3,自引:0,他引:3
为了提高ICM(Independent Continuous and Mapping,即独立、连续及映射)方法求解结构拓扑优化问题的效率,本文改进了阶跃函数及其反函数的近似逼近函数——磨光函数和过滤函数。首先,分别对ICM方法的磨光函数和过滤函数按其近似性质进行了分类,分别提出了左磨函数及上磨函数和快滤函数、慢滤函数诸概念。然后得到了区分左磨函数和上磨函数、快滤函数和慢滤函数的两个判别定理;并得到了上磨函数、快滤函数、左磨函数及慢滤函数的对应定理。进而给出了磨光函数和过滤函数的使用准则及构造方法。采用高精度逼近阶跃函数的指数类函数做左磨函数,建立近似程度更高的结构拓扑优化模型。上述策略带来了模型非线性程度的提高,增加了求解难度。为此,针对该模型给出了精确对偶映射下的序列二次近似解法。最后,以位移约束下结构重量最轻化问题为例,叙述了相应的算法。与以往采用幂函数做磨光函数时算例结果的比较表明,该模型的提法合理,算法更加有效。由于提高了对阶跃函数及其反函数的逼近程度,从而显著减少了优化迭代的次数。 相似文献
13.
The present paper studies topology optimization of truss structures in multiple loading cases and with stress constraints.
It is pointed out in the paper that the special difficulty of adding bars and/or deleting bars from structure in the numerical
algorithm of truss topology optimization is caused by the discontinuity of stress functions at the zero cross sectional area
in the conventional formulation. In a new formulation, we replace the stress constraints by new constraints. The new constraints
retain the same feasibility of the stress constraints, but are continuous in the closed interval up to zero cross sectional
area. The new formulation enables us to solve topology optimization problem in the frame of the existing FEM software and
mathematical programming techniques. Powell constrained variable metric method is applied to a number of examples of truss
topology optimization. Numerical performances of the two formulations are compared. It is shown that in the conventional formulation
the iteration of numerical algorithm may be blocked by discontinuity of the stress constraint and often stops at a nonoptimum
solution. And in the new formulation the bar adding and bar deleting is done rationally and a local optimum, even the global
optimum can be obtained by iteration.
The project supported by the National Natural Science Foundation of China 相似文献
14.
针对进化或拓扑优化方法的不足,提出了一种基于遗传算法的新型进化式拓扑优化方法--三角网格进化法,该方法不仅能够同时进行拓扑,形状与截面变量优化设计,而且在优化过程中实现了退化和进化的统一,提高了优化效率。另外本文还首次对结构类型变量进行了优化计算,取得了有益的结果。最后几个数值算例证明了本方法的可行性和有效性。 相似文献
15.
16.
A new approach for the solution of singular optimum in structural topology optimization 总被引:4,自引:0,他引:4
In order to overcome the difficulties caused by singular optima, in the present paper, a new method for the solutions of structural
topology optimization problems is proposed. The distinctive feature of this method is that instead of solving the original
optimization problem directly, we turn to seeking the solutions of a sequence of approximated problems which are formulated
by relaxing the constraints of the original problem to some extent. The approximated problem can be solved efficiently by
employing the algorithms developed for sizing optimization problems because its solution is not singular. It can also be proved
that when the relaxation parameter is tending to zero, the solution of the approximated problem will converge to the solution
of the original problem uniformly. Numerical examples illustrate the effectiveness and validity of the present approach. Results
are also compared with those obtained by traditional methods.
The project supported by the National Natural Science Foundation of China under project No. 19572023 相似文献
17.
结构拓扑优化的变量连接,是通过对设计变量之间添加约束关系,从而得到特定的拓扑优化构型,使得优化结果能够满足工程上的特殊要求和工艺制造技术的限制。针对拓扑优化中的几类过滤形式及灵敏度分析,给出了考虑变量连接的计算公式;基于自主研发的SiPESC软件集成化平台,在SiPESC .TOPO拓扑优化模块上进行二次开发,构建了拓扑优化的变量连接算法框架,其核心思想是基于面向对象设计方法和软件设计模式,实现算法与数据分离。详细阐述了变量连接的作用方式,以及软件框架通用接口设计方案,并通过数值算例验证了其在静力问题、动力问题和热传导问题上的可行性。 相似文献