基于深度学习的复合材料铺层优化方法

复合材料在航空工程中有着广泛的应用,其中铺层优化对提高结构承载能力至关重要。本文中首先运用基于神经网络的深度学习方法快速预测设计结果,然后依据预测结果,通过遗传算法搜索复合材料层合板设计问题的最优解,提高铺层设计效率,最后对比应用深度学习技术与传统优化算法得到的优化结果,证明了深度学习技术在层合板铺层优化中的便捷性和有效性。     

基于改进双种群遗传算法的复合材料层合板铺层优化设计

针对复合材料层合板铺层优化设计过程中易出现的算法早熟及收敛慢的问题,结合多种群遗传算法和自适应遗传算法,提出一种引入自适应算子的改进双种群遗传算法。在算法过程中以对称复合材料层合板为例进行验证,通过Patran建立有限元初始模型,采用Matlab编写遗传算法主程序及数据传递程序,实现对Nastran的输入输出文件的读写,并在以Tsai-Wu准则为基础确立的适应度函数下,对复合材料层合板的铺层顺序进行优化。对比传统遗传算法,结果表明该改进算法能够明显提高优化效率,并能够有效收敛于全局最优解,对解决复合材料结构优化设计问题有一定的指导意义。     

层合板热膨胀系数优化设计的遗传算法实现

本文针对在复合材料设计领域中热膨胀系数的设计问题，用遗传算法对给定层数和铺层角度的层合板进行了铺层顺序的优化研究。使其满足特定的要求，通过计算看出，遗传算法可以快速得到结果，表明了遗传算法在复合材料设计领域可以作为一种有效的设计手段。所完成的优化软件具有一定的工程应用价值。     

提高层压板屈曲强度的铺层顺序优化方法讨论

复合材料层压板铺层优化设计对于提高飞机结构承载能力至关重要。本文基于遗传算法和随机正态分布优化算法,以轴压复合材料层压板屈曲荷载为目标函数,开展铺层顺序优化设计。采用MATLAB语言编写两种优化算法程序,对不同铺层数和边界条件的对称均衡层压板进行铺层顺序优化。结果表明:采用整数编码、罚函数处理约束条件的遗传算法能够优化层合板铺层顺序问题;当铺层总数为24层时,四边简支和四边固支层合板的最优铺层相比最差铺层的承载力分别提高了27%和15%,优化铺层设计效果明显。随机正态分布优化算法计算结果与遗传算法结果一致,随机正态分布优化算法更简单实用、易收敛,更适用于层压板的铺层顺序优化设计。     

基于随机机会约束规划的复合材料层合板固有频率的铺层角优化设计

对具有随机弹性常数的复合材料层合板的固有频率优化问题进行研究。从单层板的刚度矩阵出发,给出层合板的刚度特性及一阶固有频率最大化的确定性模型,再根据Liu提出的随机机会约束理论,建立了一阶固有频率最大化的Maximax及Minimax的机会约束规划模型,根据试验及随机理论假设弹性常数服从经验分布,给出了弹性常数的经验分布随机数产生及处理随机机会约束的算法,再给出了基于随机模拟的遗传算法求解方法。最后给出了一个具体的复合材料层合矩形板在四边简支边界条件下基频优化的实例,验证了基于随机模拟的遗传算法求解的有效性。     

用遗传算法实现特殊泊松比的层合板设计

本文针对在复合材料层合板设计中泊松比的设计问题,用遗传算法对给定层数的层合板进行了泊松比的优化研究,使其满足特定的要求.通过计算看出,与传统优化方法不同,遗传算法不需要特定的初始条件就可以快速得到结果,表明了遗传算法在复合材料设计领域可以作为一种有效的设计手段,所完成的优化软件具有一定的工程应用价值.     

基于改进自适应遗传算法的复合材料铺层优化设计

为提高采用遗传算法的复合材料铺层优化计算效率,提出了改进的自适应遗传算法,其交叉和变异算子在迭代过程中能够根据种群迭代收敛趋势进行自适应调整。通过二维Rasrtigrin测试函数试验,表明改进自适应遗传算法能够增加全局收敛次数,提高收敛速度。在此基础上,编写整数编码的改进自适应遗传算法,并结合经典层合板理论,以质量轻量化为目标,结构强度和刚度为约束条件,对复合材料层合板进行铺层优化设计,算例迭代过程显示收敛到最优解自适应遗传算法需进化27代,改进自适应遗传算法需进化19代。结果表明,提出的改进自适应遗传算法能够明显提高复合材料铺层优化设计效率。     

复合材料声呐导流罩铺层的遗传算法优化设计

以复合材料力学经典层合板理论为基础,利用MATLAB编程联合ANSYS有限元分析,建立了复合材料声呐导流罩铺层结构优化平台。首先对单向载荷下的层合板铺层结构进行优化,并和实验结果相比较,验证了该优化平台的有效性。结果表明,优化后的层合板强度有了较大提高。在此基础之上,根据水下实际载荷工况,对声呐导流罩的铺层进行了优化设计,并利用ANSYS有限元分析对优化后的铺层进行应力应变分析。结果表明,声呐导流罩强度性能显著提高,实现了声呐导流罩铺层的优化设计。     

基于蚁群算法的复合材料层合板的铺层顺序优化

本文将蚁群算法引入到复合材料层合板铺层顺序的优化设计中,对给定各角度铺层数的层合板进行优化设计。本文采用遗传算法(GA)中的整数编码方法,将层合板的铺层顺序优化问题转化为求解旅行商问题(TSP),接着为满足工程实际的需要,对基本蚁群算法作了适当修改,以提高优化效率及性能。算例结果表明,对于确定层合板最佳铺层顺序的这种离散变量优化问题,蚁群算法是一种行之有效的方法,且编程简单、可移植性好,对工程结构优化设计有一定的参考价值。     

应用ANSYS的复合材料层合板静强度预测

复合材料层合板的静强度预测对于层合板设计和应用都具有重要的意义。为了提高预测的有效性,减少实验成本,在基于单层板理论的逐渐累积损伤的静强度预测方法的基础上,提出了用于二维机织复合材料静强度的预测方法。首先,以ANSYS有限元分析软件作为平台,利用其自带的APDL参数化语言建立有限元模型,并将铺层方式设定为[(0,90)/(±45)/(0,90)/(±45)/(0,90)]s,然后经过ANSYS软件的计算,得到最终的预测结果,最后制作该铺层方式的复合材料层合板并作拉伸实验,得到其实验强度,通过对比预测结果和实验结果验证了该预测方法的有效性。     

Effect of stacking sequence on the compressive performance of impacted aramid fiber/glass fiber hybrid composite

Aramid fiber/glass fiber hybrid composites were prepared to examine the compressive performance of impacted composites. The effect of stacking sequence and surface treatment on compression after impact (CAI) performance of three‐layer hybrid composites was investigated with respect to delamination area. As the impact velocity increased, the laminates exhibited a significant reduction of compressive strength owing to larger delamination area within laminate. The surface treatment aramid fiber reduced the delamination area and enhanced the resistance to buckling. The strength reduction of laminate AAA was considerable because of wide delaminated region, whereas the residual strength of laminate GGG was not affected by impact energy because the laminate absorbed most of impact energy through formation of fiber cracks rather than delamination. Considering stacking sequence, the laminate GAG and GAA exhibited an energy threshold due to insensitivity to impact damage. As a result, the residual performance of composite was primarily dominated by delamination extent rather than fiber cracks.     

Numerical simulation study of ballistic performance of Al2O3/aramid-carbon hybrid FRP laminate composite structures subject to impact loading

Aramid-carbon hybrid fiber reinforced polymer (FRP) laminates have been widely used in the design of the ceramic/FRP laminate composite armor systems for their outstanding mechanical properties. In order to figure out the influence of the stacking sequence of the aramid-carbon hybrid FRP laminate on the ballistic performance of the ceramic/FRP laminate composite structures, a finite element model (FEM) which is used to simulate the condition of the 7.62 M61 AP against the Al₂O₃/aramid-carbon hybrid FRP laminate composite structure is carried out in ABAQUS/Explicit software, and the JH-2 constitutive model and 3D-Hashin failure criteria which is programmed by a user subroutine VUMAT are utilized to describe the mechanical behavior of Al₂O₃ and FRP laminate in this FEM. The accuracy of this FEM is checked by a verification FEM and the method of establishing this FEM is proved to be reasonable. The simulation results reveals that when the carbon fiber is stacked at the top of the aramid-carbon hybrid FRP laminate, the ballistic performance and the integrity of the geometry structure of the ceramic/FRP laminate composite structure is the best when the proportion of the carbon fiber in the hybrid FRP laminate is a constant, which provides an innovative way to enhance the ballistic performance of the ceramic/FRP laminate composite armor system.     

A study on the lap shear strength of a co-cured single lap joint

In this paper, the lap shear strength of a co-cured single lap joint subjected to a tensile load was investigated by experimental analysis. Co-cured joint specimens with several different bonding parameters such as bond length, surface roughness, and stacking sequence of the composite laminate were fabricated and tested. The dependence of the lap shear strength of the co-cured joint on the bonding parameters was investigated from the experimental results. The failure mechanism of the co-cured single lap joint was partially cohesive failure. The lap shear strength of the co-cured single lap joint was significantly affected by the bond length and the stacking sequence of the composite laminate. However, the effect of surface roughness on the lap shear strength of the co-cured single lap joint was not so significant.     

Tensile behavior of symmetrical laminates made of PCBA and PMMA layers

The stiffness and strength of a composite material in form of laminate is obtained from the properties of the constituent laminae. The stacking sequence of the laminate affects the mechanical behavior. The interface between different laminae is also an important factor, since it influences the stresses that are developed in the laminate, and hence the strengths. In this study, a theoretical investigation of the mechanical behavior of symmetrical laminates made of isotropic layers [Lexan (PCBA) and Plexiglas (PMMA)] was attempted. An analysis based on the lamination theory was performed to determine the stress distribution and strains as well as the elastic constants. Experimental measurements with specimens made of laminates with different stacking sequences were carried out. The obtained values were compared with the theoretical values given by the lamination theory and mechanics of materials approach. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 101: 3215–3226, 2006     

Genetic algorithm for short-term scheduling of make-and-pack batch production process

This paper considers a scheduling problem in industrial make-and-pack batch production process. This process equips with sequence-dependent changeover time, multipurpose storage units with limited capacity, storage time, batch splitting, partial equipment connectivity and transfer time. The objective is to make a production plan to satisfy al constraints while meeting demand requirement of packed products from various product fam-ilies. This problem is NP-hard and the problem size is exponentially large for a realistic-sized problem. Therefore, we propose a genetic algorithm to handle this problem. Solutions to the problems are represented by chromo-somes of product family sequences. These sequences are decoded to assign the resource for producing packed products according to forward assignment strategy and resource selection rules. These techniques greatly reduce unnecessary search space and improve search speed. In addition, design of experiment is carefully utilized to de-termine appropriate parameter settings. Ant colony optimization and Tabu search are also implemented for com-parison. At the end of each heuristics, local search is applied for the packed product sequence to improve makespan. In an experimental analysis, al heuristics show the capability to solve large instances within reason-able computational time. In al problem instances, genetic algorithm averagely outperforms ant colony optimiza-tion and Tabu search with slightly longer computational time.     

Superposition of the notched strength of composite laminates

An improved macroscopic model for predicting the strength of a composite laminate containing a circular notch is developed. Two constants are introduced which uniquely determine the notch sensitivity of a given material. A superposition method for the notched strength of composite laminates is developed which allows data for arbitrary materials and laminate configurations to be superimposed upon a single master curve. The influence of material orthotropy upon notched strength is discussed. A relative notch sensitivity parameter is introduced which allows quantification of the notch sensitivity of a given composite material system, stacking sequence, or laminate configuration.     

Stress analyses of composite tee joints at elevated temperature

Thermal–structural coupling nonlinear finite element analyses are conducted in this paper to determine three-dimensional stresses of a composite tee joint, which is formed when a right angled plate is adhesively bonded to a base plate at elevated temperature. The von-Mises stresses of the adhesive layer of the tee joint with three different laminate stacking sequences viz. unidirectional [0]₈, cross-ply [(0/90)_s]₂, and angle-ply [(+45/?45)_s]₂ laminates have been evaluated when the tee joint is subjected to an out-of-plane loading through the right angled plate in addition to an elevated temperature applied at the undersurface of the base plate. The effects of laminate stacking sequence and temperature on von-Mises stresses have been investigated in this paper. The effects of the coefficient of thermal expansion and thermal conduction of the adhesive layer on von-Mises stresses have also been studied. Conclusions about the stresses of the composite tee joint with different stacking sequence, different coefficient of thermal expansion, and different thermal conduction of the adhesive layer are drawn.     

Stiffness and strength of asymmetric laminates made of isotropic PCBA and PMMA layers

In this work the stiffness and strength of a composite material in the form of an asymmetric laminate are obtained from the properties of the constituent laminae made of isotropic layers, Lexan (PCBA) and Plexiglas (PMMA). The influence of the stacking sequence of the laminate and the interface between the laminae that affects the mechanical properties are investigated. A theoretical analysis based on the lamination theory was performed to determine the stress and strain distribution as well as the elastic constants. Experimental measurements with specimens made of asymmetric laminates of various stacking sequences were carried out. The obtained values were compared with the theoretical values given by the lamination theory and mechanics of materials approach. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 101: 4493–4503, 2006     

基于遗传算法的层压板强度优化设计

本文的主要目的是研究遗传算法的优化机理并将遗传算法应用到复合材料层压板铺层的优化设计中。在深入理解力学原理,熟悉掌握数值计算方法的基础上对层压板进行强度优化设计。因此,在深入研究单层复合材料以及叠层复合材料的弹性特征理论基础和改进的遗传算法优化理论之后,针对复合材料层压结构遗传算法优化设计中层压结构参数具有离散型的特点,提出了适合复合材料层压结构遗传算法优化设计的整数编码策略,以整数来表征层压结构参数,这种编码方式简单直观、使用方便、易于进行遗传算法的交叉及变异操作。在分析层压结构强度的基础上,针对结构强度优化的目标构造了可用于遗传算法的适应度函数。同时参考了一定的铺层规则,在铺层角度限制为工程中常用的四种角度的前提下,应用遗传算法对层压板进行了铺层优化设计。设计结果表明,由于遗传算法特有的处理离散型问题的优势,在层压板铺层的优化设计中应用遗传算法是可行和可信的。