基于径向基函数神经网络的白车身减重优化研究

针对白车身变量多、性能响应复杂问题,在车身结构的板厚优化设计中引入近似模型方法,提高设计效率。以某轿车车身结构为轻量化设计对象,通过灵敏度分析确定优化设计变量,基于径向基函数神经网络近似模型进行全局优化,在不降低刚度和模态性能的情况下实现白车身减重目标。通过最优拉丁超立方试验设计构造样本点,用径向基函数神经网络法构造刚度和模态的近似模型,并用自适应模拟退火法进行优化求解。结果表明,径向基函数神经网络模型能较好地模拟车身结构刚度和模态响应问题,提高了整体的设计效率。通过有限元模型的验证,基于近似模型的优化结果精度较高,实例白车身减重达5.73%。     

机床床身结构优化设计方法

结合基于近似模型的尺寸优化及拓扑优化技术,提出了机床床身结构优化设计的策略.首先以床身结构的刚度和基频为设计目标,以床身内部横隔板间距及厚度、垫铁间距为设计变量,通过中心复合设计法生成实验所需的设计变量的样本点,对生成的样本点进行有限元分析得到样本点的响应,借助最小二乘法建立近似模型二阶响应曲面.采用折衷法建立多目标优化模型,并用遗传算法进行求解.在此基础上,采用密度法对床身结构进行拓扑优化.实例计算结果表明,优化后床身质量减轻了4.45%,而刚度提高了5.8%.     

仿生轻质结构在飞机大开口区的应用及其优化设计

将一种新型的仿甲虫鞘翅轻质结构应用于飞机大开口区的筋板结构设计,根据飞机大开口区对结构静/动态性能和散热性能的要求,分别构造目标函数,并采用拉丁超立方试验设计方法确定样本点数,建立有限元参数化模型进行仿真计算,利用响应面法得到各目标函数的响应面拟合函数,之后设定结构的多功能协同优化目标,并通过线性搜索法进行协同优化设计,优化后计算得到仿生轻质结构的散热性能、抗压刚度分别是优化前的14.3倍和2.1倍,动态性能指标得到明显改善,在相同载荷条件下结构减重11%;进一步以飞机大开口区加筋结构的一阶屈曲因子和最大位移作为优化约束条件,并以结构总质量最小为优化目标,构造各自响应面拟合函数,基于遗传算法优化筋板结构整体布局。优化后开口区加筋结构总质量减少15%,屈曲因子为1.02,较优化前提高21%,结构最大位移为12.1 mm,较优化前减少20%,优化效果显著。     

基于刚度和模态的某自卸车车架轻量化设计

利用有限元模态分析理论,分析了某自卸车车架的动态特性,结合车架模态试验验证了模型的合理性。在已验证的模型基础下求解了车架的弯曲和扭转刚度。进行灵敏度分析以确定有效的设计变量。以车架总质量为目标函数,以刚度和1阶固有频率为约束条件,建立轻量化优化模型。通过拉丁方试验设计方法选取样本点,运用径向基函数构造模型各响应的近似模型,采用遗传算法求解近似模型。获得的优化结果参数代入实际模型后结果表明车架总质量在满足刚度和1阶模态频率约束要求下下降了18.37 kg。     

利用神经网络实现复杂结构的多目标优化设计

结构优化设计中常常包含大量的有限元计算。现代多目标优化设计的发展趋势是以 Pareto遗传算法为代表的随机搜索方法 ,能够搜索到整个 Pareto最优解集 ,但计算量相当大 ,如果每次迭代都要涉及有限元计算 ,将是非常耗时的工作。本文在利用 Kolm ogorov多层神经网络映射存在定理的基础上导出的用神经网络进行结构近似分析的方法 ,用均匀试验设计方法选取特征样本点供神经网络训练 ,将神经网络与 Pareto遗传算法有机地结合 ,使多目标优化的计算效率进一步提高     

碟形直线超声电机优化设计

为了有效和快速地实现直线型超声电机的设计,提出了一种基于响应面法的直线超声电机优化设计方法并应用于蝶形直线型超声电机.响应面法是一种试验设计与数理统计相结合的优化方法.首先,利用有限元法建立定子的参数化模型并选定设计变量;其次,利用试验设计方法在变量空间里选取样本点,对各个样本点所对应的结构利用APDL建立有限元模型并进行模态分析和谐响应分析,得到对应各样本点的响应值,利用这些样本点和响应值建立定子响应面近似模型;最后,利用单纯形法找到最佳的设计方案.优化结果显示,蝶形直线超声电机的多方面设计要求都得以实现.     

基于响应面的三自由度超声电机定子设计优化

针对超声电机定子设计过程存在的定子设计模态与干扰模态分离不彻底、定子振动幅度小、定子振动时局部内应力较大等问题,提出了基于响应面模型的电机定子设计优化方法。首先,通过选取合适的定子设计尺寸作为变量,拉丁超立方抽样方法在设计空间内选取样本点;其次,对各样本点的组合得到的定子尺寸结构进行有限元分析,得到定子的模态和谐响应值从而建立定子的响应面模型;然后,利用遗传算法在响应面模型的基础上对定子进行优化;最后,使用有限元软件对优化前后的定子进行建模计算验证了优化结果的正确性。     

基于响应面法的无轨门机T型架优化设计

针对无轨门机T型架设计存在冗余的现象,在满足设计要求的前提下,对T型架进行优化设计。以某无轨门机T型架的结构尺寸与板厚为初始设计变量,以刚度、强度与结构自重为输出响应,通过ISIGHT中的最优拉丁超立方设计方法生成初始样本点,利用试验分析筛选出最终设计变量。利用ANSYS与ISIGHT软件,建立响应面近似模型,以T型架的刚度与强度为约束条件,其结构自重最轻为优化目标,采用多岛遗传算法进行全局寻优。结果表明:该优化方法使T型架的质量减轻了36.12%,轻量化效果较为明显,为无轨门机T型架的设计与改进提供了一定的理论依据。     

剐齿机床立柱动力学模型修正技术研究

针对机床立柱与床身接触刚度复杂,在有限元建模时边界条件难以确定,会导致机床立柱有限元模态分析结果与模态试验结果存在偏差的问题,通过响应面法与粒子群优化算法组合的方法修正机床立柱的动力学模型。以SKGC-500剐齿机床立柱为研究对象,将立柱与床身之间的导轨连接处理成弹簧阻尼单元,设计变量选择法向刚度和切向刚度。用正交试验设计方法选择有限元分析样本点并拟合二阶多项式响应面模型,建立以固有频率最小二乘法为目标函数,以接触刚度为约束条件的优化模型,用粒子群优化算法对固有频率最小二乘进行优化。结果表明,修正后的立柱动力学模型与实际固有频率有更好的一致性。     

基于振动特性的多学科多目标摩托车优化

以摩托车强度、刚度、轻量化为约束条件,以响应面法为基础建立多学科多目标优化方案.运用Hypermesh建立有限元模型,通过Radioss进行计算,并以Hyperstudy为优化平台.采用正交实验设计对优化变量进行灵敏度分析,运用拉丁方实验设计获取采样数据点,利用移动最小二乘法构建响应面模型,结合多学科、多目标优化设计方法进行结构优化,实现提高车架结构特性和摩托车动态特性的目标.     

Dynamic analysis and structural optimization of a fiber optic sensor using neural networks

The objective of this work is to apply artificial neural networks for solving inverse problems in the structural optimization of a fiber optic pressure sensor. For the sensor under investigation to achieve a desired accuracy, the change in the distance between the tips of the two fibers due to the applied pressure should not interfere with the phase change due to the change in the density of the air between the two fibers. Therefore, accurate dynamic analysis and structural optimization of the sensor is essential to ensure the accuracy of the measurements provided by the sensor. To this end, a normal mode analysis and a transient response analysis of the sensor were performed by combining commercial finite element analysis package, MSC/NASTRAN, and MATLAB. Furthermore, a parametric study on the design of the sensor was performed to minimize the size of the sensor while fulfilling a number of constraints. In performing the parametric study, the need for a relationship between the design parameters and the response of the sensor was fulfilled by using a neural network. The whole process of the dynamic analysis using commercial finite element analysis package and the parameter optimization of the sensor were automated within the MATLAB environment.     

Parametric study and optimization of a food can corrugation design using a response surface method

This paper presents the parametric design and functional optimization of a thin-walled food container with a corrugated surface. The configuration of the can corrugation should be designed to minimize the use of raw material subject to the constraints of the targeted structural performance. In the present study, the failure behaviors and the buckling strengths of a commercial food can under paneling pressure and axial loading are investigated with a series of experiments and finite element analyses. Full factorial design is implemented to study the effects of the geometric parameters of the corrugation (e.g., depth, radius, spacing and number of beadings) on its strength. Parameter optimization using a rotatable central composite design is employed to identify an optimal corrugation design by approximating the response surfaces of the can strength in terms of the significant design variables. The obtained surfaces are derived through the analysis of variance, and the suitability of the response is justified. A light- weight can body is then achieved by reduction of the can body thickness according to the required strength characteristics. Finite element analysis of the optimal model is also performed to confirm the predicted results. By using the proposed procedure, the can-body weight can be reduced by up to 12% compared with the original design.     

轻型货车EQ1060F车身的结构优化设计研究

主要研究轻型货车 EQ1 0 6 0 F驾驶室车身结构优化设计中的一些关键问题 ,包括结构有限元分析、优化方法选择、敏度分析和结构重分析 ,选择了序列二次规划法用于结构优化问题且收敛效果较好。该研究为汽车车身设计提供了一条新的设计思路。     

轻型货车EQl060F车身的结构优化设计研究

主要研究轻型货车EQl060F驾驶室车身结构优化设计中的一些关键问题,包括结构有限元分析、优化方法选择、敏度分析和结构重分析,选择了序列二次规划法用于结构优化问题且收敛效果较好。该研究为汽车车身设计提供了一条新的设计思路。     

有限元的神经网络计算方法研究

根据有限元总刚矩阵经修正后具有正定性的特点以及弹性体位能函数的具体形式，提出一种新的神经网络有限元计算模型，即模型中神经网络的能量函数与有限元的优化目标函数相等，从而避免由于神经网络自身结构的原因而带来的计算误差。同时，避免采用基于模拟退火算法等随机神经网络优化计算方法时求解结果的随机性和设定初始退火温度To、内循环次数判据、最终停止判据等人为因素的影响。理论分析和计算机仿真表明，文中提出的方法可靠、有效。     

基于材料-结构协同设计的层合板多级优化方法

提出一种以材料参数与结构参数为变量的复合材料层合板多级优化设计新方法。以单层纤维含量、层厚和铺层角为设计参数,建立了刚度和强度约束下的结构减重设计模型,基于有限元分析,运用所提出的三级优化策略完成结构轻量化设计。本文给出了不同载荷类型和位移边界条件下层合板优化算例,设计结果验证了方法的有效性。     

A Metamodeling Method Based on Support Vector Regression for Robust Optimization

Metamodeling techniques have been used in robust optimization to reduce the high computational cost of the uncertainty analysis and improve the performance of robust optimization problems with computationaUy expensive simulation models.Existing metamodels main focus on polynomial regression(PR),neural networks(NN)and Kriging models,these metamodeis are not well suited for large-scale robust optimization problems with small size training sets and high nonlinearity.To address the problem,a reduced approximation model technique based on support vector regression(SVR)is introduced in order to improve the accuracy of metamodels.A robust optimization method based on SVR is presented for problems that involve high dimension and nonlinear.First appropriate design parameter samples are selected by experimental design theories,then the response samples are obtained from the simulations such as finite element analysis,the SVR metamodel is constructed and treated as the mean and the variance of the objective performance functions.Combining other constraints,the robust optimization model is formed which can be solved by genetic algorithm(GA).The applicability of the method developed is demonstrated using a case of two-bar structure system study.The performances of SVR were compared with those of PR,Kriging and back-propagation neural networks(BPNN),the comparison results show that the prediction accuracy of the SVR metamodel was higher than those of other metamodels under uncertainty.The robust optimization solutions are near to the real result,and the proposed method is found to be accurate and efficient for robust optimization.This reaserch provides an efficient method for robust optimization problems with complex structure.     

基于ANN和GA的机床结构优化设计

机床主轴是机床中的重要部件,其设计对机床性能影响很大。在对主轴设计的各项要求中,刚度和振动稳定性是主要考虑的指标,将其归纳在一个含区域和性能约束的主轴结构优化设计数学模型中。借助有限元软件可获得不同主轴尺寸下的各项性能指标,利用人工神经网络(ANN)实现主轴各项性能指标的近似分析,并采用遗传算法(GA)实现了性能指标为设计变量隐式表达下的机床主轴结构优化设计。     

Optimum design of composite laminated plates via a multi-objective function

This paper deals with the optimum design of composite laminated plates under stiffness and gauge constraints. A multi-objective function which combines the plate weight and the strain energy stored in the plate by weighting parameters is introduced. This objective function is minimized while satisfying constraints such as the structural deformation and the limits on design variables. Both ply orientation angles and ply thicknesses of the composite plate are used as the design variables. The stiffness analysis is performed by the finite element method in which a triangular element is used that is suitable for the analysis of thin to thick plates and includes the transverse shear effects. Analyses of the derivatives of the objective function and the constraint functions with respect to the design variables is performed analytically. The mathematical programming method called the constrained variable metric is used to solve this optimum problem. An example is provided for the optimal design of a rectangular laminated plate.     

Lightweight design of an electric bus body structure with analytical target cascading

Lightweight designs of new-energy vehicles can reduce energy consumption, thereby improving driving mileage. In this study, a lightweight design of a newly developed multi-material electric bus body structure is examined in combination with analytical target cascading (ATC). By proposing an ATC-based two-level optimization strategy, the original lightweight design problem is decomposed into the system level and three subsystem levels. The system-level optimization model is related to mass minimization with all the structural modal frequency constraints, while each subsystem-level optimization model is related to the sub-structural performance objective with sub-structure mass constraints. To enhance the interaction between two-level systems, each subsystem-level objective is reformulated as a penalty-based function coordinated with the system-level objective. To guarantee the accuracy of the model-based analysis, a finite element model is validated through experimental modal test. A sequential quadratic programming algorithm is used to address the defined optimization problem for effective convergence. Compared with the initial design, the total mass is reduced by 49 kg, and the torsional stiffness is increased by 17.5%. In addition, the obtained design is also validated through strength analysis.