CFRP/Ti叠层构件钻孔工艺参数多目标优化方法

碳纤维增强复合材料/钛合金(CFRP/Ti)叠层构件以其优异性能广泛应用于航空航天领域。为保证孔同轴度和加工效率,通常采用同一参数对CFRP/Ti叠层构件进行一体钻孔。而CFRP和钛合金的材料属性和切削性能各异,一体钻孔会出现不同程度的CFRP撕裂、钛合金出口毛刺、孔径偏差等多种质量缺陷。由于工艺参数对各个质量缺陷的影响不同,甚至存在矛盾,难以找到合适的参数使得各质量缺陷均达到最小。为得到最优工艺参数,提高钻孔综合质量,对CFRP/Ti叠层构件钻孔工艺参数多目标优化方法进行了研究。根据实际工艺情况建立了CFRP/Ti叠层构件钻孔工艺参数多目标优化模型;结合钻孔工艺参数离散化的特点,基于帕累托占优原理构建了帕累托最优工艺参数集穷举搜索算法,并基于宽容分层序列法构建最优工艺参数决策算法;以T700-TDE85/Ti-6Al-4V叠层构件为对象,得到了最优工艺参数,验证了方法的可行性和有效性。     

电静液作动器多目标优化设计

电静液作动器(EHA)设计需要考虑质量、刚度、效率等指标的综合最优,而且这些目标之间通常相互矛盾,是一个多目标优化设计难题。提出一种针对EHA的多目标优化设计方法,给出了EHA质量、刚度、效率的评价模型,通过多目标粒子群优化算法对连杆长度、舵面初始角度、泵的排量等关键设计变量进行智能优化,获得帕累托优化设计集合,并进一步采用层次分析法对帕累托集合中的方案进行综合决策,按照给定的多目标间的权重,实现了方案的排序和EAH的优化设计。     

车身覆盖件冲压工艺多目标优化系统的集成与应用

采用有限元分析技术和优化技术，以实现车身覆盖件的冲压工艺多目标优化。定义了材料、工艺和几何3类主要设计变量，其交集为最佳设计变量空间。针对覆盖件的整体成形性，建立破裂、起皱和变形不足3类目标函数，采用增量有限元和全量有限元的整合策略，求解目标函数值。与其他算法相比，基于帕累托法的多目标遗传算法能求解最佳解空间的所有解，完全满足不同成形性之间的权衡需要。最后，以车身发动机罩外板为例，验证了该优化系统的应用。     

灰色关联分析法在覆盖件成形优化中的应用

由于汽车覆盖件塑性变形过程复杂,容易产生多种质量缺陷,最优成形工艺参数难以确定。现将正交试验设计方法、灰色关联分析法与冲压数值仿真相结合,针对拉延工艺参数进行优化设计,借助板料成形仿真软件Pam Stamp对某汽车翼子板拉延过程数值仿真。通过仿真数据分别计算单目标函数关联度系数,多目标函数关联度和自变量对理想质量目标平均关联度,从而全面衡量压边力、摩擦因子及拉延筋几何尺寸对翼子板拉延成形质量影响,确定最优拉延工艺参数组合,为汽车覆盖件成形工艺参数设计提供理论依据。     

单晶硅的电火花线切割过程建模与优化研究

单晶硅的电火花线切割过程中,切削效率与表面粗糙度是两个重要的目标,生产过程中希望提高单晶硅切割效率的同时保证工件表面质量。运用响应曲面法建立关于单晶硅电火花线切割加工过程中切割效率与表面粗糙度的目标函数;采用第二代基于强度帕累托进化算法(SPEA-Ⅱ)得出一组关于切割效率与表面粗糙度的帕累托最优解。在相同条件下将SPEA-Ⅱ得到的帕累托最优解与第二代非支配排序遗传算法(NSGA-Ⅱ)得到的解集进行对比,测试两种算法对目标函数的优化性能。     

面向智能移动终端关键制造过程质量建模及优化分析

针对目前智能移动终端制造过程存在产品质量一致性不高、影响产品质量指标因素不明确、质量分析智能化程度不高等问题，在智能移动终端SMT产线关键工艺机理研究基础上，关联产品质量与关键工艺参数，分析产品质量影响因素。以首道印锡工序为切入点，基于改进的k-means算法对SPI检测数据记录进行聚类分析，识别样本数据异常点。针对SPI检测异常数据基于Apriori算法构建产品质量与印锡工艺参数关联模型，以最小支持度和最小置信度为判断依据，分析产生异常质量指标的重要工艺因素。在此基础上，构建质量-工艺模型反向优化刮刀速度和压力、脱模速度和距离等印锡关键工艺参数，基于粒子群算法优化求解模型，在最优质量指标输出下最佳工艺参数动态组合，结果表明，当权重系数λ=0.1时，获得最佳工艺参数组合。     

RSM优化冲焊型液力变矩器叶片冲压工艺及回弹抑制

冲焊型液力变矩器性能与叶片形状密切相关，为提高叶片成形精度，减小回弹对性能的影响，以摩擦系数、模具间隙、冲压速度为试验因素，以叶片最大减薄率和最大回弹量为优化目标，基于优化拉丁超立方试验方法，通过Dynaform对涡轮叶片冲压过程及回弹变形进行模拟获得样本数据，研究不同工艺参数对叶片成形质量的影响，并得到关于优化目标的多项式回归响应面模型。结合多目标优化算法对冲压工艺进行优化求解，获得最优解集，选取最优冲压工艺参数组合。最后对所选优化工艺参数进行仿真验证，结果显示，优化后叶片成形质量良好，减薄率和回弹均得到优化。     

火箭贮箱箱底充液拉深成形工艺的多目标优化

充液拉深成形技术能实现大型贮箱箱底的整体成形,然而成形件的质量受到许多工艺参数的影响。针对大型贮箱整体箱底构件充液拉深成形的起皱和破裂缺陷,以预胀压力、液室压力、压边力、压边圈圆角半径等工艺参数为研究对象,建立多目标优化模型。对贮箱箱底成形过程进行模拟,在此基础上,使用拉丁超立方采样法获得样本数据。采用克里金插值法(Kriging)和径向基函数(Radial basis function, RBF)建立工艺参数和质量指标之间的代理模型。利用NSGA-III算法和粒子群算法(Particle swarm optimization, PSO)确定了贮箱箱底达到目标(壁厚减薄率最小、破裂趋势最小、法兰边起皱最小、起皱趋势最小)时的最优工艺参数。最后通过实验验证了方法的有效性和结果的准确性。     

融合GRU-Attention与鲸鱼算法的流程制造工艺参数云边联动优化

为解决流程制造工艺参数优化面临的多工序耦合模型构建复杂、多目标冲突分析困难、实时和准确性难以保障等问题，提出一种融合GRU-Attention与鲸鱼算法的流程制造工艺参数云边联动优化方法。设计了适用于多工序耦合生产的训练计算云边协同架构，通过设备边缘节点与云平台的高效协同，完成了预测模型和优化模型的云端训练，边缘端数据收集、模型下载和调用计算。在此基础上，建立了基于GRU-Attention多层神经网络的生产工艺质量预测模型，将输出质量指标作为适应度，调用鲸鱼算法对生产工艺参数进行全局寻优，获得不同工序最优工艺参数组合，实现流程生产不同工序加工质量的实时预测和综合优化。最后，以某流程制丝生产线为例进行了实验验证，结果表明，所提基于深度学习的云边联动方法可实现生产质量的综合动态优化，同时可降低工艺参数调控任务的完成时间。     

帕累托耦合遗传算法对车辆振动模型优化与动力学仿真

为研究车辆行驶产生振动严重问题,提高车辆座椅的舒适性,构造了5自由度车辆振动模型,推导出车辆行驶运动控制微分方程,采用帕累托耦合遗传算法对5自由度车辆振动模型的五目标函数进行优化.优化对象包括前轮速度、后轮速度、簧载质量和前轮的相对位移、簧载质量和后轮的相对位移及车辆座椅垂直方向加速度.在五目标函数优化处理后,找出最佳运动学仿真优化值,通过Matlab/Simulation软件进行动力学仿真,与其他优化方法进行对比.仿真结果显示,在同等条件下,帕累托耦合遗传算法优化后的车辆经过地面凸起障碍物时,车辆座椅垂直方向产生加速度峰值降低了50%,车辆抖动次数较少.帕累托耦合遗传算法对车辆振动模型优化后,车辆行驶过障碍物相对平稳,改善了车辆座椅在行驶过程中的舒适性.     

红外菲涅尔透镜的注射超声辅助成型优化

为了提升聚合物红外菲涅尔透镜的光学性能,以其表面微沟槽的成型质量为目标,提出了一种高效的注射超声辅助成型方法,并对工艺参数进行了综合质量优化。首先分析了超声振动对聚合物的加热和加压效应,设计了一套一模两腔的对比试验模具;接着以红外菲涅尔透镜的调制传递函数MTF和齿形平均高度h为优化质量目标,设计了四步骤的多目标优化流程,通过试验设计、基于BP神经网络的质量目标与注射工艺参数关系建模、基于NSGA-Ⅱ的多目标优化和试验验证进行工艺参数的综合优化。实验结果表明:该多目标优化流程具有很高的精度,MTF和h的平均预测误差MPE分别为4.16%和3.32%;注射超声辅助成型的菲涅尔透镜微沟槽具有更高的复制质量,其齿沟槽平均高度h增加了15.6%,且h值的波动量随着h值的增大而增大,MTF值受齿高h均匀性的影响大于齿高h对其的影响。     

Multiobjective optimization of injection molding parameters based on soft computing and variable complexity method

The objective of this study is to propose an intelligent methodology for efficiently optimizing the injection molding parameters when multiple constraints and multiple objectives are involved. Multiple objective functions reflecting the product quality, manufacturing cost and molding efficiency were constructed for the optimization model of injection molding parameters while multiple constraint functions reflecting the requirements of clients and the restrictions in the capacity of injection molding machines were established as well. A novel methodology integrating variable complexity methods (VCMs), constrained non-dominated sorted genetic algorithm (CNSGA), back propagation neural networks (BPNNs) and Moldflow analyses was put forward to locate the Pareto optimal solutions to the constrained multiobjective optimization problem. The VCMs enabled both the knowledge-based simplification of the optimization model and the variable-precision flow analyses of different injection molding parameter schemes. The Moldflow analyses were applied to collect the precise sample data for developing BPNNs and to fine-tune the Pareto-optimal solutions after the CNSGA-based optimization while the approximate BPNNs were utilized to efficiently compute the fitness of every individual during the evolution of CNSGA. The case study of optimizing the mold and process parameters for manufacturing mice with a compound-cavity mold demonstrated the feasibility and intelligence of proposed methodology.     

Multi-objective optimization of MIMO plastic injection molding process conditions based on particle swarm optimization

Determining optimal process parameter settings critically influences productivity, quality, and cost of production in the plastic injection molding industry. Selecting the proper process conditions for the injection molding process is treated as a multi-objective optimization problem, where different objectives, such as minimizing product weight, volumetric shrinkage, or flash present trade-off behaviors. As such, various optima may exist in the objective space. This paper presents the development of an experiment-based optimization system for the process parameter optimization of multiple-input multiple-output plastic injection molding process. The development integrates Taguchi’s parameter design method, neural networks based on PSO (PSONN model), multi-objective particle swarm optimization algorithm, engineering optimization concepts, and automatically search for the Pareto-optimal solutions for different objectives. According to the illustrative applications, the research results indicate that the proposed approach can effectively help engineers identify optimal process conditions and achieve competitive advantages of product quality and costs.     

基于BP-NSGA的注塑参数多目标智能优化设计

为获得成型性能最优的注塑参数设计方案,提出了基于BP神经网络和非支配排序遗传算法的注塑参数多目标优化方法。将注塑模结构尺寸参数和注塑工艺参数作为待优化的设计变量,建立了以高质量、低成本、高效率为优化目标的注塑参数优化设计模型。基于非支配排序遗传算法获取给定参数范围内的所有Pareto最优解,并通过建立多输入和多输出的BP神经网络来快速获得非支配排序遗传算法优化进程中所有个体的适应度值。开发了基于BP神经网络与非支配排序遗传算法集成的注塑参数智能优化设计系统,并通过鼠标注塑参数设计实例,验证了其适用性和有效性。     

基于降维的驾驶员侧约束系统高维目标优化

为解决将高维目标变为单目标优化时各子目标不能同时较优,而多目标算法直接用于高维目标优化时又存在难以找到一个有代表性的Pareto非劣解集问题,在某轿车驾驶员侧约束系统的优化过程中提出了乘员损伤准则与多目标算法协同优化的方法。在已有相关损伤准则基础上根据最新版的FMVSS 208和ECE R94法规提出了适合研究问题的损伤准则;以提出的损伤准则为媒介,将一个高维目标优化问题降为一个低维目标优化问题,通过灵敏度分析、实验设计、多项式近似模型筛选出优化设计变量并得到近似模型,用多目标算法NSGA-Ⅱ对近似模型进行计算得到Pareto非劣解集,将得到的Pareto非劣解集中的每个解代入损伤准则损伤值计算公式,升序排列得到各子目标同时较优而损伤值最小的优化解。最终的优化结果表明：该方法很好地解决了乘员约束系统的高维目标优化问题,优化效果明显。     

基于Pareto解集的多目标优化方法及其应用

针对传统多目标优化设计方法的弱点，基于Pareto概念，借助遗传算法所具有的并行搜索特性，引入群体排序技术、小生境技术求得多目标优化问题的Pareto解集，实现了先寻优后决策的求解模式。实际工程算例表明，该模式可同时获得多个Pareto最优解，据此决策能有效弱化设计人员先验知识不足的影响，因而较传统多目标优化方法更为实用有效。     

公差设计多目标模型及其粒子群优化算法研究

为解决成本一公差设计模型中忽视产品质量的问题，以新型的田口质量观和Pareto最优解集概念为基础，提出了一种公差设计多目标模型。该模型将加工成本和质量损失分别作为设计目标，并以统计法公差装配成功率为约束条件，获得了比极值公差法更加宽松的公差限。改进了传统的粒子群优化算法，利用Pareto最优性重新定义粒子，然后采用快速非支配排序技术进行粒子的适应度排序，使其能够有效地对多目标模型进行求解。该算法对具体工程实例求解时，一次运行就可求得令人满意的Pareto最优解集，设计者可以根据生产实际和市场需求从中进行选取。通过对求得的Pareto进行最优前沿的分析，可得到该类零件公差设计的特性，其结果验证了公差诒计的一船规徨．     

Multi-objectives optimal model of heavy equipment using improved Strength Pareto Evolutionary Algorithm

The problem of injection molding machine’s multi-objective optimization is very important. A triple-objective optimization model with the largest mould moving speed and injecting capacities and the smallest injecting power has been created. The optimized design constraints of the optimal model are summarized. The computational efficiency of Strength Pareto Evolutionary Algorithm (SPEA) is improved by using rough set-based support vector clustering method. The number of external stocks is reduced. The optimal Pareto solution is determined by eliminating the uncertainty in the artificial priority election. The multi-objective optimization of the HT1600X1N injection molding machine is taken as an example. The SPEA-RSVC-II which is the mixed algorithm of Strength Pareto Evolutionary Algorithm and Ro′ugh-based Support Vector Clustering is applied. It shows that the new method could accelerate the population clustering operation effectively and improves the efficiency of optimized calculation.     

A combined analytical hierarchical process (AHP) and Taguchi experimental design (TED) for plastic injection molding process settings

Injection molding process is without doubt a multi-objective process if processing time, productivity, effectiveness, and the multi-criteria quality of the product are taken into consideration. Process settings affect the degree by which these objectives are realized. This work suggests a new proposal for evaluating optimal process settings through the handling of the plastic injection molding process in the same approach as a traditional multi-objective multi-criteria process. In a sense, there are numerous objective functions including cooling time, volumetric shrinkage, warpage, sink marks, residual stresses, and various process settings including temperature, pressure, etc. Within the suggested proposal, the Taguchi experimental design is used to generate a balanced set of experiments to explore the process; then, the finite element software SIMPOE is used to evaluate the behavior of the injection molding at each experimental setting. Analytical hierarchical process is then employed for multiple comparisons of the objectives and experiments as such to give the overall objective weight for each process setting (experiment). Analysis of variance is then used to evaluate the significant factors and the optimal setting of the process. This technique proved effective to obtain compliance between process design and several common manufacturer preferences, although the considered part was not changed.     

Intelligent methods for the process parameter determination of plastic injection molding

Injection molding is one of the most widely used material processing methods in producing plastic products with complex geometries and high precision. The determination of process parameters is important in obtaining qualified products and maintaining product quality. This article reviews the recent studies and developments of the intelligent methods applied in the process parameter determination of injection molding. These intelligent methods are classified into three categories: Case-based reasoning methods, expert system- based methods, and data fitting and optimization methods. A framework of process parameter determination is proposed after comprehensive discussions. Finally, the conclusions and future research topics are discussed.