薄板装配工艺系统可靠性建模与分析

提出一种计入销/孔(槽)公差面向产品质量的薄板装配工艺系统可靠性建模方法。综合考虑定位销公差、零件孔(槽)公差和定位销磨损量,构建薄板装配偏差统计数字特征模型;分析推导定位销过程磨损模型,探讨装配质量与工艺系统可靠性之间的关系,综合考虑定位销故障率和过程磨损量对工艺系统可靠性的影响,建立工艺系统结构可靠性模型和面向装配质量的可靠性模型,继而形成薄板装配工艺系统可靠性建模方法;以车身侧板装配为例,应用提出的建模方法分析车身侧板装配工艺系统可靠性。结果表明：定位销磨损、夹具布局和销/孔(槽)公差是影响车身侧板装配工艺系统可靠性的重要因素。该方法为薄板装配工艺系统可靠性研究提供了一种新的思路。     

车身多工位装配系统可靠性评估与维护策略研究

针对车身多工位装配系统夹具的衰退过程,提出了夹具元件磨损、来料零件偏差以及定位元件配合公差等多因素集成影响下的车身装配系统可靠性评估方法。在给定车身波动阈值条件下,提出了基于可靠性模型的多工位夹具部件的动态维护策略。通过一个四工位薄板件装配案例对所提方法进行了应用验证,为夹具系统维护与车身尺寸质量控制提供了理论指导。     

多工位装配过程夹具系统公差和维护综合优化设计

提出一种面向二维多工位装配过程、综合考虑装配夹具系统全寿命周期成本、产品零件孔制造成本和产品质量损失成本的公差和维护综合优化方法。分析多工位装配尺寸偏差传递关系,建立多工位装配过程产品质量损失模型。然后根据4-2-1夹具定位原则,构建考虑夹具磨损过程损失的夹具定位销副偏差统计数字特征模型。继而发展了以夹具系统全寿命周期成本、零件孔制造成本和和产品质量损失成本为装配总成本最小化的定位销公差、零件孔公差与更换周期优化模型。以汽车侧围装配过程为例,分别研究定位销公差、零件孔公差、定位销更换周期、配合间隙、平均磨损率和磨损率方差对装配总成本的影响,并优化设计定位销公差、零件孔公差和定位销更换周期。所提出的综合优化设计方法比采用定位销等公差设计、零件孔等公差设计、定位销与零件孔等公差设计和定周期更换设计的装配总成本分别减少了16.25%、11.31%、39.93%和13.54%。该方法为产品装配夹具系统高质量低成本设计提供了一种新的途径。     

动态多工位车身串并联混合装配系统可靠性评估研究

针对当前车身装配系统可靠性评估体系对偏差因素、可靠性要素间的相关性、过程衰退等方面的评价不完备的问题,对定位销定位偏差源的组成结构、结合来料零件定位孔、槽的尺寸质量和定位销本身定位能力水平的定位销磨损模型进行了研究。分析了系统要素可靠性间的相关性,根据Copula连接函数理论建立了相关性模型;对产品上所有关键产品特性(KPC)点在整体产品尺寸质量可靠性中所占权重进行了分配并建模,基于来料-产出流思想对串并联混合系统进行了简化,最终提出了一种综合考虑各偏差源动态输入的动态多工位车身串并联混合装配系统可靠性评估模型;利用此模型对车身某分总成装配系统进行了制造成本优化分配,并且给出了一套动周期预防性维护策略。研究结果表明:该可靠性模型精确度较高,适用于指导生产制造成本的优化和建立预防性维护策略。     

基于偏差传递的二维多工位装配夹具系统公差可行稳健设计

提出一种基于偏差传递的二维多工位装配夹具系统公差可行稳健设计方法。分析二维多工位装配尺寸偏差传递关系,建立由定位销和零件定位孔(槽)公差引起夹具定位偏差的偏差流状态空间模型。采用数论网格方法对定位销和零件定位孔(槽)公差进行采样,将得到的公差样本代入夹具定位偏差模型,求得夹具定位偏差样本空间,将夹具定位偏差作为状态空间模型的输入偏差,提出基于状态空间模型的二维多工位装配成功率计算方法。继而应用Taguchi正交试验直观分析方法,分析得到影响装配成功率的主要因素即夹具系统定位销副关键公差,运用回归正交组合设计拟合得到二维多工位装配成功率与夹具系统定位销副关键公差的响应面模型。以定位销、零件定位孔(槽)制造成本所构成的装配成本最小化为目标,二维多工位装配成功率为约束,建立二维多工位装配夹具系统定位销和零件定位孔(槽)公差可行稳健设计模型。以汽车车身地板二维三工位装配为例,建立其公差可行稳健设计模型并对该模型进行计算与分析,结果表明在装配成本增幅较小的情况下,采用稳健约束后可显著提高公差设计的可行稳健性。该方法为二维多工位装配夹具系统公差稳健设计提供了一种新途径。     

基于敏感度分析的多工位薄板装配过程测点优化设计

提出了一种多工位薄板装配过程测点优化设计方法。该方法利用薄板装配偏差流传递的状态空间模型确定优化设计的目标函数,采用遗传算法实现测点在多工位间坐标位置的优化设计。装配过程实验表明,该优化方法显著提高了测量数据对夹具定位偏差反应的敏感度,有助于在夹具定位偏差产生的早期发现问题,避免因夹具定位偏差引起的产品质量缺陷。     

车身柔性虚拟装配偏差模拟分析

在轿车样车试制阶段,为了减少车身尺寸功能评估周期和降低评估成本,引入了车身的虚拟装配.针对功能评估中虚拟装配关键技术,包括装配过程夹具偏差、焊接处理及多工位装配等提出解决方案,实现了在考虑制造偏差、夹具偏差、连接偏差和多工位装配基础上的车身虚拟装配.通过案例论证了提出的虚拟装配方法的可行性.结果表明,该模型的准确度较高,适用于柔性薄板零件的装配偏差分析.     

模式相关性分析在车身多偏差源诊断中的研究

针对车身装配过程中多偏差源诊断难题,提出了基于工位间多源检测数据融合分析的多偏差源诊断方法。通过对不同工位、不同设备及检测频次的偏差数据的综合利用与主成分分析的基础上,研究了零部件与总成之间偏差模式的映射方法,实现了上游车身零部件尺寸质量的诊断;根据工装CAD信息等提取夹具失效模式向量,构建各夹具失效模式对车身总偏差的贡献度评价指标,获得了各夹具元件的失效系数,实现了工位内夹具失效的诊断。经案例验证,所提出的诊断方法具有较高的准确性和实用性。     

基于数论网格法(NT-net)的二维多工位装配成功率计算方法与实例

提出一种基于数论网格法(NT-net)的二维多工位装配成功率计算方法。首先分析了数论网格法的偏差,描述了数论网格法产生glp集的原理。然后应用数论网格法对夹具定位偏差进行采样,将其代入二维多工位装配尺寸偏差传递状态空间模型中,计算得到二维多工位装配输出偏差。将计算结果和零件上测点允许偏差进行比较,统计出合格样本数,再将其除以总样本数,计算得到二维多工位装配成功率。最后,以车身地板装配为实例,在构建了车身地板二维三工位装配偏差传递模型基础上,求解其偏差传递矩阵,再运用数论网格法计算其装配成功率,并采用蒙特卡洛模拟法进行了验证和分析,结果表明了该方法的有效性。这为产品二维多工位装配成功率的预测提供了一种新的途径。     

一种白车身多工位装配偏差源诊断方法

针对白车身多工位装配偏差源耦合复杂,诊断困难的问题,提出一种新的多工位装配偏差源诊断方法。基于多工位装配偏差流传递的状态空间模型,将连续的装配过程以检测站为节点分解成不同的装配模块;引入"两种质量诊断理论",通过对比分析装配模块的计算机仿真装配偏差和实际装配偏差,获取装配模块"分质量",将偏差源定位到具体模块,降低状态空间模型状态变量维数,实现偏差源的快速诊断。最后通过实例验证了方法的有效性。     

FLEXIBLE ASSEMBLY FIXTURING LAYOUT MODELING AND OPTIMIZATION BASED ON GENETIC ALGORITHM

There are many welding fixture layout design problems of flexible parts in body-in-white assembly process, which directly cause body assemble variation. The fixture layout design quality is mainly influenced by the position and quantity of fixture locators and clamps. A general analysis model of flexible assembles deformation caused by fixture is set up based on "N-2-1" locating principle, in which the locator and clamper are treated as the same fixture layout elements. An analysis model for the flexible part deformation in fixturing is set up in order to obtain the optimization object function and constraints accordingly. The final fixture element layout could be obtained through global optimal research by using improved genetic algorithm, which effectively decreases fixture elements layout influence on flexible assembles deformation.     

A value-based maintenance optimization method for failure prevention based on reliability modeling of a hybrid assembly system

The effective maintenance policy for the fixtures of the assembly system will guarantee good process capability and product quality to a great extent. Traditional maintenance schedules for the assembly system, such as a constant reliability index as the maintenance trigger, are always based on the fixed maintenance threshold in engineering practice. It always results in down time of the process, unqualified products or the excessive maintenance costs. The aim of this paper is to present a method to optimize the maintenance schedules for a serial-parallel hybrid assembly system. Based on the production chain definition of the hybrid system, a reliability evaluation method by integrating the process performance of the hybrid assembly system and product qualities is proposed. On the basis of the reliability method, a value-based preventive maintenance optimization policy for the fixture components is presented to prevent system failures. Compared with the traditional reliability threshold method for maintenance, the proposed maintenance method generates more value on the premise that the system status and product quality are guaranteed. A case study is used to illustrate the proposed method and validate the effectiveness and advantages.     

Simultaneous optimization of fixture and joint positions for non-rigid sheet metal assembly

This paper presents an optimization methodology for non-rigid sheet metal assembly variation by considering part variation, fixture variation, fixture layout, and joint positions, as well as the assembly spring back. The proposed algorithm integrates the finite element analysis (FEA) with a powerful global optimization algorithm, called the mode-pursuing sampling (MPS) method to simultaneously search for the optimal fixture and joint positions in order to minimize the assembly variation. An example application study is presented to demonstrate the optimization procedure and its effectiveness.     

基于偏差流分析法（SOVA）的二维多工位装配顺序规划的研究

针对目前产品装配顺序通常从装配时间和装配成本角度来考虑而难以保证装配质量的情形,应用偏差流分析法,以提高装配质量为目标,提出一种新的二维多工位装配顺序优化方法。以多工位装配过程为研究对象,分析影响装配质量的偏差源,建立偏差及其传递的数学模型,应用装配稳健度概念来评价装配工艺流程方案的优劣。以汽车侧围的4个零件装配为例,运用装配稳健度分析、比较了三种不同的装配顺序,得到了较优的装配方案。实例分析和仿真表明,应用装配稳健度可以分析选择装配顺序方案,为优化设计多工位装配工艺流程提供了新的途径。     

用GAAA优化多阶段装配过程中的夹具布局

改进了遗传算法与蚁群算法的融合（GAAA）算法,利用它来解决多阶段装配过程中二维刚性零件的夹具布局优化问题,合理选择定位销的位置使得灵敏度指标最小化。通过改变遗传算法的变异算子,变异长度以及交叉、变异在蚁群算法中发生的位置,提高了GAAA的稳定性和收敛性。以汽车侧边装配为例验证了改进算法的有效性,结果表明改进后的GAAA比基本的GAAA和蚁群算法求得的结果要好,且收敛速度更快,稳定性更好。     

Multi-objective optimization design of a fixture layout considering locator displacement and force–deformation

The fixture determines the workpiece position in a machining process; therefore, an increasing amount of attention has been given to fixture layout design. While machining, the workpiece position is affected by two major sources: (a) the locator displacement and (b) the force–deformation of the workpiece–fixture system. In the beginning of this paper, a geometric model considering the shape of a locator is developed to analyze the location performance, followed by the presentation of a simplified solving method and a location layout performance index. Second, to complete the force–deformation analysis, a finite element method-based force–deformation model is built and accelerated by a new method with a lower computer memory cost. Based on these two models, multiple objects of fixture layout optimization problems are proposed, and a multi-objective genetic algorithm-based optimization method is constructed. Finally, testing examples are approved to examine the validity of the method represented in this paper. These methods can provide a more accurate prediction of the locating performance in more widely used cases, and they have faster calculating speeds with lower computer memory costs.