基于低偏差序列样本的装配公差分析方法  

Assembly tolerance analysis method based on low discrepancy sequences sample

作　　者：张岩[1] 莫蓉[1] 

机构地区：[1]西北工业大学现代设计与集成制造技术教育部重点实验室,陕西西安710072

出　　处：《计算机集成制造系统》2014年第3期579-585,共7页Computer Integrated Manufacturing Systems

基　　金：国家科技支撑计划资助项目(2012BAF10B09);国家自然科学基金资助项目(51375395)~~

摘　　要：传统Monte Carlo分析方法在进行公差数值分析及装配成功率估算时需要大量的统计样本支撑,无法适应高精度复杂产品研制过程中公差设计多组成环、高置信度、高精度、多迭代、短周期等要求。针对该问题,引入低偏差序列代替Monte Carlo方法中的伪随机数序列,并将其融入公差数值分析模型和装配成功率估算模型中,进而构建了基于低偏差序列样本的公差数值分析方法与装配成功率估算方法。在理论分析的基础上,通过对某型飞机典型组件的实例分析,验证了该方法的有效性。The traditional Monte Carlo （MC） analysis method need large numbers of statistic samples to support assembly tolerance numerical analysis and success rate estimation,which could not adapt to the demand of tolerance design in the manufacturing process of high accuracy complex products.Aiming at this problem,the Low Discrepancy Sequences （LDS） was introduced into MC method analysis process to replace the pseudo random number sequences and integrated into the tolerance numerical analysis model and assembly success rate estimation model to construct the LDS based analysis method.The effectiveness of the proposed method was verified by analyzing the application of an aircraft typical component.

关 键 词：公差分析 装配成功率 低偏差序列 蒙特卡罗方法 

分 类 号：TH16[机械工程—机械制造及自动化] TP391[自动化与计算机技术—计算机应用技术]