基于佳点集的装配公差分析方法

Assembly Tolerance Analysis Method Based on Good Point-Set

为提高传统的蒙特卡罗模拟方法进行公差数值分析时计算结果的收敛速度,基于数论中的佳点集理论构造出一种新的准蒙特卡罗法(quasi-Monte Carlo method)。在此基础上建立了基于确定性定位分析的白车身翼子板装配三维偏差模型,分别采用传统的蒙特卡罗伪随机数序列和基于佳点集的准蒙特卡罗法序列对该模型进行公差数值分析。在理论分析基础上的结果表明,在进行高精度复杂机械产品的装配公差分析过程中,与传统的蒙特卡罗模拟方法相比,所提出的方法的计算结果所需要的统计样本较少并且精度较高。

A quasi-Monte Carlo method based on good point set of number theory isproposed in order to improvethe convergencerate of calculation results of traditional Monte-Carlo method to support assembly tolerance numerical analysis. Furthera 3-D assembly variation model of fender assembly based on deterministic locating is established. Respectively the traditional Monte-Carlo pseudo-random number sequences and quasi-Monte Carlo sequences of good points are used to support assembly tolerance numerical analysis of the model. Based on theoretical analysis, The result shows that the proposed method provides analysis results with less statistical samples and higher precision compared with Monte-Carlosimulation methodin the process of tolerance design in the manufacturing process of high accuracy complex products.