论文部分内容阅读
按照惯例,产品设计师根据设计要求设定尺寸及其它变量的标称值。在确定公差时,有两种截然不同的思想方法(见参考文献1,2,3): 1.保守的设计师采用“最坏情况”法,即算术法。 2.热衷于统计的人主张采用“勾股定理”法,即几何分析法。因为这些公差影响很大,所以这个问题应引起足够的重视。统计分析,顾名思义,需要充分熟悉统计学定律。而这一点正是统计分析方法尚未得到普遍采纳的原因。本文叙述了一种公差分析方法——蒙特卡罗模拟法。此方法受到统计学家的极大重视,而且也容易为产品设计师所理解。这种分析是由于微计算机技术的发展而成为可能的,因为这些发展使我们手头有了能够进行高速数字运算的相当有效的台式计算机。作者将借助几个实例说明蒙特卡罗模拟法是如何用于设计的最佳化处理的。
By convention, product designers set the nominal values for dimensions and other variables based on design requirements. There are two very different ways of thinking when determining tolerances (see References 1, 2, and 3): 1. Conservative designers use the “worst case” method, the arithmetic method. 2. People keen on statistics advocate the use of “Pythagorean Theorem” method, that is, geometric analysis. Because of these tolerances have a great impact, so this issue should cause enough attention. Statistical analysis, as the name suggests, requires full familiarity with the law of statistics. And this is exactly why statistical analysis has not been universally adopted. This article describes a tolerance analysis method - Monte Carlo simulation. This method has received tremendous attention from statisticians and is also easily understood by product designers. This analysis is made possible by the advances in microcomputer technology that have enabled us to have on hand a reasonably efficient desktop computer capable of high-speed digital operations. The authors will illustrate several examples of how Monte Carlo simulation can be used to optimize the design.