设计与制造中的统计几何公差建模及应用

几何与尺寸公差(形位公差GD&T)广泛用于机械工程设计与制造中重要几何特征的偏差控制.相对较成熟和简单的尺度公差建模与分析,几何公差统计建模与分析更具挑战性,是当前CAD技术中尚未但亟待解决的课题之一.现提出采用统计几何模态模型(SMA)方法解决这一问题.SMA可识别刻划与解释测量数据中的几何特征信号模式及其变化,从而服务于制造中的(公差)质量检验、诊断及变化模式的统计建模.在设计中SMA可进行模态重组综合,从而再现或仿真几何特征偏差的随机变化进行统计几何公差分析.

A statistical GD&T tolerance model for design and manufacturing

Geometric and Dimensional Tolerancing （GD＆T） is a critical tool for geometric variation control in both design and manufacturing. Due to uncertainties in manufacturing any designed feature cannot be perfectly manufactured. The impact of the variations on critical features to product performances and functions shall be fully understood and controlled. This is accomplished, in dimensional tolerancing, by modeling and replicating the random uncertainties for tolerance design and by monitoring/controlling the variations in manufacturing. Unfortunately, neither available technique nor emerging method is capable of resolving the challenges in statistical GD~-T toleraneing. To this end, a mode-based mathematic method is developed for statistical GD＆T tolerancing. It allows tolerance control by characterizing/monitoring variation signals in manufacturing and by conveying/composing the variation patterns in GD＆T tolerance design simulation. Case studies are provided for validation and demonstration.