基于Bootstrap方法的在线检测系统不确定度分析

On-line Detection System Uncertainty Analysis Based on Bootstrap Method

在线检测系统常用于数控加工零件后的精度检测,检测过程必然存在检测误差。但是在经典的误差理论中,由于被测量自身定义和测量手段的不完善,使得真值不可知,造成严格意义上的测量误差不可求。针对这一问题,提出一种基于Bootstrap方法的在线检测系统不确定度A类评定方法。实验结果表明,经过误差补偿后的在线检测系统,其检测精度和标准不确定度有了较大幅度的改善。

On-line detection system is often used for part detection after numerical control machining,and the inspection process must exist errors.But in the classical error theory,because of its definition and measurement method is not perfect,true value is not known and measurement error can not be calculated.To this problem,a class A uncertainty evaluation method of on-line detection system is proposed based on the Bootstrap method,the research on error compensation for detection system is carried out to improve the precision of on-line inspection system.The results show that the accuracy and the standard uncertainty of the on-line detection system developed by ourselves have been improved greatly after error compensation.