接触式轴孔内径在线测量中测头系统误差分析  

System error analysis of probe in on line contact measurement of the axis hole diameter

作　　者：何宇[1] 雷玉勇[1] 鄢烈忠 唐炼[1] 曾修文 

机构地区：[1]西华大学机械工程学院,四川成都610039

出　　处：《现代机械》2017年第5期35-39,共5页Modern Machinery

基　　金：四川省科技厅项目资助(2014JY0173);西华大学研究生创新基金资助(ycjj2016033)

摘　　要：为提高轴孔内径在线测量精度,分析了接触式测量过程中,测头半径误差与测头预行程误差对测量结果的影响以及其补偿方法。针对测量过程中测头半径误差,通过计算被测点的法矢,可进行有效的测头半径误差补偿;提出了基于BP神经网络的预行程误差预测模型,根据标准球实际测得的预行程误差,在对所建立的神经网络训练的基础上,对触发式测头在不同法矢上的预行程误差进行预测,并将预测结果与标准球相应的实际测得的结果进行比较,比较结果验证了该预行程误差预测模型的有效性。In order to improve the accuracy of shaft hole diameter online measurement,the influence of probe radius error and probe pre-travel error on measurement results was analyzed in the process of contact measurement. In view of the probe radius error in measuring process,by calculating the normal vector of the measured point,the radius error of the probe can be compensated effectively. Pre-travel error prediction model based on BP neural network was proposed. According to the actual pre-travel error of standard ball,the pre-travel errors of trigger probe on the different normal vector were predicted based on the training of the neural network. The predicted results are compared with the actual measured results of the standard ball,and the validity of the prediction model is verified.

关 键 词：接触式测量 测头半径误差 测头预行程误差 BP神经网络 

分 类 号：TP69[自动化与计算机技术—控制理论与控制工程]