摘要
在数控机床热误差控制补偿技术中,合理选择温度变量,是建立补偿模型的关键所在。先采用模糊聚类对温度变量进行分组,然后根据温度变量与热变形的相关性在每组中选择典型的温度变量,最后根据回归分析中经过修正的相关系数选择测温点。以XH718数控机床为实验对象,结果表明,此方法既能减少测温点,又能保证模型精度。
In the technology of thermal error compensation in NC machine tool, selecting the temperature variables rationally is key for setting the model of compensation. The temperature variables are grouped by the fuzzy group analysis. Then, according to the degree of correlation of the temperature variables and the thermal errors, one representative is selected from each group. Last, the correlation coefficient revised in regression analysis are adopted to select the temperature measurement points. Through experiments on NC machine tool XH718, it is showed that the method can reduced the numbers of temperature measurement points, also assure accuracy of the model.
出处
《机械》
2006年第5期37-39,共3页
Machinery
基金
成都市科技局项目资助(O5HJDZ299)
关键词
数控机床
热变形
模糊聚类
相关系数
测温点
NC machine tool
thermal deformation
fuzzy group
correlation coefficient
temperature measurement point
