基于灰色关联和模糊聚类的机床温度测点优化

Optimization of Temperature Measuring Points for Machine Tools Based on Grey Correlation and Fuzzy Clustering Analysis

针对机床热误差补偿技术中温度测点优化选择的问题,提出采用基于灰色关联分析和模糊聚类分析相结合的方法对机床温度测点进行优化选择。采用灰色关联分析法计算温度变量与主轴热误差之间的相关系数,并据此优选温度变量,采用模糊聚类分析法对所选择的温度变量进行聚类,确定关键温度变量,结合关键温度变量建立热误差线性回归模型。在精密卧式加工中心MCH63上对该方法进行了试验验证,结果表明,温度测点的数量由29个减少到6个,机床轴向热误差由41.3μm减小到7.6μm。

A method combining grey correlation analysis and fuzzy clustering analysis was used to optimize the temperature measuring points of thermal error compensation for machine tools. Grey cor relation coefficient between thermal errors and the temperature measuring points was computed by grey correlation analysis. The selected temperature measuring points optimized by grey correlation co- efficient were classified by fuzzy clustering analysis to determine the key temperature variables. I.inear regression model of thermal error was established with the key temperature variables. The method was tested in precision horizontal machining center MCH63. The results show that the number of tem- perature measuring points reduces from 29 to 6, the axial thermal error reduces from 41.3/,m to 7.6μm.