基于PLS的高速电主轴热变形建模与分析

Thermal Deformation Modeling and Analysis of High Speed Motorized Spindle Based on PLS

目的提出偏最小二乘方法(Partial Least Squares,PLS),对不同工况下的电主轴热变形进行预测,并分析多个温度变量和三维热变形的相关关系.方法以型号为150MD24Z7.5的电主轴为研究对象,采用精密传感器测量稳态条件下电主轴的热变形和温升数据,根据PLS模型内部分析机理,利用提取的主成分对变量的解释能力、精度及变量整体相关关系进行分析.结果分析结果验证了自变量温升之间存在多重相关性,采用PLS方法建模有益于模型精度的提高,模型提取的主成分能够反应原始变量的大多数信息,自变量温升和因变量热变形之间存在明显的线性相关关系,PLS模型的预测精度优于多元回归模型.结论电主轴热变形的PLS模型预测精度较高,适用于实际加工工况,对热误差的补偿具有补充意义.

This paper was proposed to model the thermal deformation of the motorized spindle at different operating condition and analysis the correlation between multiple temperature variables and three-dimensional thermal deformation based on the partial least square method.A precision sensor is used to measure the thermal deformation and temperature rise of the 150MD24Z7.5 motorized spindle under steady-state conditions.According to the internal analysis mechanism of the PLS model,the extracted principal components are used to analyze the interpretability,accuracy and overall correlation of the variables.The model verifies the multiple correlations between the temperature rises of independent variables.The principal components extracted by the model can reflect most of the information of the original variables.There is a clear linear correlation between the temperature rise of the independent variable and the thermal deformation of the dependent variable.The prediction accuracy of the PLS model is superior to the multiple regression model.The PLS model of the thermal deformation of the motorized spindle has high prediction accuracy,which is suitable for actual processing conditions and has supplementary significance for the compensation of thermal errors.