硬脆材料点磨削表面硬度多元回归数值模拟研究  

The simulation of multiple regression on surface hardness in point grinding hardness-brittle materials

作　　者：马廉洁[1,2] 李琛[2] 巩亚东[1] 曹小兵[2] 陈小辉[2] 

机构地区：[1]东北大学机械工程与自动化学院,沈阳110819 [2]东北大学秦皇岛分校控制工程学院,秦皇岛066004

出　　处：《现代制造工程》2015年第8期63-66,共4页Modern Manufacturing Engineering

基　　金：国家自然科学基金资助项目(51275083)

摘　　要：通过低膨胀微晶玻璃点磨削试验,分析了磨削速度、进给速度及磨削深度对表面显微硬度影响的变化趋势,并对原始数据进行换元处理,使得新变量x1、x2、x3与表面显微硬度近似呈线性关系。在换元处理的基础上,基于最小二乘法建立了表面显微硬度多元回归预测模型,并进行了模型检验和试验验证。F检验结果表明,回归模型和回归系数具有较高的显著性;相关系数检验结果表明,模型与各因素之间高度相关。模型计算值与试验值基本吻合,说明模型具有较高的可靠度。By the point grinding experiment of low-expansion glass ceramics, the trend of grinding speed, feed speed and grinding depth on the surface microhardness are analyzed. The relationship between new various xl ,x2 ,x3 and surface microhardness was linear while the origin data was dealed with conversion various technology. Then, base on the least squares method, the surface hardness of the multiple regression predict model was builded. The model test and experimental verification were carried out. F- test results showed that the regression model and regression coefficients were high significance. Correlation coefficient test results showed it was high correlation between the model and influence factors. Calculated values of model were consistent with the exper- imental values. It indicated that model has high reliability.

关 键 词：多元回归 点磨削 表面显微硬度 低膨胀微晶玻璃 

分 类 号：TH161[机械工程—机械制造及自动化]