依据正交试验结果,利用线性回归分析方法,应用Matlab软件建立了砂带磨削镁合金AZ91D时表面粗糙度的预测模型,对该模型进行回归方程和回归系数显著性检验;并对该模型进行优化,即进一步提高模型的预测效果;然后对该模型进行了试验验证。预测表面粗糙度具有可靠性:砂带粒度和砂带线速度对表面粗糙度的影响很大;磨削压力对表面粗糙度有一定的影响;接触轮硬度和工件轴向进给速度对表面粗糙度的影响不明显。
Based on the results of orthogonal test, belt grinding surface roughness prediction model of AZ91D mag nesium alloy is established using linear regression method. The significant tests of regression equation and regres sion coefficient are conducted and the model was optimized to improve the prediction precision. The results indicate that the model has reliability on the prediction of surface roughness, abrasive belt grinding pressure has certain in fluence on the surface roughness, and grain size of belt and the belt linear speed have high significant influence on surface roughness, while the contact wheel hardness and workpiece axial on surface roughness. feeding speed have no significant influence