影响激光近净成形残余应力的关键参数的多元回归分析

Multiple Regression Analysis of the Key Parameters of Laser Engineering Net Shaping on Residual Stress

激光近净成形的激光功率、扫描速度和分层高度是成形的关键参数。掌握关键参数与影响成形质量和变形的残余应力的关系是降低残余应力与选择合适成形参数的有效途径。利用生死单元技术对钛合金激光近净成形过程进行热力耦合数值求解,得到8组不同关键参数情况下有限元模型的残余应力分布云图,并对顶层三个位置的残余应力进行采样。对不同参数与残余应力做多元回归分析,得到多元回归方程。通过有限元分析和实验验证多元回归方程的正确性,证明多元回归方法是减少不同参数下成形件残余应力的预测与实验时间、对既定参数残余应力评估的有效方法。

Key forming parameters of laser engineering net shaping are laser power, scanning speed and stratification height. The relationship between the key parameters and residual stress must be mastered, because the residual stress influences the forming quality and deformation. Mastering the relationship is an effective way to reduce residual stress and to choose forming parameters. Using the life and death element technology and thermo-mechanical coupled field model, 8 groups of different key parameters of the laser engineering net shaping of titanium alloy are computed. The distribution cloud of the residual stress is obtained and three samples are selected. Through the multivariate regression of different key parameters and the residual stress, the multiple regression equation is obtained. The validity of the multiple regression equation is proved by finite element analysis and experiment. The multiple regression analysis is the effective way to reduce the time of simulation and experience of the different key parameters and to estimate the residual stress of the established parameters.