多元非线性回归的球头立铣刀几何参数优化

Optimization of geometric parameters of ball-end cutter based on the multiple nonlinear regression

研究球头立铣刀几何参数优化问题。由于球头立铣刀的几何角度对其切削性能有着重要的影响,因此选择合适的几何参数十分重要。为改善铣刀优化设计提出了以前角、螺旋角和一次后角为试验因素的多元非线性铣削力预报模型的建模方法。通过有限元仿真分析,获得了正交试验数据样本,利用回归技术对其进行分析,确定了铣削力预报模型,采用遗传算法对铣刀几何角度进行优化,通过MATLAB软件仿真最终得到了铣削力最小时的最优几何参数值。结果表明:所选几何参数与铣削力确实有非线性回归关系,且优化后的几何参数组合所对应的铣削力比正交组合试验方案所得铣削力最少降低了2.02%,铣削温度最多降低了83℃。

Research on the optimization of geometric parameters of ball-end cutter. Due to the geometric parameters of the ball-end cutter has an important effect on the milling performance, so the choice of geometrical parameters is very important. To improve the situation,proposed a new method to establish the cutting force regression equation between the cutting force and the experi- mental factors （rake angle.relief angle,helical angle） ,which is obtained with multivariate nonlinear regression method. Orthogo- nal experiment data sample is obtained through FEM analysis. The data sample is analyzed with regression techniques. Desired ge- ometric parameters are obtained with the use of the Genetic Algorithm （GA） tool box of MATLAB. The results showed that： se- lected geometric parameters and milling force do have a nonlinear regression relationship, and compared to the former situation, the milling force of the optimized combination of geometric parameters decreased by 2.02%, milling temperature decreased by 83℃ at most.