采用陶瓷刀片和CBN刀片干切削淬硬GCr15轴承钢, 测量了不同切削参数下切削后工件的表面粗糙度; 基于微粒群优化算法建立了表面粗糙度预测模型, 并与线性回归法建立的经验公式进行了比较; 用扫描电子显微镜观察了切屑形态。结果表明: 采用微粒群优化算法建立的表面粗糙度预测模型具有一定的可靠性, 与线性回归法相比, 能更精确地预测出加工工件的表面粗糙度; 切削参数中对表面粗糙度影响最大的是进给量, 其次是背吃刀量, 切削速度的影响最小; 锯齿状切屑能降低切削温度, 提高工件表面质量; 用陶瓷刀片和CBN刀片切削获得的最低表面粗糙度分别可达0.48 μm和0.56 μm。

The quenched GCr15 bearing steel was cut by ceramic and CBN blades, and the surface toughness of the steel cut with different parameters was measured. The surface toughness model was built on the base of particle swarm optimization (PSO) algorithm, and was compared with that getting from empirical formula built by linear regression method. The chip morphology was observed by scanning electron microscopy. The results show that the prediction model of surface roughness built by PSO algorithm was more reliable comparing with the model built by linear regression method, it could predict the surface roughness of the work piece more precisely. The dominant factor affecting surface roughness was the feed rate, the second was the depth of cut, and the effect of cutting speed was minimum. The saw-tooth chips could decrease the cutting temperature, and improve the surface quality of the work piece. The surface toughness obtained by ceramic and CBN blades cutting could reach 0.48 μm and 0.56 μm, respectively.