【摘要】 磨削表面质量的优劣直接影响零件的性能与使用寿命,其主要由表面粗糙度表征。运用最小二乘支持向量机预测理论与参数优化算法建立了高速磨削表面粗糙度预测模型,搭建了磨削闭环系统。通过实验分析得出,对试样内工艺参数组合下表面粗糙度进行预测时,平均相对误差为MRE=0.0095,均方根误差为MSE=0.0050;对试样外工艺参数组合下表面粗糙度进行预测时,平均相对误差为MRE=0.0119,均方根误差为MSE=0.0054。高速磨削表面粗糙度预测引导了磨削参数的设定,形成了磨削过程的闭环反馈控制,提高了高速磨削加工的自动化水平,磨削精度、效率,同时降低了磨削的废品率。更多还原

【Abstract】 The surface quality of grinding directly affects the performance and service life of the workpiece,which is mainly characterized by surface roughness.Based on the least squares support vector machines theoryand the parameter optimization algorithm,a prediction model of surface roughness for high-speed grinding is established,and a closed loop grinding system is built.The experimental results show that the average relative error(MRE)equals 0.0095 and the root mean square error(MSE)equals 0.0050 when the surface roughness is predicted by the combination of process parameters in the samples;The average relative error equals 0.0119 and the root mean square error equals 0.0054 when the surface roughness is predicted by the process parameters beyond the samples. Thesurface roughness prediction of high speed grinding guidesthe setting of grinding parameters,and forms a closed loop feedback control system of grinding process,and improves the level of automation of high-speed grinding,grinding precision,grinding efficiency,and reduce the reject rate.更多还原