摘要
表面粗糙度是评价磨削加工质量的重要指标,表面粗糙度预测是实现表面粗糙度在线控制的前提。针对现有神经网络方法在预测外圆纵向磨削表面粗糙度方面存在的不足,提出了一种新的基于支持向量回归机的外圆纵向磨削表面粗糙度预测方法。在分析了影响外圆纵向磨削表面粗糙度预测主要因素的基础上,建立了基于支持向量回归机的外圆纵向磨削表面粗糙度预测模型。应用实例的仿真结果表明,所建立的预测模型具有较强的泛化能力和较高的预测精度。
Surface roughness is an important index of evaluating ability of grinding machining, and surface roughness prediction is the precondition of realizing online controlling for surface roughness. Aiming at the deficiencies of existing methods of neural network in surface roughness prediction for cylindrical longitudinal grinding, a new surface roughness prediction method for cylindrical longitudinal grinding is proposed based on support vector regression. Analyzes the major factors which affects surface roughness prediction,based on which the surface roughness prediction model for cylindrical longitudinal grinding based on support vector regression is proposed. The simulation results of application instance show that the prediction model has strong generalization ability and high prediction precision.
出处
《机械设计与制造》
北大核心
2016年第8期131-134,共4页
Machinery Design & Manufacture
基金
国家自然科学基金项目资助(1120278)
关键词
外圆纵向磨削
表面粗糙度
支持向量机
支持向量回归机
Cylindrical Longitudinal Grinding
Surface Roughness
Support Vector Machine
Support Vector Regr-ession
