摘要
应用RBF神经网络建立了高速铣削模具型腔时已加工表面粗糙度的预测模型,预测值与实测值非常接近,预测精度略高于回归模型的精度。利用该模型对高速铣削表面粗糙度进行了预报,并分析了工艺参数的影响规律,验证了模型对质量监测及工艺参数优化的可行性及实用性。结果表明,通过合理选择工艺参数,尤其在控制切削深度和切削宽度的情况下,可获得Ra0.3μm以下的已加工表面粗糙度。
The predictive model of surface roughness in highspeed milling of mold cavity was developed based on the RBF artificial neural network method. The predictive results agree very well with those obtained from experiments and the predictive accuracy is slightly higher than that of regression model Using the forecast model, the surface roughness in high-speed milling was forecast and the influence of process parame- ters was analyzed, which verifies the feasibility and practicability of the model on the quality monitoring and process parameter optimization. It indicates that the surface roughness less than 0. 3~m can be achieved on condition that all machining parameters, especially milling depth and milling width are well selected.
出处
《组合机床与自动化加工技术》
北大核心
2013年第6期6-8,共3页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金资助项目(61006075)
广东省教育部产学研结合项目(2011B090400253)
关键词
高速铣削
表面粗糙度
RBF神经网络
预测模型
high-speed milling
surface roughness
RBF neural network
predictive model
作者简介
陈英俊（1962-），男，广东潮州人，肇庆学院电子信息与机电工程学院教师，副教授，主要从事机电产品设计与制造方面的科研与教学工作，（E-mail）chenyj1228@sina．com。
