零件表面粗糙度的在线检测方法研究  

Research on On-line Detecting Method for Part Surface Roughness

作　　者：陈爱弟[1] 王信义[1] 

机构地区：[1]北京理工大学机械工程与自动化学院

出　　处：《中国机械工程》2002年第6期494-496,523,共4页China Mechanical Engineering

基　　金：国防科技基金预研资助项目 (96J18.3 .1BQ0 14 7)

摘　　要：根据零件表面粗糙度形成的复杂性 ,建立了基于具有结构和参数学习的模糊神经网络逻辑系统 ,通过该模糊神经网络搜索最优推理规则 ,并且通过最优模糊推理规则来在线检测零件表面粗糙度。According to the complexity of part surface roughness,a fuzzy neural network based logic(FNNL) system, which has learning function of structure and parameter is established. The system can search optimum reasoning rules for on-line detecting surface roughness.The Scheme of on-line detecting method can be used for additional surface roughness detecting, and the detecting results are identical to that from the real measurement.

关 键 词：在线检测 模糊神经网络 模糊聚类 表面粗糙度 机械零件 

分 类 号：TH161.14[机械工程—机械制造及自动化] TG8[金属学及工艺—公差测量技术]