基于SOM和HMM结合的刀具磨损状态监测研究

针对端面铣刀磨损状态的识别问题,提出了基于自组织特征映射神经网络和隐马尔可夫模型结合的方法。该方法对铣削力信号进行预处理和相关特征提取,用自组织特征映射对信号特征矢量进行量化编码,所得码本作为隐马尔可夫模型的输入向量,分别训练三个不同磨损阶段的隐马尔可夫模型来对未知的刀具磨损状态进行监测与识别。实验结果表明,该方法能够对刀具磨损状态进行准确的识别,对自动化生产具有现实意义。

Research on Tool Wear Condition Monitoring Based on Combination of SOM and HMM

Because of the identification problems of cutting tool wear condtition,a method based on the combination of SOM and HMM was proposed.At first,the singals of milling force were pretreated and the related features were extracted,then the feature vectors were presorted and coded into code book of integer numbers by using SOM as import vectors of HMM to build up 3-HMMs for different tool wear stages.And then the unknown state of tool wear was monitored and indentified.Finally,the experimental results indicate t...