Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (32): 233-236.

• 工程与应用 • Previous Articles     Next Articles

End mill wear monitoring by hybrid intelligent data fusion

ZHENG Jin-xing   

  1. College of Electrical and Mechanical Engineering,Harbin Engineering University,Harbin 150001,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-11 Published:2007-11-11
  • Contact: ZHENG Jin-xing

基于混合智能计算的铣刀状态监测

郑金兴   

  1. 哈尔滨工程大学 大学机电学院,哈尔滨 150001
  • 通讯作者: 郑金兴

Abstract: Hybrid intelligent data fusion for monitoring end milling tool wear is presented in this paper.Signals of cutting force and vibration is measured with multi-sensors and features extraction in frequency domain and time-frequency domain using wavelet package decomposition.Several hybrid intelligent data fusion methods,which are wavelet neural networks,Generic Algorithm Neural Networks(GA-NN),and wavelet generic algorithm neural networks for predicting tool wear value are debated.The results show experimently all of these presented methods effectively implement tool wear monitoring and prediction,and the characters of these methods are analyzed.

Key words: tool wear, multi-sensors, hybrid intelligent data fusion, wavelet package decomposition

摘要: 提出了一种基于混合智能融合技术进行铣刀磨损量监测和预测方法。利用多传感器对切削力和振动信号进行监测,通过频率变换提取切削力特征量,采用小波包分解技术提取振动信号特征量。通过信号特征值的组合,分别探讨了几种混合智能数据融合技术-小波神经网络,遗传神经网络,遗传小波神经网络对刀具磨损量的预测效果。实验分析表明,提出的几种基于多传感器的混合智能数据融合技术均能够有效地完成刀具磨损量监测和预测,同时对它们各自的特点进行了比较分析。

关键词: 刀具磨损, 多传感器, 混合智能数据融合, 小波包分解