东北大学学报:自然科学版 ›› 2017, Vol. 38 ›› Issue (5): 675-679.DOI: 10.12068/j.issn.1005-3026.2017.05.014

• 机械工程 • 上一篇    下一篇

铣削加工颤振稳定性可靠度的计算方法

黄贤振, 胡森, 张义民   

  1. (东北大学 机械工程与自动化学院, 辽宁 沈阳110819)
  • 收稿日期:2016-03-31 修回日期:2016-03-31 出版日期:2017-05-15 发布日期:2017-05-11
  • 通讯作者: 黄贤振
  • 作者简介:黄贤振(1982-),男,山东定陶人,东北大学副教授,博士; 张义民(1958-),男,吉林长春人,东北大学教授,教育部“长江学者奖励计划”特聘教授.
  • 基金资助:

    国家自然科学基金资助项目(51105062,51305071); 教育部新世纪优秀人才支持计划项目(NCET-13-0103); 中央高校基本科研业务费专项资金资助项目(N140304003).

On the Calculation Method for the Reliability of Chatter Stability in CNC Milling

HUANG Xian-zhen, HU Sen, ZHANG Yi-min   

  1. School of Mechanical Engineering & Automation, Northeastern University, Shenyang 110819, China.
  • Received:2016-03-31 Revised:2016-03-31 Online:2017-05-15 Published:2017-05-11
  • Contact: HUANG Xian-zhen
  • About author:-
  • Supported by:

    -

摘要:

铣削过程中的系统参数往往具有随机性,严重影响了铣削加工的稳定性. 利用BP神经网络综合分析了随机因素对铣削加工过程的影响,提出了一种铣削加工再生型颤振稳定性可靠度计算方法.建立铣削加工再生型颤振动力学模型,使用全离散法获取铣削稳定性叶瓣图.利用神经网络拟合极限轴向切深的函数表达式,再分别使用蒙特卡罗法和一次二阶矩法进行可靠度计算.结果表明,基于BP神经网络的方法兼具高效和精确的优点.

关键词: 再生型颤振, 铣削, 稳定性叶瓣图, 神经网络, 可靠性

Abstract:

The system parameters in the process of milling generally have randomness, which influences the stability of milling. The neural network was used to make a comprehensive analysis of the effects of random factors in milling, and a method for the reliability analysis of regenerative chatter stability in milling was proposed. A dynamic model for regenerative chatters of milling was established, and the stability lobe diagram was obtained by means of full discretization.The neural network was applied to approximate the functional relationship of the limit axial cutting depth,and the reliability was then calculated with the Monte Carlo method and first order second moment method, respectively. Finally, the results of an example were used to demonstrate the efficiency and accuracy of the proposed method based on the neural network.

Key words: regenerative chatter, milling, stability lobe diagram, neural network, reliability

中图分类号: