• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2020, Vol. 56 ›› Issue (2): 27-34,42.doi: 10.3901/JME.2020.02.027

• 仪器科学与技术 • 上一篇    下一篇

复合材料钻削表面粗糙度在线监测与加工参数自适应优化

安华1, 王喆1, 王国锋1, 宋庆月1, 刘峰2, 钟才川1   

  1. 1. 天津大学机械工程学院 天津 300350;
    2. 天津金岸重工有限公司 天津 300350
  • 收稿日期:2019-01-20 修回日期:2019-09-01 出版日期:2020-01-20 发布日期:2020-03-11
  • 通讯作者: 王国锋(通信作者),男,1975年出生,博士,教授,博士研究生导师。主要研究方向为刀具状态监控、非平稳信号处理、加工过程辩识、智能控制。E-mail:gfwangmail@tju.edu.cn
  • 作者简介:安华,男,1995年出生。主要研究方向为智能监测及控制。E-mail:245367067@qq.com
  • 基金资助:
    国家自然科学基金(51675369)、天津市自然科学基金(17JCZDJC40100)、航空科学基金(2017ZE25003)、国防基础科研计划(JCKY2018205C002)和2018年度天津市交通运输科技发展项目计划(2018-b10)资助项目。

Research on On-line Monitoring of Surface Roughness in Composite Drilling and Adaptive Optimization of Parameters

AN Hua1, WANG Zhe1, WANG Guofeng1, SONG Qingyue1, LIU Feng2, ZHONG Caichuan1   

  1. 1. School of Mechanical Engineering, Tianjin University, Tianjin 300350;
    2. Tianjin JinAn Heavy Equipment Co,. Ltd., Tianjin 300350
  • Received:2019-01-20 Revised:2019-09-01 Online:2020-01-20 Published:2020-03-11

摘要: 针对碳纤维复合材料钻孔加工过程中无法实现加工质量与效率统一以及离线优化得到的钻削参数没有考虑变参数刀具磨损等不确定因素影响问题,提出一种新的粗糙度在线监测与加工参数自适应优化方法。采用一组新的无量纲特征以实现变参数刀具磨损监测;以刀具磨损状态值、进给速度以及主轴转速构成特征向量,进而建立基于支持向量回归的孔壁粗糙度在线监测模型。当监测系统判定孔壁粗糙度不合格时,利用模拟退火算法在当前刀具磨损状态下进行钻削参数优化。利用复合材料变参数钻削试验来验证上述方法的有效性。试验结果表明,该方法能够准确判定粗糙度的质量状态,同时有效实现钻削参数自适应优化,解决了复合材料制孔过程中加工质量与效率统一的问题。

关键词: 碳纤维复合材料, 变参数, 刀具磨损, 粗糙度, 支持向量机回归, 模拟退火

Abstract: A new method for on-line monitoring of surface roughness and adaptive optimization of processing parameters is proposed to overcome the drawback that the current drilling process of carbon fiber reinforced plastics (CFRP) can't obtain the uniformity of machining quality and efficiency, as well as the drilling parameters obtained by offline optimization don't consider the influence of uncertain factors such as tool wear. A new set of dimensionless features is used to achieve variable parameter tool wear monitoring, and then the SVR-based online monitoring model of roughness is established by the feature vector which is formed by tool wear state value, feed rate and spindle rate. When the hole wall roughness is unqualified, the simulated annealing algorithm is used to optimize the drilling parameters under the present tool wear condition. The drilling experiment with the variable cutting parameters is used to verity the effectiveness of the proposed method in the end. The results show that the proposed method is able to not only realize the online monitoring of roughness effectively and optimize the cutting parameters adaptively, but also solve the problem of compromising the quality and efficiency in drilling process of CFRP.

Key words: carbon fiber reinforced plastics, variable parameter, tool wear, roughness, support vector regression, simulated annealing

中图分类号: