摘要
文章对碳纤维增强复合材料(carbon fiber reinforced plastic, CFRP)进行了钻削加工有限元仿真研究,通过正交试验的方法分析了切削用量和刀具几何参数对分层因子的影响作用;建立了关于分层因子的人工神经网络预测模型与多元线性回归预测模型,2种模型预测值的最大相对误差分别为3.1%、7.4%,得出人工神经网络预测模型具有更高的预测精度。研究结果表明:分层因子总体上随主轴转速的增加而降低,随进给量和顶角的增加而增加,对分层因子的影响程度由大到小依次为进给量、转速、顶角;在对CFRP进行钻削加工时,宜适当提高主轴转速、降低进给量和钻头顶角,以减少钻削引起的分层缺陷,进一步改善钻孔质量。该文所采用的方法为CFRP钻削加工中切削力、刀具磨损等问题的分析提供了一定的理论参考。
The finite element simulation study of drilling carbon fiber reinforced plastic(CFRP) composites was conducted. The influence of cutting parameters and cutting tool geometric parameters on the delamination factor was analyzed by using orthogonal experiment method and the results show that the delamination factor generally decreases with the increase of the spindle speed, and increases with the increase of the feed rate and point angle. The ranking of influencing significance on the delamination factor is feed rate, spindle speed, and point angle. Both artificial neural network(ANN) and multiple linear regression models for predicting the delamination factor were established. The maximum relative prediction errors of the two models were 3.1% and 7.4%, respectively. It is concluded that the artificial neural network model has higher prediction accuracy. It is recommended to rise the spindle speed and decrease the feed rate and point angle in drilling CFRP composites, in order to reduce the delamination factor and further improve the drilling quality. Moreover, the method used in this paper also provides a theoretical reference for the study of cutting force and tool wear in the drilling of carbon fiber composite materials.
作者
张浩
安立宝
张迎信
ZHANG Hao;AN Libao;ZHANG Yingxin(College of Mechanical Engineering,North China University of Science and Technology,Tangshan 063210,China)
出处
《合肥工业大学学报（自然科学版）》
CAS
北大核心
2019年第12期1608-1614,共7页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金资助项目(51472074)
河北省引进海外高层次人才“百人计划”资助项目(E2012100005)
关键词
碳纤维增强复合材料(CFRP)
钻削
有限元仿真
分层因子
人工神经网络
carbon fiber reinforced plastic(CFRP)
drilling process
finite element simulation
delamination factor
artificial neural network(ANN)
