摘要
深孔加工中直线度是衡量孔的加工质量的重要指标之一。综合考虑直线度受到转速n、进给量s、切削液流速v、切削深度t4个切削参数影响,以此4个参数为输入建立BP神经网络系统。对BP神经网络进一步优化,应用基于进化算法的BP神经网络来对BTA钻削直线度进行预测,并通过Matlab编程以及绘制出直线度预测值与真实值折线图,达到了很好的效果,拓展了BTA钻削研究思路。
The straightness of deep hole machining is one of the important indexes of the machining quality of hole.Obviously,in metal cutting,the straightness of the hole is affected by four cutting parameters： rotate speed、feed quantity、cutting fluid velocity、cutting depth,how to predict the straightness of the hole according to the specific cutting parameters is of great important.Using the BP neural networks to predict the straightness exist several disadvantages：the network training is easy to fall into local minimum value,the network learning convergence speed is slow,the structure of the network is difficult to determine,the generalization ability of the network cannot be guaranteed. A BP neural network based on evolutionary algorithm is applied to predict drilling straightness,the BP neural networks are further optimized,and then drawing a linear prediction and a real value line chart based on Matlab,and the result is pretty good,it also provides a new study and analysis method for the studying of BTA drilling.
作者
靳伟贺
苗鸿宾
王婷
夏昊
JIN Weihe;MIAO Hongbin;WANG Ting;XIA Hao(School of Mechanical Engineering, North University of China, Taiyuan 030051, China; Shanxi Province Deep Hole Machining Center, Taiyuan 030051 ,China)
出处
《机械设计与研究》
CSCD
北大核心
2018年第5期162-166,共5页
Machine Design And Research
基金
山西省回国留学人员基金资助项目(2015-077)
