摘要
文章在现有刀具磨损检测算法的研究基础上,针对铣削刀具磨损检测提出了一种基于卷积神经网络的刀具磨损检测算法。对原始的AlexNet卷积神经网络参数以及训练算法进行了优化,同时搭建了实验平台,进行了机床切削实验并采集了大量的工件纹理图片,使用这些图片样本对所改进的方法进行验证并与其他算法进行了对比研究。结果表明,改进后的AlexNet卷积神经网络算法能够更好地对刀具磨损程度进行判断。
On the basis of the existing tool wear detection algorithm, a detection algorithm for milling tool wear based on convolutional neural network(CNN) is proposed, in this paper. The original AlexNet CNN parameters and training algorithms were optimized, and an experimental platform was built. The machine cutting experiment was carried out and a large number of workpiece texture photos were captured. The improved method was verified using these samples, and was researched compared with other algorithms. The results show that the improved AlexNet CNN algorithm can better judge the degree of tool wear.
引文
[1]杨建国,肖蓉,李蓓智,等.基于机器视觉的刀具磨损检测技术[J].东华大学学报(自然科学版),2012,38(5):505-508,518.
[2]秦国华,易鑫,李怡冉,等.刀具磨损的自动检测及检测系统[J].光学精密工程,2014,22(12):3332-3341.
[3]García-Ordás M T,Alegre E,Alaiz R,et al.Tool wear monitoring using an online,automatic and low cost system based on local texture[J].Mechanical Systems&Signal Processing,2018(112):98-112.
[4]何翔,任小洪.基于改进BP神经网络的刀具磨损状态检测[J].工业控制计算机,2015(10):49-51,53.
[5]李姗姗,刘丽冰,李莉,等.基于区域生长法的数控刀具磨损状态检测方法[J].制造技术与机床,2017(2):132-136.
[6]C Bradley,Y S Wong.Surface texture indicators of tool wear-A machine vision approach[J].International Journal of Advanced Manufacturing Technology,2001,17(6):435-443.
[7]郑建明,李鹏阳,李言,等.基于Hough变换的刀具磨损监测加工表面纹理特征提取[J].机械科学与技术,2009,28(6):711-715.
[8]孙林丽,李言,郑建明,等.基于表面二维PCA重构图像的刀具磨损分形特征研究[J].机械科学与技术,2010,29(3):395-398,403.
[9]张存吉,姚锡凡,张剑铭,等.基于深度学习的刀具磨损监测方法[J].计算机集成制造系统,2017,23(10):2146-2155.
[10]Li Lihong,An Qingbin.An in-depth study of tool wear monitoring technique based on image segmentation and texture analysis[J].Measurement,2016(79):44-52.
[11]Wang Guofeng,Zhang Yanchao,Liu Chang,et al.A new tool wear monitoring method based on multi-scale PCA[J].Journal of Intelligent Manufacturing,2016(7):1-10.
[12]王晓强,张云,周华民,等.基于隐马尔科夫模型的刀具磨损连续监测[J].组合机床与自动化加工技术,2016(10):87-90.
[13]Anti'c A,Popovi'c B,Krstanovi'c L,et al.Novel texturebased descriptors for tool wear condition monitoring[J].Mechanical Systems&Signal Processing,2018(98):1-5.
[14]Yang Zeqing,Liu Libing,Peng Kai,et al.Monitoring Method of High-speed Tool Wear Level based on Machine Vision[J].International Journal of Signal Processing,Image Processing and Pattern Recognition,2017(10):23-38.