刀具磨损
钻探
机械加工
过程(计算)
支持向量机
钻头
计算机科学
振动
特征提取
人工智能
工程类
机械工程
声学
操作系统
物理
作者
Yaochen Shi,Zeqi Li,Zhi-Yuan Wei,Xuelian Yu,Chunmei Yin,Yi-Shi Bai
标识
DOI:10.1166/jno.2022.3173
摘要
Aiming at the problems of not being able to directly monitor the wear state of drill bit during vibration drilling and not being able to collect relevant dynamic data online during machining, a digital twin-driven online monitoring method for vibration drilling bit wear was proposed. Firstly, feature extraction of multi-source signals in drilling process is carried out by wavelet analysis, and a double neural network model for bit wear recognition is established. Based on this, an online monitoring method for bit wear is proposed. The digital twinning system for bit wear is implemented, and the dynamic data in drilling process is collected online, and the simulation of bit wear process is realized synchronously. Finally, the proposed prediction method is compared with Support vector machine (SVM) recognition method. The results show that the proposed method can effectively predict the tool wear condition and realize the real-time identification of tool wear degree in machining process.
科研通智能强力驱动
Strongly Powered by AbleSci AI