摘要
制造业自动化水平的提高对数控机床的自我诊断能力提出了新的要求,而人工智能的发展为此开辟了新的解决方案。为了更高效、全面地对刀具磨损状态进行评估,通过采集立铣刀切削时的力和加速度信号,并对其时域、频域与小波能量特征的信号的特征值进行提取,建立了一种基于随机森林算法(Random forest)的刀具磨损状态评估模型。实验数据的对比验证中,随机森林模型对107组测试样本的刀具磨损状态评估准确率达到99.1%,且其建立模型的时间少于1 s。结果表明,随机森林算法具有高效与高准确度的特点,能为刀具磨损状态的在线监测系统的建立奠定基础。
The new requirement for the self-diagnosis ability of CNC machine tools has been put forward in order to improve the manufacturing automation level,and the development of artificial intelligence has opened aiming at this purpose.In order to evaluate the tool wear state more efficiently and comprehensively,the force and acceleration signals in the end milling are collected.The eigenvalues of the time domain,frequency domain and wavelet energy of the signals are extracted.And a model for evaluating the tool wear state via Random Forest algorithm is established.In the comparative verification of the experimental data,the accuracy of the tool wear state of the 107 sets of test samples by using the random forest model is of 99.1%,and the time for establishing the model is below 1 s.The result shows that the random forest algorithm has the characteristics of high efficiency and high accuracy,which lays a foundation for establishing the online monitoring system for tool wear state.
作者
李凡
谢峰
李楠
Li Fan;Xie Feng;Li Nan(School of Electrical Engineering and Automation, Anhui University, Hefei 230601, China)
出处
《机械科学与技术》
CSCD
北大核心
2020年第3期419-424,共6页
Mechanical Science and Technology for Aerospace Engineering
基金
安徽省科技攻关项目(1301022079)资助。
关键词
机器学习
刀具磨损
特征提取
随机森林算法
machine learning
wear of tool
feature extraction
random forest algorithm
