摘要
在高速铣削加工中,为了判断更换刀具的最佳时间,迫切地需要对刀具的剩余使用寿命进行准确地预测,但预测中常常会存在历史数据不足的问题.因此,本文提出了一种解决小样本空间的刀具剩余使用寿命预测方法.该方法基于支持向量回归(SVR)方法,通过随机分形搜索(SFS)算法优化模型中的关键参数.相比于传统方法,本文所采用的方法可获得更优的模型参数和更快的收敛速度.最后,将所采用的方法与隐马尔可夫模型(HMM)方法进行比较,平均精确度由0.6277提高至0.8199,为刀具的更换提供了可靠的参考.
In high speed milling operations,it is urgent to accurately predict the remaining useful life of cutting tools to determine the best time to replace them,but there is often a problem of insufficient historical data in the prediction.Therefore,a method for remaining useful life prediction of cutting tools in small sample space is proposed,which is based on the support vector regression(SVR)method.And the stochastic fractal search(SFS)algorithm is used to optimize the key parameters of SVR.Compared with the traditional method,it obtains better model parameters and faster convergence speed.Finally,the as-adopted method is compared with the hidden Markov model(HMM)approach.The accuracy rate increases from 0.6277 to 0.8199,which provides a reliable reference for tool replacement.
作者
黄贤振
孙良仕
高娓
李禹雄
HUANG Xian-zhen;SUN Liang-shi;GAO Wei;LI Yu-xiong(School of Mechanical Engineering&Automation,Northeastern University,Shenyang 110819,China;China National Heavy Duty Truck Group Jinan Truck Co.,Ltd.,Jinan 250116,China)
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2023年第6期824-831,共8页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(51975110).
关键词
高速铣削
刀具磨损
剩余使用寿命
随机分形搜索
支持向量回归
high-speed milling
tool wear
remaining useful life
stochastic fractal search
support vector regression
