Tool wear state recognition based on LS-SVM with the PSO algorithm
LIU Chengying1,3, WU Hao2, WANG Liping1,3, ZHANG Zhi4
1. Department of Mechanical Engineering, Tsinghua University, Beijing 100084, China;
2. School of Mechatronics Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China;
3. Beijing Key Lab of Precision/Ultra-precision Manufacturing Equipments and Control, Tsinghua University, Beijing 100084, China;
4. Department of Aircraft Engineering, Navy Aeronautical Engineering Academy, Yantai 264001, China
Abstract：A tool wear state monitoring system was developed based on acoustic emissions to monitor the tool wear state. Typical acoustic signals were analyzed to determine the square root amplitude, absolute mean, mean square error and maximum sound level from the time domain to characterize the tool wear. Neural networks can easily fall into a local minimum and have slow learning convergence rates so a tool wear state recognition method was developed based on a least square support vector machine (LS-SVM). The LS-SVM performance depends on the penalty factor and the kernel parameter, so a particle swarm optimization algorithm was used to automatically optimize the LS-SVM parameters. The optimized LS-SVM model is then shown to be more accurate than the basic LS-SVM model.
刘成颖, 吴昊, 王立平, 张智. 基于PSO优化LS-SVM的刀具磨损状态识别[J]. 清华大学学报（自然科学版）, 2017, 57(9): 975-979.
LIU Chengying, WU Hao, WANG Liping, ZHANG Zhi. Tool wear state recognition based on LS-SVM with the PSO algorithm. Journal of Tsinghua University(Science and Technology), 2017, 57(9): 975-979.
Abellan-Nebot J V, Subirón F R. A review of machining monitoring systems based on artificial intelligence process models[J]. The International Journal of Advanced Manufacturing Technology, 2010, 47(1):237-257.
[2]
高宏力. 切削加工过程中刀具磨损的智能监测技术研究[D]. 成都:西南交通大学, 2005.GAO Hongli. The Investigation of Intelligent Tool Wear Monitoring Techniques for Metal Cutting Process[D]. Chengdu:Southwest Jiaotong University. 2005. (in Chinese)
[3]
关山. 基于声发射信号多特征分析与融合的刀具磨损分类与预测技术[D]. 长春:吉林大学, 2011.GUAN Shan. Technology of Classification and Forecast of Tools Wear Based on Multi-feature Analysis and Fusion of AE[D].Changchun:Jilin University, 2011.(in Chinese)
[4]
Özel T, Karpat Y. Predictive modeling of surface roughness and tool wear in hard turning using regression and neural networks[J]. International Journal of Machine Tools & Manufacture, 2005, 45(4-5):467-479.
[5]
陈爱弟, 王信义, 王忠民. 基于模糊聚类的刀具磨损量在线监测方法[J]. 北京理工大学学报, 2000, 20(3):276-280.CHEN Aidi, WANG Xinyi, WANG Zhongmin. A method of online monitoring tool wear based on fuzzy classification[J]. Journal of Beijing Institute of Technology, 2000, 20(3):276-280. (in Chinese)
[6]
Suykens J A K, Vandewalle J. Least squares support vector machine classifiers[J]. Neural processing letters, 1999, 9(3):293-300.
[7]
杨先勇, 周晓军, 张文斌,等. 基于局域波法和KPCA-LSSVM的滚动轴承故障诊断[J]. 浙江大学学报(工学版), 2010, 44(8):1519-1524.YANG Xianyong, ZHOU Xiaojun, ZHANG Wenbing, et al. Rolling bearing fault diagnosis based on local wave method and KPAC-LSSVM[J]. Journal of Zhejiang University (Engineering Science), 2010, 44(8):1519-1524(in Chinese)
[8]
Heidari M, Homaei H, Golestanian H, et al. Fault diagnosis of gearboxes using wavelet support vector machine, least square support vector machine and wavelet packet transform[J]. Journal of Vibroengineering, 2016, 18(2):860-875.
[9]
Liu Z, Cao H, Chen X, et al. Multi-fault classification based on wavelet SVM with PSO algorithm to analyze vibration signals from rolling element bearings[J]. Neurocomputing, 2013, 99(1):399-410.
[10]
李威霖, 傅攀, 张尔卿. 基于粒子群优化LS-SVM的车刀磨损量识别技术研究[J]. 计算机应用研究, 2014, 31(4):1094-1097.LI Weilin, FU Pan, ZHANG Erqing. Application of particle swarm optimization-least square support vector machine in tool wear monitoring[J]. Application Research of Computers, 2014, 31(4):1094-1097. (in Chinese)
[11]
Shrivastava N A, Khosravi A, Panigrahi B K. Prediction interval estimation of electricity prices using PSO-tuned support vector machines[J]. IEEE Transactions on Industrial Informatics, 2015, 11(2):322-331.
[12]
Selakov A, Cvijetinovi Ac' D, Milovi Ac' L, et al. Hybrid PSO-SVM method for short-term load forecasting during periods with significant temperature variations in city of Burbank[J]. Applied Soft Computing, 2014, 16(3):80-88.
2017, Vol. 57 
