Modeling Method for Tool Wear Prediction Based on ADNLSSVM

China Mechanical Engineering

Previous Articles Next Articles

Modeling Method for Tool Wear Prediction Based on ADNLSSVM

XIAO Pengfei1;ZHANG Chaoyong1;LUO Min1;LIN Wenwen2

1.School of Mechanical Science & Engineering, Huazhong University of Science and Technology, Wuhan, 430074
2.Mechanical Engineering & Mechanics, Ningbo University,Ningbo,Zhejiang, 315211

Online:2018-04-10 Published:2018-04-03
Supported by:
National Natural Science Foundation of China（No. 51575211）
Hubei Provincial Natural Science Foundation of China（No. 2014CFB348）

基于自适应动态无偏最小二乘支持向量机的刀具磨损预测建模

肖鹏飞1;张超勇1;罗敏1;林文文2

1.华中科技大学机械科学与工程学院，武汉，430074
2.宁波大学机械工程与力学学院，宁波，315211

基金资助:
国家自然科学基金资助项目（51575211）；
国家自然科学基金国际(地区)合作交流资助项目（51561125002）；
高等学校学科创新引智计划资助项目 (B16019)；
湖北省自然科学基金资助项目（2014CFB348）
National Natural Science Foundation of China（No. 51575211）
Hubei Provincial Natural Science Foundation of China（No. 2014CFB348）

Abstract

Abstract: In the building process of a tool condition prediction model with traditional machine learning methods, the limited number of available training samples and the fixed length of the sliding time window and prediction model resulted in lower modeling accuracy and efficiency. Dynamic model was set up to monitor the tool wear states by using an ADNLSSVM. Feature vectors were extracted by time-frequency-domain analysis from data set of open database of milling processes, and parts of them were selected by correlation analysis as model inputs. The experimental results shows better modeling efficiency and prediction accuracy.

Key words: adaptive dynamic non-bias least squares support vector machine (ADNLSSVM), adaptive sliding time window, feature vector extraction and selection, tool wear

摘要： 由于训练样本数量有限，滑动时间窗长度以及监测模型不能自适应调整和更新等因素，传统基于机器学习的刀具磨损预测模型存在精度和效率较低等问题，因此提出了一种基于自适应动态无偏最小二乘支持向量机（ADNLSSVM）的刀具磨损预测模型。采用公开数据库中的铣削加工数据集，通过时频域分析和小波包分解等手段从振动信号中提取特征量，并进一步利用相关性分析从中选择有效特征量作为模型输入。试验结果表明该方法所建模型具有较高的建模效率和预测精度。

关键词: 自适应动态无偏最小二乘支持向量机, 滑动时间窗自适应调整, 特征提取和选择, 刀具磨损

CLC Number:

TH186

XIAO Pengfei1;ZHANG Chaoyong1;LUO Min1;LIN Wenwen2. Modeling Method for Tool Wear Prediction Based on ADNLSSVM[J]. China Mechanical Engineering.

肖鹏飞1;张超勇1;罗敏1;林文文2. 基于自适应动态无偏最小二乘支持向量机的刀具磨损预测建模[J]. 中国机械工程.

References

[1]庄子杰.基于声发射和振动法的刀具磨损状态检测研究[D]. 上海：上海交通大学，2009.
ZHUANG Zijie. Tool Wear Condition Monition Based on AE and Vibration[D]. Shanghai：Shanghai Jiao Tong University,2009.
[2] TONSBOFF H K. Developments and Trends in Monitoring and Control of Machining Processes[J].Annual of the CIRP,1988,37(2):611-622.
[3]杨吟飞,李亮,何宁.一种新的刀具磨损面图像边界提取方法[J].南京航空航天大学学报,2007,39(6):786-789.
YANG Yinfei, LI Liang, HE Ning. Extraction Method for Image Boundary Based on Tool Wear［J］. Journal of Nanjing University of Aeronautics and Astronautics,2007,39(6):786-789.
[4]张臣,周来水,安鲁陵,等.球头铣刀刀具磨损建模与误差补偿[J]. 机械工程学报,2008,44(2):207-212.
ZHANG Chen, ZHOU Laishui, AN Luling, et al.Modeling and Wear-induced Error Compensation of Ball-end Milling Cutter Wear[J]. Chinese Journal of Mechanical Engineering,2008,44(2):207-212.
[5]VIJAYAKUMAR S, WU S. Sequential Support Vector Classifiers and Regression[C]// International Conference on Soft Computing. Genoa,1999:610-619.
[6]蔡艳宁, 汪洪桥, 叶雪梅.复杂系统支持向量机建模与故障预报[M].北京：国防工业出版社,2015:4.
CAI Yanning, WANG Hongqiao, YE Xuemei. Support Vecto Machine Modeling and Fault Prediction for Complex System［M］. Beijing: National Defence Industry Press,2015:4.
[7]张浩然,汪晓东.回归最小二乘支持向量机的增量和在线式学习算法[J].计算机学报,2006,29(3):394-406.
ZHANG Haoran, WANG Xiaodong. Incremental and Online Learning Algorithm for Regression Least Squares Support Vector Machine[J]. Chinese Journal of Computers,2006,29(3):394-406.
[8]李威霖，傅攀，张尔卿.基于粒子群优化LS-SVM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 Machine in Tool Wear Monitoring[J]. Application Research of Computers,2014,31(4):1094-1097.
[9]谢厚正,黄民.基于振动测试的数控机床刀具磨损监测方法[J].仪表技术与传感器,2013(2):73-76.
XIE Houzheng, HUANG Min. Research of Numerical Control Machine Tools Wear Monitoring Method Based on Vibration Test[J]. Instrument Technique and Sensor, 2013(2):73-76.
[10]丛龙慧,韩玉杰.小波分析在刀具磨损状态检测中的应用[J].林业机械与土木工程,2009,37(4):51-52.
CONG Longhui, HAN Yulin. Application of Wavelet Analysis in Tool Wear Status Detection. Forestry Machinery & Woodworking Equipment, 2009, 37(4):51-52.

[1]	DAI Wen1;ZHANG Chaoyong1;MENG Leilei1;XUE Yanshe1;XIAO Pengfei1;YIN Yong2. Prediction Model of Milling Cutter Wear Status Based on Deep Learning [J]. China Mechanical Engineering, 2020, 31(17): 2071-2078.
[2]	HE Yan1;LING Junjie1;WANG Yulin2;LI Yufeng1;WU Pengcheng1;XIAO Zhen1. In-process Tool Wear Monitoring Model Based on LSTM-CNN [J]. China Mechanical Engineering, 2020, 31(16): 1959-1967.
[3]	LI Yousheng. Influences of TaC(NbC) on Cemented Carbide Tool Wear Resistance in High Speed Milling [J]. China Mechanical Engineering, 2020, 31(13): 1601-1605.
[4]	LIU Peilei;YANG Shiqiang;BAI Lele;LI Zhuo;LI Dexin. Wear Model of Disc Hob Cutter Rings Based on Abrasive Wear [J]. China Mechanical Engineering, 2019, 30(15): 1782-1789.
[5]	LU Zhiyuan, MA Pengfei, XIAO Jianglin, WANG Meiqing, TANG Xiaoqing. On-line Monitoring of Tool Wear Conditions in Machining Processes Based on Machine Tool Data [J]. China Mechanical Engineering, 2019, 30(02): 220-225.
[6]	WANG Lihua1;YANG Jiawei1;ZHANG Yonghong1;ZHAO Xiaoping2;XIE Yangyang1. Tool Wear Condition Recognition Based on SDAE [J]. China Mechanical Engineering, 2018, 29(17): 2038-2045.
[7]	DONG Caiyun1;ZHANG Chaoyong1;MENG Leilei1;XIAO Pengfei1;LUO Min2;LIN Wenwen3. State Recognition of Milling Tool Wears Based on Harmonic Wavelet Packet and BSA Optimization LS-SVM [J]. China Mechanical Engineering, 2017, 28(17): 2080-2089,2108.
[8]	Gao Yanfeng, Xiao Jianhua. Characteristics of Cutting Tool Wear during LAM Processes for Ti-6Al-4V Alloy [J]. China Mechanical Engineering, 2016, 27(21): 2877-2883.
[9]	Huan Haixiang, , Xu Jiuhua, Su Honghua, Fu Yucan, Liang Xinghui, Ge Yingfei. Study on Tool Wear Characteristics during Turning Titanium Matrix Composite Using PCD and Carbide Tools [J]. China Mechanical Engineering, 2016, 27(14): 1877-1883.
[10]	Wu Haibing, Zuo Dunwen, Sun Quanping, Xu Feng. Tool Wear in ZrO2 Bioceramic Milling Processes [J]. China Mechanical Engineering, 2015, 26(7): 960-964,993.
[11]	Wang Ben, Liu Dongxi, Wang Minghai, Yin Wendian, Zhen Yaohui. Wear Mechanism of PCD Tool during Cutting of Isotropic Pyrolytic Graphite [J]. China Mechanical Engineering, 2015, 26(20): 2721-2729.
[12]	Guan Shan, Yan Lihong, Peng Chang. Application of Regression Algorithm of LS-SVM in Tool Wear Prediction [J]. China Mechanical Engineering, 2015, 26(2): 217-222.
[13]	Bian Rong, He Ning, Li Liang, Qian Jun, Shi Zhenyu, Chen Mingjun. Analysis on Characteristics of Milling Force in Micro-milling of ZrO₂ Ceramics [J]. China Mechanical Engineering, 2014, 25(23): 3200-3206.
[14]	SU Yu, HE Ning, LI Liang. Experimental Study on High Speed Milling Titanium Alloys Using Cryogenic Minimum Quantity Lubrication [J]. China Mechanical Engineering, 2010, 21(22): 2665-2670.
[15]	LV Dun-Jie, WANG Jie, WANG Mei, TUN Huo. Research on Tool Wear Condition Monitoring Based on Combination of SOM and HMM [J]. China Mechanical Engineering, 2010, 21(13): 1531-1535.

Modeling Method for Tool Wear Prediction Based on ADNLSSVM

基于自适应动态无偏最小二乘支持向量机的刀具磨损预测建模

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics