刀具是机械加工中的重要组成部分，刀具磨损会影响加工精度和加工效率，准确掌握加工时刀具磨损状态至关重要，提出了一种改进粒子群（improved particle swarm optimization ，IPSO）算法优化最小二乘支持向量机（least square support vector machine ，LS-SVM）的刀具磨损状态监测方法。采集铣削时的切削力信号，分别利用经验模态分解(empirical mode decomposition，EMD)和主成分分析(principal component analysis，PCA)法进行特征提取和降维，改进粒子速度、位置的更新策略和权重搜索方法提出IPSO算法，IPSO算法通过计算适应度函数对LS-SVM识别模型的惩罚因子和核参数迭代寻优。结果表明，降维后的特征可分性较强，IPSO算法寻优能力强于传统PSO和LdPSO算法，将降维后的特征当作IPSO-LS-SVM模型的输入，模型的识别精度和效率优于PSO和LdPSO优化的LS-SVM模型。
关键词：刀具磨损状态；经验模态分解；特征降维；改进粒子群；最小二乘支持向量机
Abstract
Tool wear is an important part of machining. Tool wear can affect machining accuracy and efficiency, so it is very important to accurately grasp the tool wear status during machining. An improved particle swarm optimization (IPSO) algorithm optimized the least square support vector machine (LS-SVM) tool wear status monitoring method was proposed.The cutting force signals during milling were collected, and the feature extraction and dimension reduction were carried out by empirical mode decomposition (EMD) and principal component analysis（PCA） respectively. The IPSO algorithm was proposed by improving the updating strategy of particle velocity and position and weight search method. The IPSO algorithm iteratively optimized the penalty factor and kernel parameters of the LS-SVM recognition model by calculating the fitness function.The results show that the features with dimensionality reduction are highly separable, and the optimization ability of IPSO algorithm is stronger than that of traditional PSO and LdPSO algorithms. When the features with dimensionality reduction are taken as the input of IPSO-LS-SVM model, the recognition accuracy and efficiency of the model are better than that of LS-SVM model optimized by PSO and LdPSO.
Key words: tool wear status; empirical mode decomposition; feature dimensionality reduction; improved particle swarm;least squares support vector machine
关键词
刀具磨损状态 /
经验模态分解 /
特征降维 /
改进粒子群 /
最小二乘支持向量机
{{custom_keyword}} /
Key words
tool wear status /
empirical mode decomposition /
feature dimensionality reduction /
improved particle swarm;least squares support vector machine
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 安华，王国锋，王喆，等. 基于深度学习理论的刀具状态监测及剩余寿命预测方法[J]. 电子测量与仪器学报, 2019, 33(09): 64-70.
AN Hua, WANG Guo-feng, WANG Zhe, et al. Deep learning-based method for tool condition monitoring and remaining useful life prediction[J]. Journal of Electronic Measurement and Instrumentation, 2019, 33(09): 64-70.
[2] 吕东升，徐九华，丁文锋，等. 端铣Ti40阻燃钛合金的刀具磨损破损形态及其机理[J]. 摩擦学学报, 2014, 34(03): 268-277.
LV Dong-sheng, XU Jiu-hua, DING Wen-feng, et al. Wear behavior and mechanism of coated carbide tools in face milling Ti40 burn resistant titanium alloy[J]. Tribology, 2014, 34(03): 268-277.
[3] 程训，余建波. 基于机器视觉的加工刀具磨损监测方法[J]. 浙江大学学报(工学版), 2021, 55(05): 896-904.
CHENG Xun, YU Jian-bo. Monitoring method for machining tool wear based on machine vision[J]. Journal of Zhejiang University (Engineering Science), 2021, 55(05): 896-904.
[4] Shi Xue-chun, Wang Xi-bin, Jiao Li, et al. A real-time tool failure monitoring system based on cutting force analysis[J]. International Journal of Advanced Manufacturing Technology, 2017(1): 1-17.
[5] 李亚，黄亦翔，赵路杰，等. 基于t分布邻域嵌入与XGBoost的刀具多工况磨损评估[J]. 机械工程学报, 2020, 56(01): 132-140.
LI Ya, HUANG Yi-xiang, ZHAO Lu-jie, et al. Multi-condition wear evaluation of tool based on T-SNE and XGBoost[J]. Journal of Mechanical Engineering, 2020, 56(01): 132-140.
[6] 张锴锋，袁惠群，聂鹏. 基于广义分形维数的刀具磨损状态监测[J]. 振动与冲击, 2014, 33(01): 162-164+169.
ZHANG Kai-feng, YUAN Hui-qun, NIE Peng. Tool wear condition monitoring based on generalized fractal dimensions[J]. Journal of Vibration and Shock, 2014, 33(01): 162-164+169.
[7] 孙巍伟，黄民，高延. 基于EMD-HMM的机床刀具磨损故障诊断[J]. 机床与液压, 2017, 45(13): 178-181.
SUN Wei-wei, HUANG Min, GAO Yan. CNC tools wearing fault diagnosis based on EMD-HMM[J]. Machine Tool & Hydraulics, 2017, 45(13): 178-181.
[8] 唐利平，刘海雄. 刀具磨损状态特征参数提取与识别方法研究[J]. 组合机床与自动化加工技术, 2019, 4(10): 10-14+19.
TANG Li-ping, LIU Hai-xiong. Study on cutting tool wear state characteristic parameters extraction and recognition method[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2019, 4(10): 10-14+19.
[9] 关山，康振兴，彭昶. 基于云理论与LS-SVM的刀具磨损识别方法[J]. 振动.测试与诊断, 2017, 37(05): 996-1003+1068-1069.
GUAN Shan, KANG Zhen-xing, PENG Chang. Method of tool wear recognition based on cloud theory and LS-SVM[J]. Journal of Vibration,Measurement & Diagnosis, 2017, 37(05): 996-1003+ 1068-1069.
[10] Xie Yang, Zhang Chao-yong, Liu Qiong. Tool Wear Status Recognition and Prediction Model of Milling Cutter Based on Deep Learning[J]. IEEE Access, 2020(9): 1616-1625.
[11] 李国龙，李彪，蒋林，等. 基于HHT和IPSO算法优化RBF神经网络的滚刀磨损状态识别方法[J]. 吉林大学学报(工学版), 2020, 50(06): 1998-2009.
LI Guo-long, LI Biao, JIANG lin, et al. Hob wear recognition method based on HHT and IPSO-RBF neural network[J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(06): 1998-2009.
[12] 刘成颖，吴昊，王立平，等. 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(09): 975-979.
LIU Cheng-ying, WU Hao, WANG Li-ping, et al. Tool wear state recognition basedon LS-SVM with the PSO algorithm[J]. Journal of Tsinghua University(Science and Technology) 2017, 57(09): 975-979.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1590 KB)
