基于谐波小波包和BSA优化LS-SVM的铣刀磨损状态识别研究

发布时间：2018-07-20 16:17

【摘要】：针对铣削刀具磨损状态识别问题,提出谐波小波包和最小二乘支持向量机(LS-SVM)的状态识别方法。为克服传统小波包分解的频带交叠问题,采用谐波小波包提取不同磨损状态下铣削力信号的各频段信号能量,归一化处理后,输入LS-SVM多类分类器,实现铣削刀具磨损状态的识别。针对LS-SVM的惩罚因子和核参数对模型识别精度影响较大的问题,提出回溯搜索算法(BSA)进行自动参数寻优。实验结果表明,谐波小波包比小波包在刀具磨损状态特征提取时具有更好的识别效果。与粒子群算法进行比较,证明BSA优化LS-SVM具有更高的识别精度。
[Abstract]:A state recognition method based on harmonic wavelet packet and least square support vector machine (LS-SVM) is proposed to identify the wear state of milling tool. In order to overcome the overlapping problem of traditional wavelet packet decomposition, harmonic wavelet packet is used to extract the signal energy of milling force signal in different wear states. After normalized processing, LS-SVM multi-class classifier is input. The recognition of milling tool wear state is realized. Aiming at the problem that the penalty factor and kernel parameters of LS-SVM have great influence on the model recognition accuracy, a backtracking search algorithm (BSA) is proposed for automatic parameter optimization. The experimental results show that harmonic wavelet packet has better recognition effect than wavelet packet in feature extraction of tool wear state. Compared with particle swarm optimization (PSO), it is proved that BSA optimized LS-SVM has higher recognition accuracy.
【作者单位】：华中科技大学机械学院数字制造装备与技术国家重点实验室;湖北汽车工业学院电气与信息工程学院;宁波大学机械工程与力学学院;
【基金】：国家自然科学基金资助项目(51575211,51421062);国家自然科学基金国际(地区)合作与交流项目(51561125002) 湖北省自然科学基金资助项目(2014CFB348) 高等学校学科创新引智计划资助项目(B16019)
【分类号】：TG714

【相似文献】