基于AE信号与优化SVM的刀具磨损状态识别  

Identification of Tool Wear Condition Based on AE Signal and Optimized SVM

作　　者：张锴锋[1,2] 袁惠群[2] 聂鹏[1] 王佳林[1] 

机构地区：[1]沈阳航空航天大学机电工程学院,辽宁沈阳110136 [2]东北大学机械工程与自动化学院,辽宁沈阳110819

出　　处：《测控技术》2015年第11期20-23,28,共5页Measurement & Control Technology

基　　金：国家自然科学基金重点项目(51335003)

摘　　要：根据小波包变换理论,对刀具磨损声发射(AE)信号进行滤波和能量特征值提取。采用分步式扫描的方法对传统的盒计数法进行改进,并利用改进的盒计数法计算滤波后信号的广义分形维数特征值。以上述提取的特征值为备选特征,采用支持向量机(SVM)作为分类器,利用量子遗传算法(QGA)首先对分类器的输入特征进行筛选,之后对分类器的模型参数进行优化。利用优化后的分类器对测试样本进行分类,测试结果表明,该方法可以对刀具磨损状态进行有效识别。Based on wavelet packet transformation theory,the acoustic emission（AE） signal from the process of tool wear is filtered and the energy features are extracted.The traditional box-counting method is improved with step-scan method,and the generalized fractal dimension features of the filtered signal are calculated.Taking the extracted features above as alternative signal features and the support vector machine（SVM） as the classifier,the input features are screened and the model parameters are optimized with quantum genetic algorithm（QGA）.The test samples are classified with the optimized classifier,and the test result shows that this method can effectively identify tool wear condition.

关 键 词：小波包 AE信号 广义分形维数 支持向量机 量子遗传算法 

分 类 号：TH164[机械工程—机械制造及自动化]