利用声发射信号的特征分析对冲压模具的状态判别  

State identification of a punching die based on characteristic analysis of acoustic emission signal

作　　者：骆志高[1] 王祥[1] 李举[1] 范彬彬[1] 郭啸栋[1] 

机构地区：[1]江苏大学机械工程学院,镇江212013

出　　处：《振动与冲击》2009年第3期186-189,共4页Journal of Vibration and Shock

摘　　要：根据冲压模具失效产生声发射信号的特点,确定失效信号的特征参数;以小波包分析技术提取模具信号的能量特征向量,把各频带内的能量与总能量之比为特征参数;再对各频带的特征参数进行遗传算法生成新的优化特征参数,以概率一致性原理为基础,通过模糊数学可能性理论得到特征参数的隶属度函数,进而对冲压模具的工作状态进行判定,实验证明,该方法能可靠地对模具进行状态的判别,具有很强的实用性。According to characteristics of an acoustic emission signal induced by a punching die when it failed,the characteristic parameters of failure signal were determined.The energy eigenvector of the failure die signal was extracted using wavelet packet analysis technology.The ration of the energy in different frequency bands to the total energy were taken as the characteristic parameters.Then,using genetic algorithms the characteristic parameters were optimized.The principle of probabilistic consistency and the fuzzy mathematics possibility theory were taken as the foundation to obtain the membership functions of the characteristic parameters.Last,the method was used to identify the state of a punching die.Experiments showed that the proposed method can reliably identify the state of a punching die and it is strongly appliable.

关 键 词：声发射信号 小波包分析 遗传算法 可能性理论 冲压模具 判别 

分 类 号：O536[理学—等离子体物理] TD326[理学—物理]