摘要
针对刀具磨损声发射信号的非平稳特征和BP神经网络学习算法收敛速度慢、易陷入局部极小值等问题,提出了基于经验模态分解和最小二乘支持向量机的刀具磨损状态识别方法.首先对声发射信号进行经验模态分解,将其分解为若干个固有模态函数之和,然后分别对每一个固有模态函数进行自回归建模,最后提取每一个自回归模型的系数组成特征向量,特征向量被分为两组,一组用于对最小二乘支持向量机训练,另一组用于识别刀具磨损状态.试验结果表明:该方法能很好地识别刀具磨损状态,与BP神经网络相比具有更高的识别率.
In view of the non-stationary characteristics of acoustic emission signal of tool wear,and the slow convergence rate of learning algorithm and easily dropping into the local minimum value for back propagation neural networks,a novel method of tool wear state identification based on empirical mode decomposition and least squares support vector machine was proposed.Firstly,the empirical mode decomposition method was used to decompose the collected acoustic emission signals into a number of stationary intrinsic mode function,and then autoregressive model of each intrinsic mode function was established respectively.Finally,auto regression model coefficients were selected to constitute the feature vector.The feature was divided into two groups,one group was used to train the least squares support vector machine and the other was used to identify the tool wear state.The identification result proves that this method is superior to neural network,and it has a higher identification rate.It is proved that this method is efficient and feasible.
出处
《北京航空航天大学学报》
EI
CAS
CSCD
北大核心
2011年第2期144-148,共5页
Journal of Beijing University of Aeronautics and Astronautics
基金
辽宁省教育厅重点实验室资助项目(LS2010117)
关键词
刀具磨损状态识别
最小二乘支持向量机
经验模态分解
自回归模型
tool wear condition monitoring
least squares support vector machine
empirical mode decomposition
auto regressive model