CAJ | 학술논문

针对滚动轴承故障振动信号的强噪声背景以及现实中不易获取大量典型故障样本的特点,提出一种基于柔性形态滤波和支持矢量机(Support vector machine,SVM)的滚动轴承故障诊断方法.柔性形态滤波既可以有效地提取出信号的边缘轮廓和信号的形状特征,同时又具有稳健性;SVM具有良好的分类性能,特别在小样本、非线性及高维特征空间中具有较好的推广能力;SVM分类器的惩罚因子和核函数参数采用经典粒子群优化算法进行优化,避免传统方法对初始点和样本的依赖.首先对振动信号进行柔性形态滤波,然后提取滤波后信号的故障特征频率的归一化能量为特征矢量作为SVM分类器的输入参数,用于区分滚动轴承的外圈、内圈和滚动体故障,SVM分类器的参数采用标准粒子群优化算法进行优化.试验结果表明了方法的有效性.
침대곤동축승고장진동신호적강조성배경이급현실중불역획취대량전형고장양본적특점,제출일충기우유성형태려파화지지시량궤(Support vector machine,SVM)적곤동축승고장진단방법.유성형태려파기가이유효지제취출신호적변연륜곽화신호적형상특정,동시우구유은건성;SVM구유량호적분류성능,특별재소양본、비선성급고유특정공간중구유교호적추엄능력;SVM분류기적징벌인자화핵함수삼수채용경전입자군우화산법진행우화,피면전통방법대초시점화양본적의뢰.수선대진동신호진행유성형태려파,연후제취려파후신호적고장특정빈솔적귀일화능량위특정시량작위SVM분류기적수입삼수,용우구분곤동축승적외권、내권화곤동체고장,SVM분류기적삼수채용표준입자군우화산법진행우화.시험결과표명료방법적유효성.
Based on soft morphological filters and support vector machine (SVM), a roller beating fault diagnosis method is proposed. It is very difficult to filter the fault vibration signals from the strong noise background because the roller bearing fault diagnosis is a problem of multi-class classification. Soft morphology filter can not only identify the features of fringe and shape of the signals but also give improved performance under certain conditions. Support vector machine has good classification performance especially in the small-sample, nonlinear and high dimensional features and so on. The penalty factors and kemel parameters of SVM are optimized by using particle swarm optimization to avoid dependence on initial parameters and training samples. Firstly, vibration signals are filtered by the soft morphological filters. Secondly, the normalized energy of the different characteristic frequencies is utilized to identify the fault features of input parameters of SVM classifier. The SVM parameters are optimized by using the canonical particle swarm optimization. The experiment results indicate that the modeling method is correct.