摘要
针对滚动轴承故障振动信号的非平稳、非线性特性,将经验模态分解方法和排序熵有机结合,提出一种新的基于自适应尺度的复杂度参数——本征时间尺度排序熵,用于描述不同本征模态分量的复杂程度,从而实现故障特征的量化描述。首先,将原始振动信号经过EMD分解得到若干本征模态分量,然后分别对各本征模态分量计算排序熵,即可得到不同本征时间尺度排序熵,最后利用该参数实现不同故障状态的有效区分与识别。实例分析结果表明了该方法的有效性和实用性,从而为机械设备状态监测与故障诊断提供了一种有效途径。
Aiming at the nonstationary and nonlinear characteristics of bearing fault vibration signals, a new complexity measure based on the adaptive scales, intrinsic time scale permutation entropy, is proposed. It combined the merits of both empirical mode decompostion algorithm and permutation. The new measure can describe the complex degree of different intrinsic mode functions and quantify the fault features. First, the original vibration signals are decomposed into several intrinsic mode functions by EMD algorithm; second, permutation entropies of different IMFs are computed respectively to obtain the different intrinsic time scale permutation entropy; finally, different working states are identified and classified effectively. The case analysis results validate the availability and feasibility of the proposed method. The new method can provide an effective way for condition monitoring and fault diagnosis of mechanical equipments.
出处
《燕山大学学报》
CAS
2013年第2期179-184,共6页
Journal of Yanshan University
基金
河北省自然科学基金资助项目(F2011203149)
关键词
经验模态分解
复杂度
排序熵
故障诊断
滚动轴承
empirical mode decomposition
complexity
permutation entropy
fault diagnosis
rolling bearings
