As a key component in rotating machinery, the operating reliability of bearing influences the performance and service life of the equipment directly. In order to describe bearing performance degradation(BPD) process e...As a key component in rotating machinery, the operating reliability of bearing influences the performance and service life of the equipment directly. In order to describe bearing performance degradation(BPD) process effectively, an assessment approach combining extension and ensemble empirical mode decomposition(EEMD) was proposed. First, the extension was utilized to construct the matter-element of bearing operating state, and the energy moment of intrinsic mode functions(IMFs) was used as characteristic parameter of the matter-element. Then, to determine classical domains of characteristic parameters, the mathematical statistics method was adopted. Finally, the BPD was analyzed qualitatively and quantitatively according to the comprehensive correlation degree of bearing current operating state related to its healthy state. The analytic results of bearing test-rig show that the proposed method indicates the incipient fault approximately occurring in the 81 st hour, and the method also quantitatively presents the degree of BPD. By contrast, the BPD assessment based on time-domain features extraction method could not achieve the above two results effectively.展开更多
文摘滚动轴承全寿命周期性能退化监测是设备主动维修技术重要的组成部分,对损伤状态进行有效评估可以实现设备接近零停机运行,发挥机器的最大生产力。为有效描绘滚动轴承性能退化趋势,提出一种基于流形学习的模糊C均值(Fuzzy C-means algorithm,FCM)方法。首先提取监测信号的时域、频域特征及小波包时频域特征组成高维特征集,然后按确定的本征维数提取高维特征集的低维流形特征,进而建立基于局部线性嵌入流行学习(Locally linear embedding,LLE)的模糊C均值模型评估轴承当前运行状态。通过IMS滚动轴承全寿命试验,验证了该方法能够有效描绘滚动轴承性能退化阶段,为预知维修提供了重要信息。
基金Project(51575232)supported by the National Natural Science Foundation of ChinaProject(201215020)supported by the Natural Science Foundation of Jilin Province,China
文摘As a key component in rotating machinery, the operating reliability of bearing influences the performance and service life of the equipment directly. In order to describe bearing performance degradation(BPD) process effectively, an assessment approach combining extension and ensemble empirical mode decomposition(EEMD) was proposed. First, the extension was utilized to construct the matter-element of bearing operating state, and the energy moment of intrinsic mode functions(IMFs) was used as characteristic parameter of the matter-element. Then, to determine classical domains of characteristic parameters, the mathematical statistics method was adopted. Finally, the BPD was analyzed qualitatively and quantitatively according to the comprehensive correlation degree of bearing current operating state related to its healthy state. The analytic results of bearing test-rig show that the proposed method indicates the incipient fault approximately occurring in the 81 st hour, and the method also quantitatively presents the degree of BPD. By contrast, the BPD assessment based on time-domain features extraction method could not achieve the above two results effectively.
