欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出了一种本征时间尺度分解模糊熵和GG模糊聚类的滚动轴承故障诊断方法。首先,将滚动轴承的振动信号进行ITD分解,得到若干个固有旋转分量和一个趋势项。然后,将PR分量分别与原始信号进行相关性分析,筛选出前3个含主要特征信息的PR分量,并将筛选的PR分量的模糊熵作为特征向量。最后,将特征向量输入到GG分类器中进行聚类识别。通过模糊熵、样本熵和近似熵对比,实验结果表明模糊熵能更好的表征故障信号的特征信息;通过GG聚类、GK聚类和FCM聚类对比,实验结果表明GG聚类效果明显优于FCM、GK的聚类效果。因此,实验证明了基于ITD模糊熵和GG聚类的滚动轴承故障诊断方法的有效性和优越性。
英文摘要：
A new method for rolling bearing fault diagnosis based on the intrinsic time-scale decomposition（ITD） fuzzy entropy and Gath-Geva（GG） clustering algorithm is introduced. Firstly, rolling bearing vibration signal is decomposed with ITD to obtain several proper rotation（PR） components and a tendency item. Secondly, the first three PR components, which content the primary feature information, were chosen by the criteria of correlation with the original signal, and the fuzzy entropies of each PR component are composed eigenvectors. Finally, the constructed eigenvectors are put into GG classifier to recognize different fault types. In the experiment, the fuzzy entropy compares with sample entropy and approximate entropy, the comparison experimental results show that the fuzzy entropy can characterize the feature information of the fault signal better; the GG clustering compares with Gustafaon-Kessel（GK） clustering and fuzzy center means（FCM） clustering. The comparison experimental results show that the result of GG clustering is superior to FCM and GK＇s. So, experimental results show that the rolling bearing fault diagnosis method based on ITD fuzzy entropy and GG clustering is effective and superior.
同期刊论文项目
同项目期刊论文
期刊信息
位置：