An adaptive slope entropy combined with hierarchical entropy applied to rolling bearing fault diagnosis
Zhe Zhang, Yingwei Liu, Yuxuan Han, Pengfei Huangfu, Zhiyuan Ma, Weichen Shi, Ke Feng
Abstract
This article proposed an improved slope entropy (Islope) algorithm. In the original algorithm, the state mode is determined by parameters α and β. However, the wide range of parameter choices and strong randomness in the original algorithm may decrease slope entropy capability when unreasonable parameters are selected. Therefore, an adaptive parameter selection is proposed. In order to better demonstrate that Islope has better extraction ability, a robustness test is carried out, and the Islope can work better in the noise state. On the other hand, the single scale can’t fully express the time series. For this reason, the Islope is combined with the algorithm of hierarchical entropy. A hierarchical improved slope entropy is proposed. This method was applied to two experimental tests. At the same time, the sparrow search algorithm is used to optimize the parameters of the support vector machine, and the final experimental result is verified 100%.