Aiming at the challenges of feature extraction from rolling bearing vibration signals and the low accuracy of fault diagnosis,a novel fault diagnosis method based on improved multiscale sample entropy(IMSE)and parameter-optimized variational mode decomposition(VMD)is pro-posed.This method initially employs the IMSE to perform smooth coarse-graining on the original time series,replacing the average value with the maximum value of each sequence to represent the coarse-grained information,thus avoiding the data loss issue inherent in multiscale sample entropy(MSE).By optimizing VMD parameters through a combination of scale spectrum and summation fuzzy entropy,the optimal mode components are obtained,and the reconstruction signal is selected.The IMSE values of the reconstructed signals serve as feature vectors input into a support vector machine for fault diagnosis.Experiment results demonstrate that the proposed method obtains more accurate fault signal features and increases the fault diagnosis accuracy.
关键词
滚动轴承故障诊断/变分模态分解/尺度谱/求和模糊熵/多尺度样本熵
Key words
fault diagnosis of rolling bearing/variational modal decomposition/scale spectra/summed fuzzy entropy/multiscale sample entropy
维普
万方数据