A mechanical part fault diagnosis method based on improved multiscale weighted permutation entropy and multiclass LSTSVM

•The improved VMD (IVMD) is proposed and used as a noise reduction filter, which improves the anti-noise performance of IMWPE.•A novel fine composite coarse-grained method is proposed in IMWPE, which reduces signal discontinuity and information loss.•A new weighting method is proposed in IMWPE, the permutation order and amplitude information of the signal are considered.•LSTSVM is extended to multi-class LSTSVM through one vs one (OVO) or binary tree (BT) strategies, and combined with IMWPE to achieve fault diagnosis. Multi-scale permutation entropy (MPE) has poor stability, accuracy and noise immunity, recognition accuracy of twin SVM is low, so a diagnosis method based on improved multi-scale weighted permutation entropy (IMWPE) and least squares TSVM (LSTSVM) is proposed. The innovations include that the improved variational mode decomposition (IVMD) based on energy and mutual information is used to suppress noise, and novel composite coarse-grained and permutation pattern weighting methods are introduced to construct IMWPE, after which the noise immunity, stability and accuracy of features have been improved. LSTSVM is extended to a multi-class LSTSVM through one vs one (OVO) and binary tree (BT) strategies, the performance of the proposed methodology is significantly improved. The results show that signal-to-noise ratio of signal is enhanced, the stability of IMWPE is improved, the diagnosis accuracies for bearing and check valve are increased by 17.084 % and 12.498 %, which provides a reference for fault diagnosis under noise.