A novel intelligent detection method for rolling bearing based on IVMD and instantaneous energy distribution-permutation entropy

Due to the strong non-stationary properties of the vibration signal, fault diagnosis of rolling bearing under different working conditions has become a difficulty. To address this issue, a new scheme based on improved variational mode decomposition (IVMD) and instantaneous energy distribution-permutation entropy (IED-PE) is developed for recognizing fault category of the rolling bearing. In this approach, IVMD with cross-correlation criterion is provided to decompose the collected data samples into several sub-signals and determine adaptively the mode number. Next, a novel feature extraction technique named IED-PE is proposed to obtain the three-dimensional (3D) eigenvector, which can improve the recognition degree of fault category. Finally, 3D eigenvector is imported into k-nearest neighbor (KNN) classifier for achieving the multi-fault recognition. Experimental studies show that the presented scheme is not only capable of extracting accurately fault features, but can distinguish availably multi-class fault patterns. The research offers a new perspective for intelligent fault detection of rolling bearing.