Coordinated approach fusing time-shift multiscale dispersion entropy and vibrational Harris hawks optimization-based SVM for fault diagnosis of rolling bearing. (March 2021)

• Record Type: Journal Article
• Authors: Shao, Kaixuan; Fu, Wenlong; Tan, Jiawen; Wang, Kai
• Is Part Of: Measurement. Volume 173(2021)
• Publication Date: 2021-03
• Subjects: Fault diagnosis -- Variational mode decomposition -- Time-shift multiscale dispersion entropy -- Vibrational Harris hawks optimization -- Support vector machine; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2020.108580
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: An improved feature extraction method is proposed by aggregating time-shift multiscale technique and dispersion entropy. A novel optimization algorithm called vibrational HHO is proposed to select optimal parameters of SVM adaptively. Effectiveness of VHHO is verified by performance analysis with benchmark functions. Superiority of the novel diagnosis model is ascertained by engineering experiment and comparative analysis. Abstract: To fully mine the effective fault information and improve the fault diagnosis accuracy, a novel fault diagnosis approach for rolling bearings is proposed by integrating variational mode decomposition (VMD), time-shift multiscale dispersion entropy (TSMDE) and support vector machine (SVM) optimized by vibrational Harris hawks optimization algorithm (VHHO). Firstly, vibration signals with different fault types are decomposed into several intrinsic mode functions (IMFs) by VMD. Subsequently, the proposed TSMDE aggregating time-shift procedure and dispersion entropy is employed to extract multiscale fault features from IMFs. Afterwards, the proposed VHHO that adopts a periodic mutation mechanism to enhance the original Harris hawks optimization (HHO) is devoted to search the optimal parameters of SVM, with which different faults are recognized. Finally, simulations and applications are conducted to evaluate the proposed coordinated VMD-TSMDE-VHHO-SVM approach, and the results reveal that the proposed approach can achieve better diagnosis performance than other comparative ones.