Title: Rub-impact fault identification based on EMD and stochastic resonance

Authors: Mingyue Yu; Jinghan Zhang; Liqiu Liu

Addresses: Shenyang Aerospace University, Shenyang, China ' Shenyang Aerospace University, Shenyang, China ' Shenyang Aerospace University, Shenyang, China

Abstract: An approach combining empirical mode decomposition (EMD) and adaptive stochastic resonance (SR) has been brought forward to make effective identification of rub-impact fault. Firstly, vibration signals were decomposed by EMD to obtain intrinsic modal function (IMF). Secondly, concerning about the different sensibility of IMFs to fault characteristic information, two signal evaluation indexes, margin factor and information entropy, have been brought in to choose the sensitive IMFs from the wear degree and uncertainty of signal, which can embody fault characteristic information better and make signal reconstruction. Thirdly, to further strengthen the characteristic information of fault, information entropy was chosen as fitness function of artificial fish swarm algorithm (AFSA) to optimise the parameter of adaptive SR and give SR treatment to reconstructed signals. Finally, according to the frequency spectrum of signal after SR, rub-impact fault is identified. The result indicates that the proposed method can correctly identify rub-impact faults.

Keywords: stochastic resonance; rub-impact fault; information entropy; margin factor; feature extraction.

DOI: 10.1504/IJISE.2024.138029

International Journal of Industrial and Systems Engineering, 2024 Vol.46 No.4, pp.509 - 530

Received: 13 Jun 2022
Accepted: 24 Jun 2022
Published online: 17 Apr 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article