Title: A method for extracting bearing degradation data of ship propulsion systems based on cumulative features

Authors: Hui Ma; Chunlong Ma; Wenjun Xia

Addresses: School of Mechano-electronic Engineering, Suzhou Vocational University, No. 106 Zhineng Avenue, Wuzhong District, Suzhou, Jiangsu, 215000, China; School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China, No. 2006 Xiyuan Avenue, Pidu District, Chengdu, Sichuan, 611731, China ' School of Automobile and Traffic Engineering, Heilongjiang Institute of Technology, No. 999 Hongqi Avenue, Daowai District, Harbin, Heilongjiang, 150000, China; College of Shipbuilding Engineering, Harbin Engineering University, No. 145 Nantong Street, Nangang District, Harbin, Heilongjiang, 150000, China ' School of Automobile and Traffic Engineering, Heilongjiang Institute of Technology, No. 999 Hongqi Avenue, Daowai District, Harbin, Heilongjiang, 150000, China

Abstract: This article proposes a method for extracting degradation data of rotating components in ship transmission systems based on accumulated features, aiming to address the challenges faced by traditional methods in extracting early degradation features and improve the accuracy and reliability of prediction models. By combining multi-resolution signal decomposition techniques and accumulative feature processing, the proposed method refines the classical features and inverse trigonometric function features, extracting monotonic degradation signal features. At the same time, the EM smoothing method is used to smooth the feature signals, further improving the availability and accuracy of the data. The experimental results show that this method can significantly improve the accuracy of life prediction for rotating components, providing strong support for the maintenance and optimisation of marine transmission systems. The effectiveness of accumulating feature processing in enhancing data prediction capabilities was verified through monotonicity, correlation, and trend evaluation indexes.

Keywords: degradation data; bearing; cumulative features; ship propulsion systems; expectation maximisation.

DOI: 10.1504/IJOSM.2025.150800

International Journal of Ocean Systems Management, 2025 Vol.2 No.2, pp.127 - 152

Received: 19 Oct 2024
Accepted: 25 Jan 2025
Published online: 23 Dec 2025 *

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