Title: Design and analysis of a test rig for simulating faults in hydrodynamic bearings and spectral analysis

Authors: Muhammad Umar Khan; Muhammad Ukasha Bin Khalid; Areeba Akhtar; Muhammad Owais Khan

Addresses: Department of Mechanical Engineering, University of Engineering and Technology, Peshawar, 25000, Pakistan ' Department of Mechanical Engineering, University of Engineering and Technology, Peshawar, 25000, Pakistan ' Department of Mechanical Engineering, University of Engineering and Technology, Peshawar, 25000, Pakistan ' Department of Mechanical Engineering, University of Engineering and Technology, Peshawar, 25000, Pakistan

Abstract: Hydrodynamic bearings play an important role in various industrial applications, underscoring the importance for efficient defect detection and predictive maintenance. In this study a developed test apparatus to simulate bearing faults and employs spectral analysis for fault detection. A 3D model of the test rig was designed using CAD software to simulate defects such as misalignment and imbalance. Vibrational and related sensor data were collected and processed to derive spectral information essential for defect detection. Spectral data were utilised in conjunction with long short-term memory (LSTM) algorithms, which are particularly effective in identifying defect patterns in bearings. The LSTM algorithm was determined to be highly effective, demonstrating significant accuracy in the detection of various defects in hydrodynamic bearings. The findings suggest that integrating spectral analysis with LSTM algorithms offers a powerful approach to reducing maintenance costs and improving equipment reliability in industrial systems.

Keywords: hydrodynamic bearings; fault diagnosis; spectral analysis; predictive maintenance; long short-term memory algorithm; LSTM.

DOI: 10.1504/IJOSM.2025.150835

International Journal of Ocean Systems Management, 2025 Vol.2 No.2, pp.229 - 252

Received: 14 Feb 2025
Accepted: 05 Mar 2025
Published online: 23 Dec 2025 *

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