Full-Line Idler Fault Monitoring in Belt Conveyors via UWFBG-DAS and Characteristic Energy Feature Analysis

Area of Science:

• Engineering
• Materials Science
• Signal Processing

Background:

• Belt-conveyor idler monitoring is difficult due to varying stiffness and industrial vibrations.
• Existing methods struggle with long-distance and heterogeneous industrial environments.

Purpose of the Study:

• To develop and validate a novel sensing method for reliable, long-distance monitoring of belt-conveyor idlers.
• To address challenges posed by spatially varying structural stiffness and strong industrial vibrations.

Main Methods:

• Development of an ultra-weak fiber Bragg grating distributed acoustic sensing (UWFBG-DAS) system.
• Utilized finite-element modeling to optimize sensor deployment.
• Employed envelope demodulation and variational mode decomposition (VMD) for fault-sensitive signal isolation.
• Implemented zone-specific self-referencing thresholds for energy accumulation analysis.

Main Results:

• Successfully deployed 1.2 km of sensing cable in a coal-fired power plant.
• Observed characteristic energy increases in 9 out of 10 idler replacement tests.
• Achieved 90.3% classification accuracy for three representative fault types using stratified five-fold cross-validation.

Conclusions:

• The UWFBG-DAS method combined with characteristic energy analysis is feasible for long-distance idler monitoring.
• The approach is effective even under spatially heterogeneous industrial conditions.
• Demonstrated high accuracy in fault detection and classification for belt-conveyor idlers.