Water ingressed lubrication oil degradation with cavitation effect detection in gearbox using sound pressure level analysis

Abstract

This study introduces a novel methodology for diagnosing water ingress-induced lubrication oil degradation and cavitation effects in industrial gearboxes using Sound Pressure Level (SPL) analysis, offering a real-time, non-invasive alternative to conventional vibration-based or spectroscopic techniques. Experimental investigations reveal a strong correlation between elevated SPL values and water contamination in lubrication oil, enabling early detection of gearbox failures while emphasizing the importance of maintaining optimal oil levels to minimize mechanical losses, enhance efficiency, and extend operational lifespan. To address the complexities of fault diagnosis under interaction effects between gear faults and oil types, a machine learning-based classification framework was developed using SPL features, achieving 96.3% accuracy in detecting lubrication degradation and in classifying combined faults. These findings validate SPL as a cost-effective and scalable diagnostic tool that complements existing condition monitoring techniques, offering a comprehensive approach to gearbox health assessment. By integrating SPL analysis with other methods, this research paves the way for intelligent maintenance strategies that enhance gearbox reliability, reduce downtime, and deliver significant cost savings in industrial applications.