Lin Shuiquan
ABSTRACT
Gearbox fault diagnosis is a critical aspect of ensuring the reliability and efficiency of industrial machinery. This study developed an AI-powered vibration signal fault diagnosis and analysis system to accurately assess gearbox conditions, identify faults, and support predictive maintenance. The research employed descriptive and developmental design, integrating qualitative and quantitative approaches. Key challenges in gearbox fault diagnosis were identified, including structural complexity, non-stationary operating conditions, and limitations of traditional methods. Significant predictors of gearbox faults were determined, including waveform indices (T1, T2, T3) and frequency-based indices (F1, F2, F3). The study utilized time-domain, frequency-domain, and time-frequency domain analysis to improve diagnostic accuracy. An artificial intelligence fault diagnosis system was designed to process large volumes of data in real time, enhancing fault prediction and reducing maintenance costs. The developed system demonstrated a very high level of compliance with the ISO 25010 Software Quality Standard, ensuring its functionality, performance efficiency, reliability, and security. The findings indicate that AI-driven fault diagnosis can significantly improve gearbox monitoring, enabling early fault detection and preventive maintenance. Recommended enhancements include improved signal processing, strengthened data security, remote monitoring capabilities, and an optimized user interface. This research contributes to the advancement of intelligent fault diagnosis technologies, promoting industrial automation and digital transformation
Keywords: Gearbox fault diagnosis, artificial intelligence, vibration signal analysis, predictive maintenance, ISO 25010, industrial automation
https://doi.org/10.57180/vpoh3377