ML-Driven Smart Manufacturing for Accurate Mechanical Fault Detection and Classification
Keywords:
accuracy, anomaly detection, deep learning, fault classification, machine learning, predictive maintenance, reliabilityAbstract
Mechanical faults in industrial machinery can lead to significant operational disruptions, increased maintenance costs, and safety hazards. In 2023, the global market for predictive maintenance technologies was valued at over $10 billion, reflecting the growing emphasis on early fault detection and prevention. Accurate classification of mechanical faults is critical for maintaining equipment
reliability and minimizing downtime. Traditional methods for fault diagnosis often involve manual inspection and heuristic analysis, which can be labor-intensive and prone to errors, particularly when dealing with complex machinery or large-scale operations. Manual fault detection methods typically rely on visual inspections, vibration analysis, and other diagnostic techniques that require substantial
expertise and are often reactive rather than proactive. These approaches can struggle to detect early signs of failure or subtle anomalies, leading to delayed maintenance actions and potential system failures
References
Upadhyay, A.; Alaküla, M.; Márquez-Fernández, F.J. Characterization of Onboard Condition Monitoring Techniques for Stator Insulation Systems in Electric Vehicles—A Review. In Proceedings of the IECON 2019—45th Annual Conference of the IEEE Industrial Electronics Society, Lisbon, Portugal, 14–17 October 2019; Volume 1, pp. 3179–3186.
Xian, R.; Wang, L.; Zhang, B.; Li, J.; Xian, R.; Li, J. Identification Method of Interturn Short Circuit Fault for Distribution Transformer Based on Power Loss Variation. IEEE Trans. Ind. Inform. 2003, 20, 2444–2454.
