Fault diagnosis of gearbox and electromechanical system using hybrid deep learning architectures

Abstract

Gearboxes, being essential elements in sectors such as manufacturing, transportation, and power generation, are highly susceptible to failures, including gear cracks, misalignment, and wear. Such faults can cause catastrophic system breakdowns, prolonged production downtimes, and costly repairs. Early detection of these faults is crucial for preventing system failures and ensuring smooth operation. Similarly, electromechanical (EM) systems are widely used in industries for various applications. EM systems mostly have an electric motor as a prime mover and a mechanical load, such as a rotor, gearbox, pumps, etc., coupled. EM systems may have combined faults, i.e., faults in motors and faults in loads. Diagnosing combined faults is challenging due to overlapping symptoms and their compounded effects. Hence, advanced fault detection and classification methods are necessary to improve the reliability of gearboxes and EM systems, optimize maintenance scheduling, reduce downtime, and enhance productivity while cutting costs.