Predictive Maintenance of Gearbox: A Cost-Effective IoT Approach for Remaining Useful Life Estimation

Rotating components, such as gearboxes, play a crucial role in various industrial applications by transferring speed and torque. Monitoring the health of gearboxes is essential to enhance productivity and reduce maintenance costs. This study focuses on predictive maintenance techniques for gearboxes, with a specific emphasis on calculating the remaining useful life (RUL) of the system. The primary objective of this research is to identify faults and estimate the RUL of gearboxes. Time domain data, obtained from gearboxes using NodeMCU (ESP8266), is utilized for RUL calculation. Real-time data is captured through the Internet of Things (IoT) platform called ThingSpeak. Machine learning algorithms implemented in MATLAB are employed to estimate the RUL. Notably, the wear along the face width yielded the maximum RUL in this investigation. The present research introduces a cost-effective and user-friendly IoT technique for predicting the RUL of gears. By employing IoT-based condition monitoring and predictive maintenance, industrial stakeholders can optimize maintenance schedules, minimize downtime, and reduce overall maintenance costs associated with gearbox failures.