Machine Learning Motor Vibration Monitoring System with a Service Estimation Date

Abstract

This paper presents machine learning motor vibration with service estimation date. AI and machine learning algorithms are used to evaluate electric motor vibration patterns and predict maintenance and repair needs. Efficiency, downtime, and maintenance and repair schedule optimisation are project goals. Over time, machine learning algorithms analyse electric motor data to identify vibration patterns. This will predict maintenance and repair needs. KNN, CatBoost, and Neural Network were studied. Machine learning algorithms predicted maintenance and repair needs with over 90% accuracy. Algorithms also calculated the service estimation date, improving maintenance and repair scheduling. It improved maintenance and repair programmes, reduced downtime, and increased reliability. An ESP8266 and a vibration sensor to record and send electric motor data to ThingsSpeak, an IoT website that analyses it. Machine learning algorithms is then used to classify the motor vibration to determine the service estimation date. This project taught me how to maintain and repair electric motors using machine learning and AI algorithms.