Predictive Maintenance of Industrial Appliances Integrated with Machine Learning

Abstract

This paper presents a novel approach to enhance the maintenance paradigm of electric fans through the integration of machine learning and Internet of Things (IoT) technologies. The system employs an Arduino Uno R3 microcontroller to collect real-time data from vibration, temperature, and humidity sensors strategically placed on the fan. Leveraging this data, a machine learning model is developed to predict potential faults or anomalies, enabling proactive maintenance actions. The methodology involves the creation of a comprehensive dataset through continuous monitoring of the fan's operating conditions. The paper details the integration of sensors, the architecture of the system, and the implementation of machine learning algorithms for predictive analysis. The chosen model demonstrates its efficacy in forecasting maintenance requirements, showcasing 90% accuracy in the experiment.