Industrial IOT Data Processing and Failure Prediction using Machine Learning Algorithms on Edge Device
This paper presents a work for improving performance of industrial internet of things (IIoT) system through data processing at the edge layer using edge computing.Shock and vibration are practically constant in industrial machinery, causing fatigue and wear on parts and equipment. The result is inaccuracy and finally system failure. In any industry failures must be as low as possible. Understanding possible issues in order to implement preventative measures, maintenance, or equipment replacement is crucial for maintaining performance and avoiding costly downtime and damage. Failures and needless maintenance can affect the overall productivity of the industry. Predictive maintenance (PdM), is one of the most notable use case of industrial internet of things. Idea of improvising industrial IoT using failure prediction through data processing in the edge layer is proposed. Synthetic data is used to predict the failure in the system and computation is done on the edge layer.
Keywords - Predictive Maintenance, Machine Learning, Data Analytics, Edge Computing.