Energy Management and Anomaly Detection in Condition Monitoring for Industrial Internet of Things Using Machine Learning

Different concepts of condition and energy monitoring systems in manufacturing facilities have been studied extensively, in relation to the improvement and enhancement of the decision-making processes in industries. Internet of Things (IoT) communication networks has also provided more integrated machine connectivity for real time data, and so its application in industrial processes has enabled effective energy usage and condition monitoring for sustainable management. In this paper, the operational status of the machines categorically ascertained within a short time interval and maintenance is predicted by the system in response on user interface application Node-RED dashboards and Python Shell environment. Furthermore, a portable and scalable wireless sensor network using the IEEE 802.15.4e protocol has been integrated with Machine Learning (ML) algorithm to analyze the anomaly detection in the condition and energy monitoring sensor datasets. As a result, the 99.16% accuracy of this supervised learning model is observed.