AN AUTOMATED DATA LOGGER SYSTEM FOR REAL-TIME MONITORING AND ANOMALY DETECTION IN INDUSTRIAL IOT ENVIRONMENT

Abstract

In Industrial Internet-of-Things, a data logger must possess critical features such as real-time data acquisition, scalable storage capabilities, robust anomaly detection, and efficient dashboard integration for user-friendly monitoring, ensuring comprehensive data management and system reliability across industrial environments. Nevertheless, current data loggers offer very little data storage, have few intelligent features, and frequently have an interface that is difficult to use. Additionally, these loggers struggle with efficient data management, leading to storage issues and poor user experience. The integration of Industrial Internet of Things technology facilitates efficient mass data collection by enabling seamless connectivity and real-time monitoring. In this work, a system that features a user-friendly dashboard, enhanced with Grafana for advanced data visualization and management, built on Node-RED for flexible and streamlined development was proposed. A Raspberry Pi was chosen as a gateway due to its capability to process real-time data and send the data to the database. The system is capable of reading data from multiple sensors, which is stored in InfluxDB, a reliable time-series database. Moreover, the dashboard supports factory workflow and environmental monitoring from any location. The system also alerts users when an anomaly is detected, enabling proactive management and timely response. The anomaly message was sent directly from Raspberry Pi to reduce processing time, as demonstrated in the performance test results. The developed product underwent user evaluation, scoring grade A in usability testing with an impressive score of 91.25%, indicating a high level of user satisfaction and effectiveness.