Edge Architecture for the Integration of Soft Models Based Industrial AI Control into Industry 4.0 Cyber-Physical Systems

@Article {
author = {Garcia, Ander},
title = {Edge Architecture for the Integration of Soft Models Based Industrial AI Control into Industry 4.0 Cyber-Physical Systems},
pages = {67},
keywds = {
Artificial Intelligence; Control; Cyber-Physical Systems; Edge Computing; Industry 4.0
}
abstract = {

Traditionally PLC and SCADA systems programmed by automation engineers have been responsible for the control of industrial machines and processes. Industry 4.0 paradigm has merged OT and IT domains, proposing new alternatives for this task. Industry 4.0 approaches start capturing OT industrial data and making it available to the IT domain. Then, this data is visualized, stored and/or analyzed to gain insights of the industrial processes. As a final step, AI models access real-time data to generate predictions and/or control industrial processes. However, this process requires OT and IT knowledge not present in many industrial companies, mainly SMEs. This paper proposes a micro-service edge architecture based on the MING (Mosquitto, InfluxDB, Node-RED and Grafana) stack to ease the integration of soft AI models to control a cyber-physical industrial system. The architecture has been successfully validated controlling the vacuum generation process of an industrial machine. Soft AI models applied to real-time data of the machine analyze the vacuum value to decide when the most suitable time is (i) to start the second pump of the machine, (ii) to finish the process, and (iii) to stop the process due to the detection of humidity

}
isbn = {978-303142535-6},
date = {2023-01-01},
}