Session: 02-13-02: Industry 4.0 Aspects

Paper Number: 68686

Start Time: Tuesday, 01:00 PM

68686 - Intelligent Process Control Following Industry 4.0 Trends 

Industrial Internet of Things (IIOT) and Machine Learning (ML) are currently in development as an important manufacturing paradigm with active contemporary research exploring its trends in enhancing Industry 4.0 on a global scope. Manufacturing encompasses a broad umbrella of technologies and this paper will focus specifically on hydraulic process control implementation through IIOT and ML along Industry 4.0 trends. With the vast amounts and types of data generated by implementing closed-loop systems that communicate instruments and controls through IIOT and ML applications, new automated methods are required to securely manage big data in a robust architecture. The expanding capabilities of IIoT technologies increase the demands of customers for intelligent and interconnected instruments and controls.  ML algorithms improve process control when managing big data collection in closed-loop systems. There is a need to explore new methodologies utilizing IIoT and ML technologies on Manufacturing Data Management (MDM) in order to improve process control, implement big data collection, analyse the data, and provide feedback. New Programmable Logic Controllers (PLCs) move the industry standard forward for manufacturing and process control. As Original Equipment Manufacturers (OEMs) and researchers produce new technologies to bridge the gap, high-value capabilities previously denied to either technology are now possible. The goal of this paper is to propose a new approach analysing IIoT and ML technologies to improve process control using big data collection, analysis, and its feedback to support MDM. Some researchers have been exploring approaches analysing mainly IIoT to support some process controls and other researchers have been exploring mainly ML approaches to support a variety of other process controls. However, very few of them combine both (IIoT and ML) technologies to support hydraulic process control. In some advanced manufacturing systems, the merge of IIoT and ML technologies is the key to solving the complex feedback control for process variables using big data analysis. A new approach to implement both (IIoT and ML) technologies to support process control is part of the research novelty contribution of this paper. Part of the methodology used in this paper demonstrating the research contribution is through a proof of concept using a Process Control Training Bench (PCTB) including two water tanks connected with a network of pipes and a pump to circulate the water through the hydraulic system. The PCTB incorporates four sensors to monitor the main process control behaviour. The closed-loop system uses four different instruments to sense the process variables Pressure (P), Flow (F), Temperature (T), and Vibration (V) making the optimal process control feedback. The four data variables (P, F, T and V) are sensed and transmitted to perform the closed-loop feedback using PLCs and Human Machine Interactions (HMIs). The PCTB uses SIMATIC S7-1500 PLC and IOT 2040 PLC through a conventional Siemens HMI (all from Siemens) to control the hydraulic system. The instruments (sensors) implemented in the hydraulic process control hardware include different manufactures, but the controls (PLCs + HMIs) are Siemens brands. The software to perform the closed loop feedback is Totally Integrated Automation (TIA V 16). Some preliminary results of the current research are that for some cases the P, F, T, and V could be collected and transmitted directly to the PLCs and HMI implementing the control feedback. However, there are cases when the combination of the four variables require using a ML algorithm to implement the closed-loop feedback before the PLC is takes an action. In these cases, the IOT 2040 PLC carry out the MDM not only collecting the massive amount of data in the cloud, but also processing through the ML algorithm to implement the closed-loop feedback. This paper will present a new approach analysing IIoT and ML technologies and use the PCTB to show how to improve hydraulic process control using big data collection, analysis, and its feedback to support the MDM.

Presenting Author: Grayson McMichael Kennesaw State University

Authors:

David Guerra-Zubiaga Kennesaw State University
Grayson McMichael Kennesaw State University
Diana Segura-Velandia Loughborough University
Maria Aslam Loughborough University
Seung-Woo Yim Kennesaw State University
Zack Anderson Kennesaw State University
Yee Mey Goh Loughborough University