Session: 03-15-03: Smart Manufacturing and Robotics for the Future III

Paper Number: 160631

Converting Virtual Commissioning of a Pick and Place Robot Into Digital Twin 

Virtual commissioning is a growing technique in the industry, which allows engineers to validate a new system in a simulated virtual environment before implementing a physical model. The industry has now moved to the use of digital twin systems within the virtual commissioning movement. This approach lowers commissioning time, reduces costly errors, and enhances system reliability by allowing the industry to simulate the process virtually and receive real-time data to correct errors. However, the problem within this movement is the inaccuracy of data and the model limitations of the system. The industry has yet to acknowledge this problem. The motivation of this work is to show how a digital twin can be improved within a simple pick-and-place system and output reliable data that will help the industry move forward while highlighting the gap in research.

To better understand this complexity, this research explores how digital twins can support high-quality operations via real-time monitoring and proactive maintenance strategies. This research aims to highlight the research gap in the industry of the improvement of data collection within digital models. Within our literature review, we will highlight key improvements within industry 4.0/5.0 and show how these can be used within a digital twin system to improve its data capability. Sensor fusion, which integrates various sensor data types into a cohesive data structure, has shown its capability of analyzing system health, detecting faults, and failure prediction. By combining virtual commissioning and real-time sensor fusion, industries have the potential to adopt the benefits of digital validation into real-time operation, enabling continuous adjustment and process improvement without increasing risk to the physical system.

This case study develops and integrates a digital twin for a MiniFESTO robot, reflecting a real-world system environment, through real-time data acquisition and sensor fusion. The primary objective of this case study is to develop a digital twin and measure and analyze the data within the simple model while improving on its accuracy. With the implementation of the digital twin, we can analyze and predict the behavior of the system, giving an efficient way of monitoring performance. To accomplish this, we will apply Siemens Product Lifecycle Management tools, Tecnomatix, TIA Portal, and Manufacturing System Integration (MSI) techniques to carry out our methodology of analyzing 2 variables within the system. To validate our proposed model, we will focus on evaluating the difference between the physical and digital wattage and speed of the system. The differences will highlight potential inconsistencies between the two models, providing valuable insight into the accuracy of digital twins, and help tuning of similar models that can improve performance prediction.

The findings in this research show that it is possible to minimize the error between the physical and digital models. We have shown that through real-time sensor data and synchronization, the digital twin will mirror the behavior of the physical system with +/- 5% accuracy, bettering system predictions and operation. With the reduced error, we can reliably predict performance, leading to improved efficiency. According to the research scope, we were able to develop a digital twin for the robot and manipulate the conveyor belt speed in real time. As a part of further research within the project, we plan to explore different next-gen automated tools within virtual commissioning and find ways to improve the digital twin. Tools like deep learning, machine learning, and industrial Internet of Things methods. This will show the diverse ways a digital can be improved and adapted to different situations and problems within the industry. Our findings demonstrate a strategy on how to implement a digital twin in a day-to-day operation while showcasing the benefit of having a virtual representation of a physical system.

Presenting Author: David Guerra-Zubiaga Kennesaw State University

Presenting Author Biography: Dr. Guerra-Zubiaga is an accomplished academic-industrial executive with a strong background in educational leadership and industry engagement. As an Engineering Faculty Professor with over 16 years of experience, he has played a key role in advancing academic initiatives, fostering student success, and supporting faculty in a shared governance environment.
In addition to his academic career, Dr. Guerra-Zubiaga has 15+ years of full-time experience in research and development within the industry. His expertise lies in developing new products and integrating complex technologies. He has successfully led international industrial projects totaling $14.2 million in research funding. Notably, in 2014 and 2016, he secured a $340 million in-kind software grant from Siemens PLM Software.
His research contributions include 2 patents, 1 book, and over 100 international publications. He has also supervised 25 postgraduate theses. Since 2011, he has served as Senior Associate Editor (North America) for the International Journal of Computer Integrated Manufacturing. Additionally, he has been actively involved with ASME-IMECE since 2017, currently serving as a Track Organizer for Advanced Manufacturing.
Today, Dr. Guerra-Zubiaga is an Associate Professor in the Robotics and Mechatronics Engineering Department at Kennesaw State University and holds the title of Siemens Endowed Professor of Mechatronics.

Authors:

David Guerra-Zubiaga Kennesaw State University
Gershom Richards Kennesaw State University
David Luna Kennesaw State University
Carter Corbin Kennesaw State University
Theodor Myklebusthaug Kennesaw State University
Oscar Tharp Kennesaw State Unicersity
Juan Crisantos Kennesaw State University