Shao will advance smart ultrasonic metal welding with NSF CAREER award
Assistant Professor Chenhui Shao has won a NSF Faculty Early Career Development (CAREER) award for his project “Dynamic Process-Attribute-Data-Performance Modeling to Enable Smart Ultrasonic Metal Welding.” The CAREER Program offers NSF’s most prestigious awards in support of early-career faculty who have the potential to serve as academic role models in research and education and to lead advances in the mission of their department or organization.
The award will provide Shao with a five-year $500K grant dedicated to research on the fundamentals of ultrasonic metal welding. By establishing relationships between process conditions, microstructural weld attributes, online sensing data, and weld performance, Shao plans to advance the fundamental understanding of process mechanisms in ultrasonic metal welding. He also aims to create a suite of machine learning-based decision-making tools that will ultimately enable smart ultrasonic metal welding.
“All these deterministic and random factors can help determine the final quality, but we don’t know why. People have some insights about them, but they are not linked,” said Shao. “By using data science, we can potentially link them together to have a better map of the relationships between process conditions, microstructural changes, weld attributes, and sensing data to predict and control weld quality.”
Ultrasonic metal welding is a welding process which uses a clamping force and high-frequency vibrations to join thin layers of metal. This process can be used to join dissimilar metals, making it highly relevant to the electric vehicle and electronic packaging industries.
The award will also fund educational activities, both at the college and K-12 levels. Shao recently received approval for his Manufacturing Data and Quality Systems (ME 498) course to become a permanent course in the MechSE department, under the new name Data Science in Manufacturing Quality Control (ME 453).
“I really want to prepare our students to be ready for industry because as far as I was concerned, we did not have such classes when I was a student in mechanical engineering. There are many machine learning classes on campus, but they are not teaching machine learning in the engineering context,” said Shao. He uses real-world data to teach students about machine learning and quality control. “I have several collaborators from industry and I use my work with them as examples in class.”
At Dr. Martin Luther King Jr. Elementary School in Urbana, Shao has taught students about concepts of spatial sampling and inference with his cookie puzzle activity and plans to continue to use this activity for K-12 outreach. He hopes that introducing aspects of manufacturing engineering to students at a young age will increase interest in the field and help create the next generation of manufacturing engineers.
Shao believes this award will help him establish a good foundation for future research.
“This is a long-term award that will help me establish myself in this field. My vision for my career is to combine data science, controls, and manufacturing in the smart manufacturing area. This combination is unique and very important because many state-of-the-art data science tools are purely data-driven and do not incorporate or explain physics.”