A $1 million award from the U.S. Department of Energy (DOE) supports Nebraska-led research on UMNS: Ultrasonic MEMS Neuromorphic Sensor for Additive Manufacturing In-Process Monitoring. The four-year project spans 2024-28 with funding through DOE’s EPSCoR-State/National Laboratory Partnership Grants program.
The work focuses on Additive Manufacturing (AM), also known as 3D printing: now trending in industries and research. The processes’ merits versus traditional “subtractive manufacturing” include: less waste, faster production, and easier design changes. However, challenges can arise in AM’s production quality; building objects layer by layer with AM, defects can happen, from poor material quality to complex machine settings. Sensors are important during AM production, to address such issues -- yet these sensors produce a huge amount of data. For example, one type of sensor can create over 75 gigabytes of images every second -- more than what Netflix streams! Powerful computers and memory are needed to handle all that data.
A team at the University of Nebraska-Lincoln (UNL) and Iowa State University (ISU) is focusing on analog computing (a different way of processing information than regular digital computers) that listens to sounds made during the AM production process. By analyzing these sounds in real time, this approach can detect defects as they happen -- without storing huge amounts of data. This also makes the system safer from cyberattacks, since no data is saved.
Fadi Alsaleem, associate professor with UNL’s Durham School of Architectural Engineering and Construction, leads the research team with Jinying Zhu, a UNL Civil Engineering professor, and Joseph Turner, a professor at ISU. Together they bring expertise in analog computing, sound sensing, and additive manufacturing; also on board is David Mascarenas from Los Alamos National Laboratory.
According to Alsaleem, this project is the first time anyone has tried to use analog computing for real-time monitoring of AM. The work also connects with DOE goals in advanced and smart manufacturing. He adds that this technology could benefit the national security area of tracking materials. During the project, graduate students gain hands-on experience in areas like computer modeling and machine learning—skills that are valuable in our nation’s workforce development.