Metal, melted, mastered

Imagine a fleet of submarines sitting idle on a military base in the Pacific because they contain malfunctioning or aging parts.

Each submarine needs a unique replacement, available only from a single machinist more than 5,000 miles away. After months of waiting, the parts finally arrive — only for mechanics to discover a defect, and the process has to start over. 

This has been the reality for decades. Researchers at Virginia Tech see a more efficient way forward. 

A new study led by Prahalada Rao, associate professor in the College of Engineering, could reshape the way submarines and aircraft are built. He recently published research in the journal Materials and Design that shows how using artificial intelligence (AI) to monitor wire-arc additive manufacturing — essentially welding in 3D — can detect flaws as parts are built, correct them in real time, and ensure they’re ready to use once the print is completed.

The breakthrough comes at a time when the Navy urgently needs faster, more reliable production to maintain its fleet. 

“We’ve always relied on conventional machining, but it takes months to produce even a single part,” Rao said. “Additive manufacturing gives us the ability to make those parts much faster and with less waste, which opens up a new way of thinking about how we build.” 

Good melt, bad melt 

The defense and aerospace supply chains previously relied on mom-and-pop machine shops. These small operations had the expertise to machine, cast, or forge critical components for submarines and aircraft from solid blocks of metal. But their methods were slow, wasteful, resource intensive, and overreliant on a shrinking workforce. Flaws often weren’t discovered until a part was finished, wasting weeks of labor and forcing manufacturers to scrap parts.

After the Cold War, many of these shops closed, and the skills retired with them. Additive manufacturing has helped fill that gap, offering a faster way to produce complex components and dramatically shortening the time it takes to get parts into service. 

Not all additive manufacturing works the same way. Some methods are slow and suited for small, intricate parts, like laser powder bed fusion. Rao is currently working on faster approaches known as wire-arc printing and laser-wire printing. 

“Wire-arc additive manufacturing is basically welding in 3D,” Rao said. “If laser powder bed fusion additive manufacturing produces a pint of material a day, wire-arc is keg-sized. You can deposit 40 or 50 kilograms of material in just one day. The challenge is making sure that much metal goes down without a single flaw.” 

That’s where AI comes in.

Instead of discovering cracks or pores after a part is finished, Rao’s team is training algorithms to spot the warning signs as the metal is being laid down. By studying the “melt pool” of hot metal and comparing what “good” and “bad” prints look like, AI learns to recognize defects in real time and signal adjustments before they spread.

“When the melt pool looked good, the part turned out how we wanted. When it looked bad, we knew what would happen,” Rao said. “So we built a machine learning algorithm that was able to predict with about 90 percent certainty when things were going wrong.”

Mom-and-pop turned Industry 4.0

Rao is part of a larger team of researchers in Virginia Tech Made: Center for Advanced Manufacturing. Virginia Tech Made cultivates cross-campus collaborations, expands partnerships with industry and government, and trains students and manufacturing professionals based on the university's expertise in advanced materials, manufacturing technologies, computational design, data analytics, and digital infrastructure.

As part of the Grado Department of Industrial and Systems Engineering, Rao applies his background in systems thinking to make manufacturing not just faster, but smarter.

“Faster, better, cheaper,” Rao said. “I want to make it better through quality control, faster by not wasting time redoing parts, and cheaper by reducing defects. That’s something we do very well in process control.”

Rao and his team carry out much of their work in his Kelly Hall lab, but they’re also making use of the Learning Factory, Virginia Tech’s hands-on hub for manufacturing education. A smaller laser powder bed fusion machine and a hybrid laser wire printing machine were recently added to the facility — a resource Rao credits to the foresight of department head Eileen Van Aken. He said her push to secure the equipment was key, both for advancing the research and for giving students the chance to train on the same technology used in industry. 

“Giving students access to the same machines they’ll see in industry is critical,” Rao said. “That’s why the laser powder bed fusion and laser wire system in the Learning Factory is so valuable. Additive is something you can train quickly, and it prepares our students to step right into the future of manufacturing.”

Original study:DOI: 10.1016/j.matdes.2025.114598

---

---