AI Technology Enhances Detection of Micro-defects in Stainless Steel

According to the American Nuclear Society website, researchers at Argonne National Laboratory (ANL) in the United States have developed a new technology that combines artificial intelligence (AI) with pulsed infrared thermography (PIT) to accurately detect micro-defects in 3D-printed stainless steel components. These defects, such as pores, typically have diameters smaller than 100 micrometers and are difficult to detect using traditional methods. However, they can significantly weaken the material’s strength, especially in extreme environments like nuclear reactors, where they may lead to serious issues. The related research findings have been published in a recent issue of Scientific Reports.

                                                 3D-Printed Stainless Steel Components

AI Empowerment Enhances Detection

PIT technology uses optical flash lamps to rapidly heat the metal surface, while a high-speed infrared camera records the thermal images formed during heat diffusion and attenuation. When internal defects, such as air-filled pores, are present, they alter the material’s physical behavior, making it difficult for heat to pass through smoothly. AI algorithms process the PIT images, filtering noise and enhancing defect visibility, thereby accurately identifying tiny defects as small as 100 micrometers in diameter. Researchers found that this new method can detect these minuscule defects that were previously undetectable with conventional techniques. This represents an exciting advancement in ensuring the structural integrity of materials used in nuclear applications.

   The principle of "AI + PIT" for detecting internal defects and reconstructing images of internal defects

New Breakthrough

Additive manufacturing, combined with 3D printing technology, is increasingly being used to produce parts for extreme environments, such as the interior of nuclear reactors, where metals must withstand high heat and radiation. The long-term strength and reliability of these components depend on the detection and prevention of defects. Traditional nondestructive testing techniques, such as X-ray imaging and ultrasonic testing, struggle to identify tiny sub-surface defects in complex-shaped 3D-printed structures. In contrast, AI-enhanced PIT offers a non-contact solution that can be scaled to components of any shape and size. This allows manufacturers to inspect the integrity of 3D-printed metal parts during the testing process without damaging them. This technology is particularly suitable for industries such as nuclear energy and aerospace, where material performance requirements are extremely high, enabling early defect detection, preventing failures, and extending component lifespan.

Application Prospects

Researchers are working to improve the technology, aiming to enhance the sensitivity of PIT and optimize AI algorithms for detecting even smaller defects. They also plan to extend the technology to other materials and manufacturing processes. This breakthrough demonstrates the potential of AI in addressing complex challenges in nondestructive evaluation and materials science. By combining advanced imaging technology with machine learning, new possibilities are emerging to ensure the safety and quality of materials used in the most demanding environments.

This advancement not only enhances the quality control of additively manufactured components but also highlights the immense potential of AI in nondestructive evaluation and materials science. In the future, researchers plan to further improve detection sensitivity and extend the technology to more materials and manufacturing processes, providing stronger guarantees for material safety in extreme environments.