[2B2] Porosity evaluation of additively manufactured parts using artificial intelligence for automated defect recognition based on artificial intelligence models
I Britto-da-Silva, E Costa-Santos and J Schulz
Carl Zeiss Industrielle Messtechnik GmbH, Germany
Additive manufacturing (AM), also known as 3D printing, has revolutionised various industries by enabling the creation of complex and customised functional components. However, ensuring the quality and integrity of these components remains a significant challenge. Traditional quality inspection methods often fall short due to the intricate geometries and internal structures characteristic of AM parts. This paper explores the primary challenges in quality inspection for AM, focusing on the detection of internal defects.
To address these challenges, we propose the integration of artificial intelligence (AI) with industrial computed tomography (CT) scanning as a comprehensive solution. Industrial CT scanning provides high-resolution, non-destructive imaging capable of revealing internal features and defects that are otherwise inaccessible. When coupled with advanced deep learning algorithms, these scans can be analysed more economically, efficiently and accurately than with conventional methods. AI enhances and speeds up the inspection process by automating defect detection, facilitating the analysis of large datasets to identify patterns and anomalies with great repeatability.
The integration of AI with CT scanning offers several advantages: improved accuracy and reliability of defect detection, reduced inspection times and an enhanced ability to segment and classify different patterns. This approach not only ensures higher quality standards for AM parts but also supports the scalability of additive manufacturing in critical industries such as aerospace, medical devices and automotive. The paper concludes with experimental results demonstrating the effectiveness of AI-driven CT scan analysis in real-world applications, comparing with traditional methods and exploring its potential to transform quality inspection in additive manufacturing.
To address these challenges, we propose the integration of artificial intelligence (AI) with industrial computed tomography (CT) scanning as a comprehensive solution. Industrial CT scanning provides high-resolution, non-destructive imaging capable of revealing internal features and defects that are otherwise inaccessible. When coupled with advanced deep learning algorithms, these scans can be analysed more economically, efficiently and accurately than with conventional methods. AI enhances and speeds up the inspection process by automating defect detection, facilitating the analysis of large datasets to identify patterns and anomalies with great repeatability.
The integration of AI with CT scanning offers several advantages: improved accuracy and reliability of defect detection, reduced inspection times and an enhanced ability to segment and classify different patterns. This approach not only ensures higher quality standards for AM parts but also supports the scalability of additive manufacturing in critical industries such as aerospace, medical devices and automotive. The paper concludes with experimental results demonstrating the effectiveness of AI-driven CT scan analysis in real-world applications, comparing with traditional methods and exploring its potential to transform quality inspection in additive manufacturing.
