Intelligent Monitoring and Defect Detection System for Powder Bed Fusion Process

Opportunity

AM, or 3D printing, has emerged as a promising manufacturing technology as it offers design freedom, accelerates time-to-market and reduces material wastage compared to conventional subtractive processes. However, ensuring that parts built via AM are consistent and meet quality requirements remains a key challenge. Inconsistencies in AM-built parts and the lack of optimised non-destructive evaluation (NDE) methods for AM processes have hindered the widespread adoption of AM, especially in industries where product certification is crucial. This invention presents a monitoring system and defect detection tool for real-time and in-situ inspection of the PBF AM process. It will benefit industrial partners, particularly AM service bureaus or AM system operators.

Technology

This technology employs a sensor fusion methodology using an optical camera and an infrared camera, providing a multi-control approach for real-time quality control of the PBF process. The defect detection tool is equipped with deep-learning capabilities using convolution neural networks for real-time defect recognition, defect classification, as well as establishing the relationships between defect signatures and quality-control metrics of the fabricated products. Compared to other process-monitoring solutions, this technology monitors a field of view beyond just the melt pool, offering observation of defect signatures, defect detection and classification, analysis of defects from acquired images, and quality evaluation of printed parts.

Figure 1: Defect detection and classification using convolutional neural networks. 

Applications & Advantages

This technology is ideal for PBF AM operators, such as service bureaus and equipment owners and may serve as the basis for the standardisation and certification of AM products.

Advantages:

• Improves print consistency, part quality and reliability.
• Minimises material and time wastage by providing real-time feedback on part quality during fabrication.
• Increases manufacturing efficiency and promotes greater acceptance of PBF parts in industries such as aerospace, automotive and medical.
• Monitors a broader field of view, capturing defect signatures beyond just the melt pool.