A joint team from Gazstroyprom and the Center for Industrial and Manufacturing Solutions at Innopolis University has introduced a solution based on artificial intelligence for assessing pipeline quality. The AI detects pores, slag inclusions, lack of fusion, and other weld seam defects with 94% accuracy.
The core of the method is a digital radiographic technique that uses matrix detectors to capture images of weld areas. The AI system automatically identifies and classifies weld defects, evaluates their size, and takes into account pipeline parameters such as wall thickness, diameter, and internal pressure.
Based on the analysis results, the system provides an accurate report on the condition of weld seams at a given site.
The core of the method is a digital radiographic technique that uses matrix detectors to capture images of weld areas. The AI system automatically identifies and classifies weld defects, evaluates their size, and takes into account pipeline parameters such as wall thickness, diameter, and internal pressure.
Based on the analysis results, the system provides an accurate report on the condition of weld seams at a given site.