Abstract
Welding Engineering is one of the most integral parts of manufacturing engineering, which involves joining two different materials via various types of processes. Complexities in welding engineering are still a challenge for welders, and ML implementation will be the most appropriate solution for weld quality assessment. The weld joints assessments area challenge, as the welders may not be able to detect the very basic issues inside the welded joint. Using image processing concepts and sensors to find the issues within the weld using the real-time monitoring supported by ML models implemented in various research works. This review rigorously presents all the ML models and sensors used for weld quality monitoring and its assessment for internal fatigues, cracks, or any other flaw.