AI shows promise in unique European trial
As reported in a on Fusion for Energy's website, Senior Technical Officer MarÃa Ortiz de Zúñiga and Cristian Casanova, Vacuum Vessel Programme Manager at the time, conducted a pilot study using data previously collected from vacuum vessel welds with defects to train an AI model to predict which welds in progress would present defects.
"Big Science projects like °ÄÃÅÁùºÏ²Ê¸ßÊÖ offer a wealth of data which is ideal for AI. They provide us with a one-of-a-kind opportunity to learn, train, extrapolate and apply these skills in other fields of manufacturing. There are very few attempts in bringing AI in nuclear engineering, so we decided to give it a try and break new ground. We are the first °ÄÃÅÁùºÏ²Ê¸ßÊÖ party to test the waters and since this pilot project, there has been an unprecedented interest and enthusiasm harnessing its potential," explains MarÃa Ortiz de Zúñiga.
The results were conclusive. After data collected from 2,000 electron beam welds was entered into the model, it predicted with 100% accuracy how many of the 100 welds in progress would pass or fail conformity checks. Similarly, the model was able to predict the results of PAUT testing (phased array ultrasonic testing) with 96% accuracy.
Read the full article on the Fusion for Energy website .