Russian Students Develop AI to Detect Pipeline Defects

A national AI competition in St. Petersburg showcased student-built neural networks for energy infrastructure, including a model that detects pipeline damage with 60% accuracy.

At the TECH SQUAD All-Russian AI competition in St. Petersburg, students from technical universities unveiled new tools aimed at digitally transforming the energy sector. The event brought together over 400 students from 34 universities across the country.

During the finals, nine teams worked on algorithms to automatically detect defects in pipeline X-rays. The winning team developed a neural network capable of identifying structural issues with up to 60% accuracy. Other participants introduced mobile diagnostic solutions for on-site inspections and lightweight algorithms designed for use on low-power devices.

TECH SQUAD is part of a broader initiative to build a skilled workforce for the digital transformation of Russia’s fuel and energy sector. Organizers say future events will feature expanded challenges and wider geographic reach, engaging even more young engineers in the process of technological modernization.