New levels of aircraft engine efficiency unlocked with AI model

The European Commission’s “Flightpath 2050” strategy aims to significantly reduce aviation emissions, with more efficient aircraft engines being a key technology to achieve this goal. Researchers at Graz University of Technology (TU Graz) in Austria, through the ARIADNE project, have developed an AI-driven model that accelerates the simulation and optimization of intermediate turbine ducts—critical components that guide airflow between high- and low-pressure turbines in aircraft engines. By leveraging a large database of flow data and testing various AI approaches, the team found that reduced order models, which focus on common data features, dramatically speed up calculations while maintaining acceptable accuracy. This allows for rapid assessment of efficiency changes when design parameters, such as duct length, are varied. The TU Graz team plans to share their extensive turbine duct database and reduced-order model online to enable other researchers to develop advanced three-dimensional simulation models. Their machine learning approach has also uncovered previously unknown dependencies and trends in engine design, opening new avenues for innovation. Given that aviation