Advanced Design of Heat Exchangers using multiscale models and machine learning

Heat exchangers (HXs) are vital components in energy systems, including aviation engine cooling. As aircraft systems become more compact and efficiency demands rise, the need for advanced HX design is more pressing than ever. The MSCA-funded ADeHEx project will design custom compact HXs for the engines of tomorrow. Specifically, it combines topology optimisation (TO) and additive manufacturing (AM) techniques to create efficient HXs. While TO for thermo-fluidic problems has advanced, its industrial applications have been hindered by computational costs. As such, ADeHEx’s goal is to develop a comprehensive design methodology for multi-scale 3D fluid-to-fluid HXs using machine learning-based de-homogenisation. This includes constructing homogenised thermo-hydraulic computational models and employing convolutional neural networks to reduce computational time.