This project explored how artificial intelligence can accelerate the design and optimisation of complex cooling systems by integrating data-driven modelling with advanced computational simulations. The research focused on jet impingement cooling, a highly efficient technique used in electronics, energy systems, and industrial processes where precise thermal management is essential. The project combined high-fidelity computational fluid dynamics (CFD) with machine learning models, including artificial neural networks (ANNs), to predict thermal and flow behaviour under varying design conditions. A comprehensive dataset was generated using CFD simulations, covering a wide range of geometric and operational parameters such as nozzle diameter, jet velocity, and jet-to-surface distance. The trained ANN model achieved remarkable accuracy in predicting the Nusselt number (heat transfer rate) and flow characteristics, significantly reducing computational time compared with traditional CFD-only approaches. A key outcome was the ability of the AI model to identify optimal configurations that balance cooling performance with energy efficiency. The research demonstrated that AI-enabled optimisation can reduce the number of design iterations required to achieve performance targets, accelerating innovation in thermal system development. The findings have strong implications for AI-enabled simulations and digital twin technologies, where real-time predictive tools are used to guide design and control decisions. Such methods can transform sectors that rely on high-precision cooling, including semiconductor manufacturing, aerospace, renewable energy, and next-generation reactor systems. By bridging physics-based and data-driven approaches, the project highlights how AI can enhance both the speed and accuracy of simulation-led engineering design.
This research represents a step towards intelligent design platforms capable of autonomously improving thermal systems, contributing to the broader goal of efficient, adaptive, and sustainable engineering solutions.