The University of Leeds and Morvern Group Limited KTP 24_25 R4

Aeroplane turbine blades are being made faster and more reliably by combining computer simulations with AI-powered cameras that watch the manufacturing process in real time. The problem is that ceramic injection moulding—used to create the complex internal cooling channels inside turbine blades—is slow and prone to defects. Each blade must withstand extreme heat and stress, so even tiny flaws in the ceramic core can scrap the entire part. Currently, tool design and quality checks rely heavily on trial and error, which drives up costs and limits production rates. This project uses computational fluid dynamics to model how molten ceramic flows into the mould, then feeds that data into an AI system that adjusts the injection process on the fly. A machine vision camera monitors each part as it forms, flagging defects before the ceramic hardens. The goal is to cut waste, speed up production, and reduce the need for manual inspection. If successful, the approach could lower the cost of turbine blades for aircraft engines and industrial gas turbines—machines that generate electricity or power compressors. That means cheaper flights, more efficient power generation, and a more resilient supply chain for critical aerospace components.