Active Materials & Manufacturing Computing & AI

Hybrid physics-based and data-driven models for the prediction of process-induced voids in advanced composites

Funder EPSRC (part of UKRI)
Recipient organisation University of Bristol
Funding Not disclosed
Period Jan 2025 — Jan 2029

In plain English

AI plain-English summary

Wind turbine blades and aircraft wings are riddled with tiny air pockets that weaken them, and this project will build a computer model to predict where those pockets form before the part is made. These voids—trapped air bubbles inside the composite material—are a major source of manufacturing waste. Currently, engineers rely on trial-and-error to adjust production parameters, scrapping defective parts and consuming extra energy in the process. The project combines physics-based equations with machine learning to model void formation during curing, something neither approach can do alone. The work is split into four phases, each building on the last, to produce a hybrid predictive tool. If successful, the model could let manufacturers optimise production in sectors such as aerospace, automotive, and renewable energy without running costly physical experiments. For wind energy, that means longer, more reliable turbine blades with fewer structural defects. For the composites industry broadly, it means less material waste and lower energy use—directly addressing the sustainability challenge of reducing carbon footprints during production. The research is applied, not fundamental: its value lies in a practical tool for factory floors.

View original technical description
The composites industry has seen remarkable growth in recent years, with the allure of strong, lightweight materials driving it. These materials support manufacturing in industries such as aerospace, automotive, and renewable energy. However, this surge in demand brings sustainability challenges, including the need to reduce waste and energy consumption during production. Addressing these challenges is critical as industries work to minimize their carbon footprints. One approach to waste reduction is through advanced modelling techniques that predict the behaviour of materials during production. Modelling reduces the need for trial-and-error in the manufacturing process, which reduces the material and energy waste. For example, in the wind energy sector, effective modelling can help optimize the quality of composite wind turbine blades, reducing defects such as voids, which are small air pockets that can compromise the material's structural integrity. This project will focus on understanding the formation of these voids and developing a model to predict this behaviour accurately. This project will be split into four distinct phases. Each phase will build upon the previous phase, with the final aim being to produce a predictive model that integrates both physics-based and machine learning (ML) approaches.

View the original record at the funder ↗

Researchers

Ryan Chung (Student)

Related Research

Grants with similar aims, by meaning.

Modelling the effect of voids in Composite Components
Development of Advanced Numerical Fluids models for complex flows found in composites
Composites: Made Faster - Rapid, physics-based simulation tools for composite manufacture
Investigation of fine-scale flows in composites processing
Robustness-performance optimisation for automated composites manufacture

Original classification

Studentship

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.