In the world of engineering, every advance represents a step forward towards excellence and efficiency. In this regard, the consortium led by CiTD has achieved significant milestones in the SM@RTM project, aimed at mitigating defects in composite parts using the resin transfer moulding (RTM) technique.

One of the project’s most notable branches, developed by CiTD in collaboration with IMDEA, focuses on optimising the RTM process through advanced computational models. Computational Fluid Dynamics (CFD) techniques have been applied to model mould filling, taking into account different mould defect scenarios, such as gaps or overlaps, inherent in manufacturing using Automated Fibre Placement (AFP). These models, supported by accurate data obtained from the collaboration between CiTD and AIMEN, also within the scope of the project, allow for a detailed simulation of the filling process, providing a deep understanding of the factors that influence the final quality of the part.

Furthermore, an innovative leap forward has been made by developing surrogate models using neural network techniques in Python. These models significantly reduce the computational time required to simulate different defect scenarios, allowing for rapid evaluation of the influence of injection parameters and mould defects on the final porosity of the part. This approach, based on data obtained from CFD models, demonstrates an effective synergy between numerical simulation and machine learning.

In the final stage of this simulation process, from complex CFD simulation models to surrogate models, including the learning of these models, CiTD has played a crucial role in the analysis and interpretation of the results obtained. Through exhaustive studies, it has been identified how different types of mould defects affect the injection pressures required during the RTM process. It has been observed that gap-type defects require lower pressures compared to overlap-type defects. Based on these findings, optimal pressures have been determined for each error distribution, thus achieving more precise and efficient control of the moulding process. In short, real-time corrective actions have been developed, a powerful tool for operators, and RTM control strategies have been established.

In summary, the advances achieved in the SM@RTM project represent a significant milestone in the pursuit of excellence in composite parts manufacturing. The combination of advanced simulation models and innovative machine learning techniques paves the way for a more efficient, reliable and adaptive RTM process.