A Digital Twin Framework of In-Line Process Optimisation for Material Extrusion-Based Additive Manufacturing

In this paper, a two-layered digital twin framework for in-situ parameter optimisation in material extrusion-based additive manufacturing is presented. The digital twin framework leverages several advanced data-driven technologies, such as surrogate modelling and machine learning, to enhance process efficiency and product quality. A case study was conducted using a fused deposition modelling printer, where a multi-source sensor net, comprising a 2D camera, a 2.5D laser scanner, and a 3D surface reconstruction system, provided valuable data insights. The results showcased successful implementation and highlighted the potential of the digital twin model in additive manufacturing. The efficiency trade-off of the multi-source sensor net was discussed, proving that applying the framework will not affect the fast-printing system. Future work will focus on closed-loop control and generalised modelling for further optimising the digital twin framework and enhancing printing product quality.