Optimal Galerkin Finite Element methods for non-isothermal Liquid Composite Moulding process simulations

Liquid Composite Moulding (LCM) describes a family of composites manufacturing processes in which a liquid thermosetting resin is injected, under pressure, into a reinforcing fibre bed. In these processes the mould containing the fibres is often preheated in order to facilitate the resin flow. Accurately modelling the thermal effects is crucial for designing an efficient manufacturing cycle. In this study, a selection of Galerkin Finite Element based transport algorithms, of both the non-monotone and monotone type, are used to simulate the convection-dominated resin flow through the porous medium. Their performances are verified against experimental test cases and their optimality is compared in terms of the prediction accuracy, computational efficiency and ease of implementation. Analytical solutions are derived for steady-state situations to validate the general numerical approach. An extension of the Mizukami-Hughes Petrov-Galerkin scheme is developed to deal with the unsteady conditions, and is found to perform very well.