Control of flow in resin transfer molding with real‐time preform permeability estimation

Abstract Variabilities in the preform structure in situ in the mold are an acknowledged challenge to achieving reliable preform saturation in liquid molding processes. Physical models offer an effective means of deriving real‐time process control decisions so as to steer the resin flow in a desired manner, which ensures complete preform saturation. An important parameter influencing the fidelity of the simulations is the preform permeability, which is a strong function of the preform microstructure. A model‐based control strategy that incorporates the ability to determine and utilize local permeability information in real‐time is of much value, and forms the focus of the paper. An intelligent model‐based controller is developed that uses virtual sensing of permeability to derive optimal decisions on controlling the injection pressures at the mold inlet ports in a resin transfer molding (RTM) process. The controller employs an artificial neural network, trained using process simulation data, as an on‐line flow simulator, and a simulated annealing algorithm to optimize the injection pressures on‐the‐fly during the process. Preform permeability is virtually sensed during the process, based on the flow front velocities and the local pressure gradient along the flow front, estimated using a fuzzy logic model. The controller, implemented on an RTM process, is shown to be able to accurately steer the flow fronts through various preform configurations.