Optimum tooling design for resin transfer molding with virtual manufacturing and artificial intelligence

Resin transfer molding (RTM) is a promising fabrication method for low to medium volume, high-performance polymer composite structures. Yet there exist several technical issues which impede a wide application base. One of these issues is tooling design. In the RTM process, the arrangement of injection gates and vents of the mold has a significant impact on product quality and process efficiency. In this paper, a systematic approach for optimum design of RTM tooling is introduced. This approach is built upon an RTM virtual manufacturing (simulation) model coupled with a neural network-genetic algorithm optimization procedure. The simulation model is employed to predict resin flow patterns (i.e, potential quality problems) and processing efficiency (mold filling time). With the simulation results, a neural network is trained to create a rapid RTM process model. Genetic algorithms are applied to this rapid RTM process model to search for the optimum solution to RTM process design. This tooling design scheme enables the engineer to determine the optimum locations of injection gates and vents for the best processing performance, i.e. short filling time and high quality level (minimum defects). The approach is illustrated with an example. (C) 2001 Elsevier Science Ltd. All rights reserved.