Numerical permeability tensor characterization of fibrous reinforcement through 1D flow analysis using particle swarm optimization

Özet

Flow simulations are performed to analyze the resin flow behavior through fibrous reinforcements to promote void-free composite manufacturing with excellent mechanical properties. These flow simulations require an essential parameter known as permeability tensor which is defined as the resistance to the flow of resin due to fibrous reinforcement. This study proposes a key strategy to determine all in-plane permeability components from a single rectilinear flow experiment. The method is based on introducing intentional disturbance to the mold domain, which transforms the one-dimensional flow into a two-dimensional flow. The process is divided into two steps, the experimental determination of flow arrival times at designated locations within the mold domain and comparing it to the numerical flow arrival times (obtained using LIMS software) via the residual sum of squares (RSS). An optimization algorithm based on particle swarm optimization (PSO) is established to reduce the RSS to get accurate permeability predictions. The validation study of the proposed strategy has been practiced for three cases. The results show that this method can effectively characterize the in-plane permeability components from a single rectilinear injection experiment.