Wrinkling Control During Deep Draping of Dry Fiber Fabrics: An Effective Numerical Optimization Approach

ABSTRACT Draping of dry fiber fabrics is the first step of high‐pressure resin transfer molding (HP‐RTM) and key stage of quality control, which is widely used in aeronautics. To minimize wrinkling intensity on the surface of the deformed fabric, this study proposes a numerical optimization approach for key forming parameters, including blank holder weight, ply orientation, and tool desk height. The in‐plane shear stiffness is modeled using a hypoelastic equation derived from experimental results, while the out‐of‐plane bending behavior is captured using the membrane‐bending decoupling (MBD) method. The in‐plane shear and out‐of‐plane bending behavior of dry fiber fabrics are modeled using a user‐defined subroutine and validated through the draping of a T‐shaped structure. Furthermore, a promoted back‐propagation neural network (BPNN) coupled with a genetic algorithm (GA) is developed to optimize draping parameters for minimum wrinkling intensity. The continuous nonlinear function of population for GA is built effectively using BPNN with limited input data. The weights and thresholds of BPNN are improved by an optimized method named whale optimization algorithm (WOA), resulting in predicted values with near‐zero absolute errors. The optimal forming parameters for the T‐shaped thin‐walled structure are determined through a numerical optimization approach and subsequently validated through experimental testing.