Mechanically enhanced model for accurate braid angle prediction on variable-diameter mandrels

Braided composites are widely used because of their high structural integrity and ability to create complex shapes. The current limitations of the analytical and kinematic methods, such as the low generalizability and computational inefficiency, restrict their application to variable-diameter mandrels. To address these challenges, this study aimed to develop a mechanically enhanced model for accurately predicting braiding angles during the braiding process, particularly for varying mandrel diameters. The proposed model incorporates yarn-yarn, yarn-guide ring, and yarn-mandrel contact interlacements to solve the convergence issues using an improved Euler method. Experimental validation was conducted on a 72-spindle radial braiding machine using mandrels with axially symmetric variable diameters, thereby enabling the precise measurement of braided angles under preset conditions. The simulation and experimental results demonstrated that the proposed model can accurately predict braid angle changes. It outperformed the extended classical model, especially at smaller preset angles. A consistent trend in the braid angle variation was observed, with the results showing a close alignment with the actual data over the steady-state phase of the process. The results revealed the robust characteristics of the braiding process and the minimal effect of the preset angles on the rate of change. The novel contributions of this research include the resolution of the premature mandrel contact issues, enhancement of the braid-angle prediction accuracy, and broadening of the applicability of the model to complex forming processes. This work has provided a reliable and efficient method for reducing the time and cost of designing braided composites, thereby aiding industries such as aerospace and automotive manufacturing.