Journal Article

Variable-Diameter Deployable Structure Composite Preforms Made by Braiding and Needle-Punching Integrated Forming Technology

Geng, Fengchen, Zhou, Zhengxi, Wang, Yaoyao, Yang, Tianzheng, Zhuang, Zhiqi, Shan, Zhongde, Sun, Zheng, Zhang, Jun, Guo, Zitong

2024

Overview

Variable-diameter deployable carbon fiber reinforced polymer (CFRP) composites possess deformation and load-bearing functions and are composed of stiff-flexible coupled preforms and matrix. The stiff-flexible coupled preform, serving as the reinforcing structure, directly determines the deployable properties, and its forming technology is currently a research challenge. This paper designs a braiding and needle-punching (BNP) composite preform forming technology suitable for stiff-flexible coupled preforms. Before forming, the preform is partitioned into flexible and rigid zones, with braiding and needle-punching performed layer by layer in the respective zones. A retractable rotating device is developed to form the stiff-flexible coupled preform, achieving a diameter variation rate of up to 26.6% for the BNP preform. A structural parameter model is also established to describe the geometric parameter changes in the deformation and load-bearing areas of the preform during deployment as a function of the braiding angle. Based on experiments, this paper explains the performance changes of BNP composites concerning the structural parameters of the preform. Experimental analysis shows that as the braiding angle increases, the tensile performance of BNP composites significantly decreases, with the change rate of tensile strength first decreasing and then increasing. Additionally, when the braiding angle is less than 21.89°, the impact toughness of BNP composites remains within the range of 83.66 ± 2 kJ/m 2 . However, when the braiding angle exceeds 21.89°, the impact toughness of BNP composites gradually decreases with increasing braiding angle. Furthermore, a hybrid agent model based on Latin hypercube sampling and error back-propagation neural network is developed to predict the tensile and impact properties of BNP composites with different structural parameters, with maximum test relative errors of 1.89% for tensile strength and 2.37% for impact toughness.

• Publisher: Springer Nature Singapore,Springer Nature B.V,SpringerOpen
• Subject: Angles (geometry) / Back propagation networks / Braided composites / Braiding / Braiding and needle-punching process / Carbon fiber reinforced plastics / Composites mechanics / Composites preform / Deployable structures / Electrical Machines and Networks / Electronics and Microelectronics / Engineering / Engineering Thermodynamics / Fiber composites / Fiber reinforced polymers / Heat and Mass Transfer / Hypercubes / Impact strength / Instrumentation / Latin hypercube sampling / Load bearing elements / Machines / Manufacturing / Mechanical Engineering / Neural networks / Original Article / Parameters / Parametric model / Power Electronics / Prediction model / Preforms / Processes / Punching / Smart Material / Tensile strength / Theoretical and Applied Mechanics / Toughness / Variable-diameter deployable structure