The Simulation of Dual Layer Fuel Tank During the Impact With the Ground Based on Parallel Computing

作     者：Li Zheng Jin Xiang-long Chen Xiang-dong 

作者机构：Shanghai Jiao Tong Univ Ctr High Performance Comp Shanghai 200240 Peoples R China 

出 版 物：《JOURNAL OF COMPUTATIONAL AND NONLINEAR DYNAMICS》 (计算和非线性动力学杂志)

年 卷 期：2007年第2卷第4期

页      面：366-373页

核心收录：

学科分类：07[理学] 0802[工学-机械工程] 0701[理学-数学] 0801[工学-力学（可授工学、理学学位）] 070101[理学-基础数学] 

主　　题：dual layer fuel tank parallel algorithm ALE finite element domain decomposition 

摘      要：The crashworthiness of a dual layer fuel tank, with the outer layer made of metal and the inner layer made of woven fabric composite material, is fundamental for the survivability of air impact with the ground in emergency. In this research, the simulation of a three-dimensional dual layer fuel tank in the impact with the ground is achieved through the multimaterial arbitrary Lagrangian-Eulerian (ALE) finite element method because of its ability to control mesh geometry independently of geometry. At the same time, the naked flexible tank in the impact with the ground is simulated for the evaluation of the outer metal tank. The ALE description is adopted for the fluid domain, while for the structural domain the Lagrangian formulation is considered. The computation of the fluid-structure interaction and the impact contact between the tank and the ground are realized by the penalty-based coupling method. Then, the dynamic behaviors of the dual layer fuel tank and the naked flexible tank in the impact are analyzed. In the meantime, the parallelism of the dual layer fuel tank is discussed because the computation of the fluid-structure interaction and the impact contact is quite nine consuming. Based on domain decomposition, the recursive coordinate bisection (RCB) is improved according to the time-consuming characteristics of fluid-filled tank in the impact. The result indicates, comparing with the RCB algorithm, that the improved recursive coordinate bisection algorithm has improved the speedup and parallel efficiency