Analysis of Metal Forming using a Shifted ICCG Algorithm

摘要

In general, Newton-Raphson algorithm is employed to solve a set of equations obtained from the rigid plastic/rigid viscoplastic finite element analysis of metal forming. In this paper, the authors present the application of an algorithm - the shifted incomplete Cholesky decomposition of stiffness matrix is combined with the conjugate gradient method, termed shifted ICCG method in the finite element method (FEM) analysis of metal forming. The performance of this algorithm in terms of CPU time, frictional boundary conditions and memory are discussed. Numerical tests were used to demonstrate the efficiency of the shifted ICCG algorithm in the modelling of metal forming. Simulation results indicate that the CPU time is larger than that by Newton-Raphson algorithm, but there is a significant saving in memory, especially for a large and complicated industrial problem. A comparison of the CPU time used in the diagonal algorithm and shifted ICCG algorithm has also been conducted. Simulation results of the metal forming processes show that the shifted ICCG algorithm is stable and applicable.