Particle swarm optimization for numerical bifurcation analysis in computational inelasticity

Summary An efficient and robust algorithm to numerically detect material instability or bifurcation is of great importance to the understanding and simulation of material failure in computational and applied mechanics. In this work, an intelligence optimizer, termed the particle swarm optimization, is introduced to the numerical solution of material bifurcation problem consisting of finding the bifurcation time as well as the corresponding bifurcation directions. The detection of material bifurcation is approached as a constrained minimization problem where the determinant of the acoustic tensor is minimized. The performance of the particle swarm optimization method is tested through numerical bifurcation analysis on both small and finite deformation material models in computational inelasticity with increasing complexity. Compared with conventional numerical approaches to detect material bifurcation, the proposed method demonstrates superior performance in terms of both computational efficiency and robustness. Copyright © 2016 John Wiley & Sons, Ltd.