A computational approach for crack identification in plate structures using XFEM, XIGA, PSO and Jaya algorithm

In this paper, a creative and intelligent approach based on an inverse problem that accurately predicts crack location in plate structures is presented. The eXtended Finite Element (XFEM) and the eXtended IsoGeometric Analysis (XIGA) are combined with two optimization techniques, namely Particle Swarm Optimization (PSO) and Jaya algorithm to predict the crack location. The superiority of XIGA is demonstrated by using various NURBS orders to reduce the number of elements, provide fast simulation and achieve best convergence compared with XFEM. Four numerical-optimization techniques are considered in this paper, namely XFEM-Jaya, XIGA-Jaya, XFEM-PSO and XIGA-PSO. In the optimization techniques, the objective function minimizes the difference between the calculated and measured displacements and strains. Convergence studies for various positions of a crack and a hole in plates are performed and the results show that Jaya algorithm significantly performs more accurate and faster than PSO. In addition, the proposed techniques are validated using experimental data and another numerical-optimization technique, i.e. XFEM coupled with Genetic Algorithm (GA), presented in literature. The comparisons show that XIGA-Jaya performs the best of all considered techniques.