Author/Authors :
Kaveh, A. School of Civil Engineering - Iran University of Science and Technology, Narmak, Tehran, Iran , Hoseini Vaez , S. R. Department of Civil Engineering - Faculty of Engineering - University of Qom, Qom , Hosseini, P. Faculty of Engineering - Mahallat Institute of Higher Education, Mahallat, Iran , Fathi, H. Department of Civil Engineering - Faculty of Engineering - University of Qom, Qom
Abstract :
A modified dolphin monitoring (MDM) is used to augment the efficiency of particle swarm optimization (PSO) and enhanced vibrating particle system (EVPS) for the numerical crack identification problems in plate structures. The extended finite element method (XFEM) is employed for modeling the fracture. The forward problem is untangled by some cycle loading phase via dynamic XFEM. Furthermore, the inverse problem is solved and compared via two PSO and EVPS algorithms. All the problems are also dissolved by means of fine and coarse meshing. The results illustrate that the function of XFEM-PSO-MDM and XFEM-EVPS-MDM is superior to XFEM-PSO and XFEM-EVPS methods. The algorithms coupled via MDM offer a higher convergence rate with more reliable results. The MDM is found to be a suitable tool which can promotes the ability of the algorithms in achieving the optimum solutions.
Keywords :
modified dolphin monitoring , plate structures , Cack detection , paticle swarm optimization , enhanced vibrating particle system
