Comparison of ABC, CPSO, DE and GA Algorithms in FRF Based Structural Damage Identification

Abstract

In this contribution, performances of well-known population based algorithms, the artificial bee colony (ABC), contemporary particle swarm optimization (CPSO), genetic algorithm (GA), and differential evolution (DE) are compared in a basic model for damage identification (DI). DI is modeled as an inverse problem with the objective function based on the difference of the frequency response functions (FRF) computed by the finite element model of the structure and the reference data measured from damaged structure. Damage parameters are determined solving the problem with the aforementioned algorithms. It was observed that DE is the best one of a given number of function evaluations and gives the most accurate results in spite of noise interference to the reference data. According to the relevant literature, this is the first study including a comparison of these algorithms in an FRF based DI study.