Particle Swarm vs.Evolutionary Optimization Techniques in a Multiobjective Framework for Damage Identification
In the context of real-world damage detection problems,the fack of a clear objective function advises to perform simultaneous optimizations of several objectives with the purpose of improving the performance of the procedure.Evolutionary algorithms have been considered to be particularly appropriate to these kinds of problems.However,evolutionary techniques require a relatively long time to obtain a Pareto front of high quality.Particle swarm optimization (PSO) is one of the newest techniques within the family of optimization algorithms.The PSO algorithm relies only on two simple PSO self-updating equations whose purpose is to try to emulate the best global individual found,as well as the best solutions found by each individual particle.Since an individual obtains useful information only from the local and global optimal individuals,it converges to the best solution quickly.PSO has become very popular because of its simplicity and convergence speed.However,there are many associated problems that require further study for extending PSO in solving multi-objective problems.The goal of this paper is to present the first application of PSO to multiobjective damage identification problems and investigate the applicability of several variations of the basic PSO technique.The potential of combining evolutionary computation and PSO concepts for damage identification problems is explored in this work by using a multiobjective evolutionary particle swarm optimization algorithm.
Damage identification multiobjective particle swarm optimization evolutionary algorithms
Ricardo Perera Sheng-En Fang Antonio Ruiz
Department of Structural Mechanics,Technical University,c/ Jose Gutierrez Abascal 2,28006 Madrid,Spa Department of Applied Mathematics,Technical University,c/ Alenza 4,28003 Madrid,Spain
国际会议
The 8th International Conference on Damage Assessment of Structures(DAMAS 2009)(第八届结构损伤评价国际学术会议)
北京
英文
661-668
2009-08-03（万方平台首次上网日期，不代表论文的发表时间）