基于改进粒子群算法的损伤检测数值仿真研究

结构损伤检测在数学上常转化为约束优化问题．首先介绍了粒子群算法（PSO）的基本理论，并在分析传统粒子群算法容易陷入局部极小原理的基础上，提出了旨在增强粒子群算法后期粒子摆脱局部极小能力的改进粒子群算法（IPSO）．5个常用测试函数的测试结果表明，改进粒子群算法的性能优于传统粒子群算法．最后通过两层钢框架多种损伤工况的数值研究，进一步验证了改进粒子群算法的优越性及其应用于损伤检测领域的可行性．

Simulation Study of Damage Detection Based on Particle Swarm Optimization

Structural damage detection often can be inverted into a constrained optimization problem.Based on the analysis of the theory of the particle swarm optimization(PSO),an improved PSO is proposed,aiming to increase the diversity of the particles in the later phase of the PSO.The improved PSO is first studied by five benchmark functions;and the result shows that it is superior to the basic PSO.Then it is simulated by the steel frame structure with single and two damage detections;the results show that the improved PSO is a usable tool for structural damage detection.