基于宏观调控策略的结构损伤检测PSO改进算法

结构损伤检测是结构健康监测过程重要的一步,数学上常常转化为求解约束优化问题。针对粒子群优化(PSO)算法易于出现的"早熟问题",采用市场经济条件下的宏观调控策略对早熟前粒子群位置进行干涉,藉以增强PSO算法抵抗局部极小的能力,达到改进PSO算法的目的。四个基准测试函数极值问题分析结果验证了改进后的PSO算法优于带权重因子的PSO算法,两层刚架单损伤和多损伤数值仿真以及三层建筑框架结构四种损伤工况试验研究进一步证明了改进后的PSO算法在结构损伤检测领域的应用是有效可行的。

An improved PSO algorithm for structural damage detection based on macro-economic control strategy

Structural damage detection is a very important process for structural health monitoring.It is often converted into a constrained optimization problem in mathematics.In this paper,an improved Particle Swarm Optimization(IPSO) algorithm is proposed for structural damage detection based on macro-economic control strategy,which aims to increase its convergence rate and thereby to obtain an acceptable solution of the premature convergence problem that is easy to occur in the later phase of the general PSO algorithm.The IPSO algorithm was empirically studied with a suite of four well-known benchmark functions,and further examined with both numerical simulations about single and multi-damage of a 2-story rigid frame and some experiments on a 3-story building model.The illustrated results show that the improved PSO algorithm is effective and applicable to the structural damage detection.