用粒子群算法选择体育馆监测系统中传感器的最优布点

Optimal placement of sensors for monitoring systems on a gymnasium using particle swarm optimization

如何合理、有效地确定传感器的位置,是结构健康监测系统正常、高效运行的前提.基于模态能量和粒子群算法进行传感器优化布置研究.首先,讨论了粒子群算法的基本原理和优点;其次,根据结构模态应变能的大小挑选出结构的主要贡献模态,进而确定监测模态阶数;然后,针对不同的传感器类型,基于位移模态定义适用于加速度传感器的适应度函数,基于应变模态定义适用于应变传感器的适应度函数;最后,以西宁市体育馆钢结构为工程背景,采用粒子群算法对其监测系统中加速度传感器和应变传感器分别进行优化布置.结果表明,该方法优化结果稳定可靠且收敛迅速.

The optimal sensor placement is a primary problem to be solved to ensure that the structural health monitoring system works normally and efficiently. A method of sensor placement based on the modal energy and particle swarm optimization is proposed. Firstly, the basic principle and the advantages of the particle swarm optimization(PSO) are discussed. Secondly, the modal shape order for optimal sensor placement is the order of significant modes selected by the modal strain energy. Thirdly, on the basis of displacement and strain modes, two expressions of the fitness function are defined for different sensors and tasks. Finally, taking the steel structure of Xining gymnasium as a case, the acceleration sensors and the strain sensors in modal parameter testing are optimized by the proposed method. The results show that the PSO method gives reliable and stable results with rapid convergence.