结构健康监测传感器优化布置的混合算法  

A Hybrid Algorithm for Optimal Sensor Placement of Structural Health Monitoring

作　　者：谢强[1] 薛松涛[1] 

机构地区：[1]同济大学土木工程学院,上海200092

出　　处：《同济大学学报（自然科学版）》2006年第6期726-731,共6页Journal of Tongji University:Natural Science

基　　金：国家自然科学基金资助项目(50508026)

摘　　要：提出了一种基于模型减缩和线性模型估计理论的、用于建筑结构健康监测中传感器布置的新算法.根据选定的主、从自由度,用改进减缩系统方法来减少初始结构的自由度数目.然后,基于线性模型估计,以所选定的目标模态为线性模型的设计矩阵,用奇异值分解处理设计矩阵.用分解后的前几个左奇异向量计算每一个自由度对于结构模态的贡献.最后用迭代算法来确定所需的传感器数量和位置.算例表明,此种混合算法适用于建筑结构监测的传感器布置计算.This paper presents a new hybrid algorithm based on both model reduction and linear model estimation for searching optimal number and locations of sensors for building structural health monitoring. According to the chosen master and slaver degrees of freedom（DOFs）, a model reduction method named Improved Reduced System is used to reduce the DOFs of the original model. Then, based on linear model estimation, the target modes are treated as a design matrix of linear model and singular value decomposition is used to decompose the target modes. The first several left singular vectors of the decomposed matrix are used to calculate contribution of each sensor candidate. Iterative method is used to choose the last set of sensor locations. Illustrations clearly show that the hybrid method is applicable to optimal sensor placement for building structural health monitoring.

关 键 词：结构健康监测 最优传感器布置 模型减缩 线性模型估计 混合算法 

分 类 号：TU973.2[建筑科学—结构工程]