桥梁结构真实模态的智能化识别

Intelligent identification of the real mode for bridge structures

随着桥梁结构健康监测系统的不断普及,模态参数识别已被广泛运用于桥梁健康监测中以获取结构自身的动力特性值.虽然随机子空间算法作为模态参数识别算法中最为广泛使用的算法之一,但其依然存在稳定图中真实模态筛选难这一难题.基于此,在已有的研究基础之上进行了深入的研究.首先,提出桥梁结构真实模态存在的一般规律,并利用试验对其进行验证;其次,基于稳定图的基本原理提出以频率值、阻尼比以及振型系数为筛选因子的真实模态智能客观筛选算法.最后,以某大型斜拉桥为研究对象进行模态参数识别,并依次将算法所得频率值、阻尼比以及振型系数与动载试验结果和MIDAS理论结果分别进行对比分析,其结果表明:所提算法能运用于实际桥梁结构的真实模态识别,且识别结果具有可靠性.

With the continuous popularization of the health detection system for bridge structures,modal parameter identifica-tion has been widely used in bridge health monitoring to obtain the dynamic characteristics of the structure. Although the sto-chastic subspace algorithm is the most widely used one in the modal parameter identification algorithm, problem still remains that it is difficult to pick out the real mode in the stability diagram. Based on this, a further study is carried out based on the re-search of the present scholars. Firstly,the general rules of the real mode for bridge structures are proposed and verified by ex-periments. Secondly, based on the basic principle of stability diagram, the real mode screening algorithm is proposed, which takes the frequency value as the main screening basis, the damping ratio and the vibration mode coefficient as the verification basis respectively. Finally, taking a large cable-stayed bridge as the object of study, we carried out the modal parameters iden-tification and made a comparison of the frequency, damping ratio and vibration coefficient between the proposed algorithm and the dynamic load test results, and also the MIDAS theoretical results respectively. The results show that the algorithm is relia-ble and thus can be used in the real modal identification for bridge structures.