基于自适应人工鱼群算法的钢梁结构损伤识别

Structural Damage Identification of Steel Beam Based on Adaptive Artificial Fish Swarm Algorithm

桥梁在投入使用后,在环境因素和荷载作用下,常常会产生结构上的损伤,伴随损伤状态运作的桥梁存在很大的安全隐患。本文介绍一种基于自适应步长的人工鱼群算法,首先介绍了人工鱼群算法的基本原理,并在此基础上对算法进行了改进,寻优过程中采用自适应步长。然后,通过基于MATLAB自行开发的有限元分析程序和锤击振动试验,得到各损伤工况下的理论和试验模态参数。最后将实际损伤和算法识别结果进行了对比分析。研究结果表明,本文的自适应人工鱼群算法能较准确找到损伤位置,并识别出损伤程度,与原算法相比在识别损伤位置和损伤的大小程度判断方面都有所提高,说明本文方法准确有效,可应用于实际结构的损伤识别。

When the bridge is in service,there will be damage somewhere under environmental influence and loadings. If the bridge is in service along with damage,it will be unsafe. In the paper,an artificial fish swarm algorithm with adaptive step is introduced. Firstly,the basis of artificial fish swarm algorithm is introduced,which is improved based on adaptive step. Secondly,analytical and experimental modal parameters under different damage cases are obtained by finite element analysis compiled on MATLAB and vibration testing with a hammer. Finally,the identified damages of different cases are compared with actual damages. It is found that the damage location and severity can be identified through the proposed adaptive artificial fish swam algorithm,which is more accurate and effective than the original algorithm,and can be applied in damage identifications of actual structure.