基于改进布谷鸟搜索的Benchmark框架损伤识别

Damage identification of the benchmark frame based on an improved cuckoo search

工程优化问题中,布谷鸟搜索存在收敛精度不高、收敛速度较慢等弊端,从发现概率和随机步长两方面对布谷鸟搜索进行改进,并成功进行了实验室结构的损伤识别。对基本布谷鸟搜索进行了改进,发现概率自适应调整,步长自适应变化。基于MATLAB建立了英属哥伦比亚大学实验室ASCE Benchmark框架的3维有限元模型,并提取了6种损伤工况的实测频率和振型。基于频率因子和振型因子建立了目标函数,分别采用基本布谷鸟搜索和改进布谷鸟搜索进行了损伤识别,结果表明,改进的布谷鸟搜索能够更好地识别结构的损伤位置和损伤程度。研究成果具有一定理论研究意义和较高的工程应用价值,可应用于实际工程的损伤识别和健康监测。

There are many disadvantages when cuckoo search （CS） is adopted in engineering optimization, mainly the low convergence accuracy and slow convergence. So, the CS was improved in the aspects of discovery probability and steplength, which was then used in damage identification. In the improved cuckoo search （ICS） the adaptive discovery probability and adaptive step length were adopted. A 3D finite element model of the ASCE Benchmark frame model in the University of British Columbia was setup with the help of the Matlab and the experimental frequencies and vibration modes in six damage cases were extracted from the laboratory data. In the end, a damage identification objective function was established making use of the frequency factor and mode shape factor, and the damage identification was carried out based on CS and ICS. It is shown that the damage location and damage degree of the structure can be identified with better accuracy based on ICS. The results are of great theoretical significance and engineering value, and it can be applied in the damage identification and health monitoring in practical engineering.