基于布谷鸟搜索算法的温度驱动损伤识别方法

Temperature-driven damage identification method based on cuckoo search

基于振动的损伤识别方法数据分析过程复杂,且易受环境不确定因素影响,为解决该问题,提出了一种基于布谷鸟搜索算法的温度驱动损伤识别方法。分析中将温度视为结构的可测激励,对基于温度诱导应变的结构损伤参数识别公式进行了推导,进而利用实测的温度诱导响应值与有限元模型计算的温度诱导响应值构造目标函数,基于改进的布谷鸟搜索算法对有限元模型中的损伤参数进行更新,从而实现损伤的准确判别。通过一根两端弹性约束的H型钢梁,以及一榀Benchmark钢框架模型的数值仿真分析,验证了所提方法的损伤识别效果。分析结果表明:基于布谷鸟搜索算法的温度驱动损伤识别方法可对不同程度、不同位置及不同数量的损伤进行有效识别,即使在较高的噪声干扰下,仍能取得较好的识别结果。相比于基于振动的有限元模型损伤参数识别方法,该方法无需施加外部激励,输入与输出数据(温度与温度响应)方便可测,损伤识别所需的数据量较少,受噪声干扰影响小,可对结构的损伤进行有效定位与量化。

In order to solve the problems of the commonly used vibration-based damage identification methods,such as complex data processing and susceptibility to environmental uncertainties,a temperature-driven damage identification based on the cuckoo search algorithm method was proposed.Regarding the temperature as the measurable excitation of the structure,the formula of damage identification based on temperature-induced strain was derived.Then the objective function was constructed using the measured temperature-induced response and the calculated response,and the damage parameters in the finite element model were updated based on the improved cuckoo search algorithm to achieve accurate damage identification.An H-shaped steel beam and a benchmark steel frame model were used for numerical analysis,and the effect of the method were demonstrated.The results show that the proposed method can identify the damage of different cases,including different degrees,different positions and different quantities.Even under the interference of noise,it can still achieve better recognition accuracy.Compared with the vibration-based finite element model damage identification method,the proposed method does not require the application of external excitation,and the input and output data(temperature and temperature response)are convenient and measurable.The amount of data required for damage identification is less,and it is less affected by noise interference,which can locate and quantify the structural damage effectively.