基于Kent混沌蜂群算法的码头排架结构损伤识别

摘要

为有效识别高桩码头的结构损伤,提出了一种基于Kent混沌人工蜂群算法的结构损伤识别方法.该算法采用Kent混沌映射和一般反向学习策略初始化蜂群,并引入锦标赛选择策略和Kent混沌搜索机制对算法性能进行改进.基于损伤结构模态参数(振型和固有频率)的计算值与测量值之差构造了目标函数,采用改进的混沌人工蜂群算法搜索最优目标函数对应的损伤因子,实现了基于改进算法的结构损伤识别.对高桩码头排架不同工况下的损伤识别结果表明,改进的混沌人工蜂群算法能够有效地识别结构损伤,且性能优于粒子群优化算法、基本人工蜂群算法和Logistic混沌人工蜂群算法.%A Kent chaos artificial bee colony (KABC) algorithm is proposed to effectively identify the damage of the highpiled wharf structure.The Kent chaotic mapping and generalized opposition-based learning strategy is utilized to initialize the artificial bee colony.Besides,the tournament selection strategy and Kent chaotic search are employed to improve the performance of the algorithm.The difference value between the calculated and measured modal parameters of damaged structure (inherent frequency and vibration mode shape) is constructed as objective function.The KABC algorithm is then occupied to search the damage factors that correspond to the optimal objective function value so that implement the damage identification.The simulation results of a high-piled bent under different operating conditions indicate that the proposed method show better performance than particle swarm optimization,basic ABC algorithm and Logistic chaos ABC algorithm and can effectively identify the structural damage.