结构损伤检测的人工鱼群算法

结构损伤检测实际上属于系统识别的问题，其最终目标是识别结构损伤前后物理参数的变化。可利用实测结构模态参数建立方程求解得到结构物理参数，该过程在数学上往往转化为求解约束优化问题。由此，尝试采用人工鱼群算法来求解这类大型土木工程约束优化问题，首先介绍了算法的参数定义、行为描述及算法流程，然后利用经典测试函数对算法计算性能进行测试，最后给出了结构损伤识别这类约束优化问题的目标函数并通过数值仿真验证了该算法的有效性和鲁棒性。考虑测量噪声影响并通过不同损伤工况的数值仿真，研究结果表明，人工鱼群算法能有效地检测出损伤单元所处位置和损伤程度，因而将其应用到结构损伤检测领域是可行的。

Artificial Fish Swarm Algorithm for Structural Damage Detection

Structural damage detection in fact belongs to the system identification problem; its ultimate goal is to identify the change in physical parameters before and after structural damages. Measured structural modal parameters can be used to obtain the structural physical parameters. The identification procedure is often transformed into solving constrained optimization problems in mathematics. This paper tries to use artificial fish swarm algorithm (AFSA) to solve the kind of large civil engineering constraint optimization problem. First, the AFSA is introduced, its parameters are defined, its behavior and procedures described. The classical test functions are then adopted to test the performance of AFSA algorithm. Finally the objective function of the constrained optimization problem on structural damage identification is given and the effectiveness and robustness of the proposed AFSA method is evaluated by using a number of numerical simulations with noise polluted data. The illustrated results show that the AFSA method can effectively locate damage elements and quantity degree of damages with better noise immunity. It is promising to be applied to the field of structural damage detection in situ.