基于学习因子优化的粒子群算法识别结构损伤

Structural Damage Identification Based on Improved Particle Swarm Optimization Algorithm with Learning Factors

采用粒子群算法识别结构损伤时,取得的实际最优解易被局部最优解覆盖。将局部最优解用于结构损伤识别,致使产生识别结果精度低、误判率高等问题。将S型动态变化的学习因子c_(1)和c_(2)引入到粒子群算法中,可使粒子快速、准确地收敛到全局最优解。该方法通过优化个体粒子和群体粒子的学习能力,有效提高了粒子群的寻优效率,实现了结构损伤的高精度识别。数值模拟结果表明:提出的新方法在结构损伤识别方面具有识别精度高、误判少、鲁棒性好等优点,可用于结构单一单元损伤或多单元损伤的识别。研究成果为结构损伤识别及结构健康监测提供了新思路。

During identifying structural damage with the particle swarm optimization algorithm,the actual optimal solution is easily covered by the local optimal solution obtained.The identification results exhibit low accuracy and high misjudgment rate when the local optimal solution is used to identify structural damage.By introducing learning factors c_(1) and c_(2) of S-type dynamic change,the animproved particle swarm algorithm is proposed to identify structural damage,which enables the particles to converge to the global optimal solution quickly and accurately.The optimization ability of the particle swarm of the presented method is effectively improved by improving the learning ability of individual particle and swarm particles,and the high-precision identification of structural damage is realized.The results indicate that the method exhibits the advantages of high identification accuracy,less misjudged damage and strong robustness in the process of structural damage identification.It can be applied to the identification of single damage and multiple damages of structure.The results can provide new ideas for structural damage identification and structural health monitoring.