基于支持向量机和粒子群算法的结构损伤识别

为了有效地进行结构的损伤识别，提出了一种基于支持向量机和粒子群算法的结构损伤识别方法。首先利用支持向量机为损伤裂缝指标、损伤位置与各阶频率和一阶振型建立函数关系，然后将利用该函数关系得到的频率和振型与实测频率和振型间的差异作为优化目标，进而实现结构的损伤识别。为提高损伤识别的精度，将优化目标转化为多目标优化问题，并利用所提出的灰色粒子群算法进行求解。实验结果表明，该方法在结构损伤识别中具有较好的效果。

Structural damage identification by support vector machine and particle swarm algorithm

A structural damage identification method based on the support vector machine and the particle swarm algorithm was proposed. The functional expressions for the crack damage index, the damage location,the eigenvibration modes and first order eigenfrequency were derived by the support vector machine. The differences between the vibration frequency and modes derived from the obtained expressions and the measured ones were taken as the optimization objective to identify the structure damage. In order to improve the precision of the damage identification, the optimization objective was transformed into a multi-objective optimization problem which was solved by the proposed grey particle swarm algorithm. The proposed identification method was proved to be effective by experiments.