分段小波分析与改进粒子群算法的两阶段桁架结构损伤识别

Segmented wavelet analysis and improved particle swarm optimization for two⁃stage truss structure damage identification

为实现桁架结构损伤的高精度识别,提出了一种基于分段小波分析与改进粒子群算法的两阶段损伤识别方法。提出了一种新的结构损伤特征参量——模态振型位移平方差,进行分段小波分析识别桁架结构的损伤位置;通过改进的粒子群算法精确识别桁架结构杆件及其端部连接损伤位置及损伤程度。算例分析表明,提出的模态振型位移平方差损伤特征参量识别损伤效果优于模态振型位移和模态振型位移差,一定程度上可以克服小波系数的“边端效应”;提出的分段小波分析法可有效解决小损伤因受损伤程度大的节点或单元影响后不易识别的问题;提出的改进粒子群算法具有跳出局部最优解的能力,寻优效率较标准粒子群算法更高,对二维桁架结构损伤的识别误差小于0.1%。

In order to achieve high⁃precision identification of truss structure damage,a two⁃stage damage identification method based on segmented wavelet analysis and improved particle swarm optimization is proposed.A new characteristic parameter of structural damage,the square difference of modal displacement,is proposed,which is used to perform segmented wavelet analysis to identify the damage position of the truss structure.The improved particle swarm optimization is used to accurately identify the damage position and damage degree of the truss structure member and their end connection.The analysis of the calculation example shows that the proposed damage characteristic parameter of modal displacement square difference is better than modal displacement and modal displacement difference in identifying damage,and it can overcome the"edge effect"of wavelet coefficients to a certain extent;The proposed segmented wavelet analysis method can effectively solve the problem that small damage is difficult to identify due to the influence of nodes or elements with large damage;The improved particle swarm optimization proposed has the ability to jump out of the local optimal solution,the optimization efficiency is higher than that of the standard particle swarm optimization,and the damage identification error of the two⁃dimensional truss structure is less than 0.1%.