基于ARMAX模型和稀疏正则化的结构损伤识别方法

Structural Damage Identification Based on ARMAXModel and Sparse Regularization

结构健康监测系统对于保障工程结构安全具有重要意义,而结构损伤识别方法是结构健康监测系统的关键组成部分。本文提出一种基于时间序列ARMAX模型和稀疏正则化的结构损伤识别方法。首先,建立与结构运动方程对应的ARMAX模型,并利用模型自回归系数提取结构的固有频率和振型;然后,将提取的结构模态参数作为损伤敏感特征,构建损伤识别求解方程;最后,结合结构损伤的稀疏特性,使用稀疏正则化算法对方程进行求解,由解向量中的非零元素可得结构损伤的位置和程度。进行了一个六层集中质量剪切结构试验,试验结果表明该方法可以准确识别出结构中损伤的位置和程度,与传统损伤识别方法相比,该方法有效提高了损伤识别的精度。

Structural health monitoring systems play an important role in the safety of civil engineering,and structural damage identification method is the key part of structural health monitoring system.A structural damage identification method based on ARMAX model and sparse regularization is proposed in this paper.First,an ARMAX model corresponding to the structural motion equation is established,and structural natural frequencies and mode shapes are extracted from the autoregressive coefficients of the model.Second,the structural modal parameters are employed as damage sensitive features,and an equation for damage identification is built.Last,the damage identification equation is solved by sparse regularization algorithm,and the nonzero entries in the solution indicate damage locations and severities.The effectiveness and accuracy of the proposed method are verified through the laboratory test of a six-story lumped-mass shear building structure.The locations and severities of damages can be identified accurately.Compared with the traditional damage identification method,the identification accuracy is significantly improved.