基于压缩感知在地震波作用下的结构损伤识别方法研究

Study of Structural Damage Identification Method based on Compressed Sensing under Seismic Wave Action

面对结构健康监测过程中的海量数据,该文将压缩感知的方法应用于数据处理。先将处理的数据用于损伤识别,后与原始的数据进行损伤识别对比,初步验证压缩感知的方法是否能应用于工程结构损伤识别中。为了能够更好地评价结构在外力作用下的健康情况,该文以地震波为例,通过MATLAB建立了一个五层刚架算例进行了数值模拟,通过模拟大量的响应数据,并运用压缩感知的方法对原始数据进行处理,结合损伤识别方法和数值算例加以验证,结果表明压缩的数据在一定范围内不影响损伤识别的进行。

In the face of massive data in the structural health monitoring process,this paper applies the compressed sensing method to data processing.The processed data is first used for damage identification,and then compared with the original data for damage identification,so as to preliminarily verify whether it can be applied to the structural damage identification in engineering.In order to better evaluate the health of the structure under the action of external force,this paper adopted seismic waves as an external force,established a five-layer rigid frame through MATLAB for numerical simulation,simulated a wealth of responding data,processed the original data with the compressed sensing method,and verified it with the damage identification method and a numerical example.The results showed that the compressed data within a certain range does not affect the damage identification.