The results show that the CNN-based damage detection method using mode shapes as the inputs has a higher locating accuracy for all damage degrees, while the method using mode curvature differences as the inputs has a lower accuracy for the targets with a low damage degree however, with the increase of the target damage degree, it gradually achieves the same good locating accuracy as mode shapes. Finally, the features extracted from each convolutional layer of the CNN are checked to reveal some internal working mechanisms of the CNN and explain the specific meanings of some features. The mode shapes and mode curvature differences are taken as the inputs of the CNN training samples, respectively, and the damage locating accuracy of the CNN is investigated. The CNN training samples (including a large number of damage scenarios) are created by ABAQUS and PYTHON scripts. This paper aims to locate damaged rods in a three-dimensional (3D) steel truss and reveals some internal working mechanisms of the convolutional neural network (CNN), which is based on the first-order modal parameters and CNN. Furthermore, the application of these two damage detection methodologies in a real case study related to the long-term monitoring of a 16th Century adobe church allowed confirming the building safe condition during almost two years of monitoring period, as well as the absence of damage after a 5.2Mw earthquake. They also evidence the importance of monitoring several modes as their sensitivity to damage depends on damage location itself. The results of the laboratory tests on a real scale adobe wall positively indicate the capabilities of these two methodologies to accurately identify damage. Two damage detection methodologies are investigated: (i) Autoregressive Models to predict the structural dynamic response taking into account the environmental parameters as input and (ii) Principal Component Analysis to detect patterns and anomalies in this response without the need of information about environmental conditions. This paper explores the accuracy of vibration-based SHM for identifying the existence of damage in adobe constructions, a widespread structural system but on which limited experimental and numerical applications of the technique are available. Structural Health Monitoring (SHM) has demonstrated to be a fundamental tool for detecting damage in early stages in existent civil engineering structures.
0 kommentar(er)
