ニューラルネットワークを用いた階層構造物の簡易異常診断

抄録

In this study, we proposed a structural health monitoring and diagnostic method for layered structures using the convolutional neural network (CNN). The method is based on the idea of having the CNN learn the features of the anomaly using a mode circle composed of the FRFs of the anomaly calculated by the mathematical model to diagnose anomalies in the actual layered structures. This method belongs to the primary diagnosis one, and its purpose is to identify the location of abnormality quickly after abnormality detection. In particular, the objective is to identify which locations and to what extent the initial abnormality in the layered structure occurs. An abnormality represents a decrease in the stiffness characteristics (spring constant) of the outer wall of a hierarchical structure when it deteriorates or is damaged. We considered a three-layered structure as a numerical example and the proposed method is applied to the three types of abnormal conditions to verify the validity of the proposed method. As a result of verification, the proposed method was clarified effective in identifying initial abnormality in the layered structure.