Author:

Abstract:

© Springer Nature Singapore Pte Ltd. 2019. The first two steps in Structural Health Monitoring (SHM) are to detect and localize damage in a structure. Transmissibility uses only output signals to extract the structural dynamic parameter and very sensitive to damage. In this paper, a simply supported beam is analyzed using a transmissibility-based damage detection methodology combined with Auto-Associative Neural Network (AANN). Firstly, the transmissibility is calculated between two points in the examined beam. Then a transmissibility indicator is taken as input for the neural network, which accounts only for the response data and is sensitive to damage. The target of AANN is the location and the severity of the damage. It is found that the network, after being trained, show good results and can be used to detect and localize damage.