EXPERIMENTAL DATA-DRIVEN STRUCTURAL HEALTH MONITORING FOR COMPOSITE PLATES UNDER DYNAMIC LOADS

Ensuring the structural integrity of aerospace components is essential for mission success, operational safety, and minimizing maintenance costs and downtime. This work proposes a deep learning-based structural health monitoring system to detect changes in the dynamic properties of composite components. Vibration responses are captured via accelerometers under varying dynamic acoustic excitations. A database of experimentally acquired signals is used to train and validate the model. Results demonstrate the system’s capability to identify dynamic alterations, highlighting its potential for efficient and reliable aerospace monitoring