Description
The use of fibre reinforced polymer (FRP) in civil construction applications has gained considerable popularity worldwide as suitable method for strengthening existing concrete structures. However, there is very little experience in the implementation of methods able to give a reliable prediction about the health of this type of structures even although sudden and brittle failure modes are likely to happen. Electromechanical impedance (EMI) method formulated from measurements obtained from PZT patches gives the ability for monitoring the performance and changes experienced by these strengthened beams at a local level, which is a key aspect considering their possible premature debonding failure modes.
In this work, a deep learning approach using convolutional variational auto-encoders for exploiting the raw impedance signatures is implemented to automatically detect anomalies in an unsupervised manner for this type of structures. To validate the effectiveness of the method, an experimental test campaign was performed. A concrete specimen strengthened with FRP and instrumented with PZT transducers in different location was subjected to different loading stages which provided different levels of damage. The results showed the potential of the method for EMI data-driven minor damage identification for real-life concrete infrastructures.
