Damage Detection in Structural Systems Using a Hybrid Method Integrating EMI with ANN

A hybrid method combining electro-mechanical impedance (EMI) technique and artificial neural network (ANN) is proposed to detect damages in structural systems. The structural members are treated as Timoshenko beams for flexural motion as well as the damages are modeled by changes in Young's modulus in the damaged area. For a structural member with surface-bonded PZT wafers, a coupled system is considered. Based on this model, EMI signatures extracted from the PZT wafers can be used to identify damages in a structural system. Then, some kinds of compressed EMI data are employed as ANN input variables instead of the raw EMI data. It is shown that the identification results by this method agree fairly well with the given conditions.