Material characterization of functionally graded material by means of elastic waves and a progressive-learning neural network

In this paper, a procedure is suggested for characterizing the material properties of functionally graded material (FGM) plate by the use of a modified hybrid numerical method (HNM) and a neural network (NN). The modified HNM is used to calculate the displacement responses of FGM plate to an incident wave for a known material property. The NN model is trained by using the results from the modified HNM. Once trained by, the NN model can be used for on-line characterization of material properties if the dynamic displacement responses on the surface of the FGM plate can be obtained. The material property so characterized is then used in the modified HNM to calculate the displacement responses. The NN model would go through a progressive retraining process until the calculated displacement responses obtained by using the characterized result is sufficiently close to the actual responses. This procedure is examined for two sets of material properties of a SiC–C FGM plate. It is found that the present procedure is very robust for determining material property distributions in the thickness direction of FGM plates.