| Abstract | This work aims to identify the different microstructures presented by cast iron, namely, lamellar, vermicular and spheroidal microstructures, through the statistical fluctuation and fractal analyses of backscattered ultrasonic signals. The signals were obtained with a broadband direct contact ultrasonic probe with a central frequency of 5 MHz. The statistical fluctuations of the ultrasonic signals were analyzed using Hurst and detrended-fluctuation analyses (DFA), and the fractal analyses were carried out by applying the minimal cover and box-counting techniques to the signals. The curves obtained for the statistical fluctuations and fractal analyses, as function of time window, were processed by using two pattern classification techniques, namely, principal-component analysis (PCA) and Karhunen-Loève expansion. For the Karhunen-Loève expansion, an approximately 100% success rate has been reached for the classification of the different microstructures, for the training and the testing sets of events. The results presented correspond to an average taken over a 100 randomly chosen sets of events. It is concluded that the statistical fluctuation and fractal analyses are effective additional tools for recognition of the different cast iron microstructures. © 2011 American Institute of Physics. |
|---|
