In this paper an approach to evaluate the corrosion level of non accessible pipes is presented. The method is based on the classification problem of the ultrasonic waves reflected by four classes of defects whose geometry is defined by well-known axial, angular and radial extensions. A set of optimal features constitutes the database for the final classifier. These features are chosen by time and frequency parameters extracted from simulated ultrasonic waves. Three types of classifier have been tested: the k-nn classifier, the Linear Vector Quantization and the Multi Layer Perceptron neural networks. The results show that the method achieves a good recognition rate when the classification is performed by the Multi Layer Perceptron neural network and that the various classes recognized could be used as reference class to evaluate the state of the pipe under test.

Neural network classification of flaws in pipes using ultrasonic waveforms / Acciani, G.; Brunetti, G.; Fornarelli, G.; Guaragnella, C.. - In: WSEAS TRANSACTIONS ON CIRCUITS AND SYSTEMS. - ISSN 1109-2734. - 4:10(2005), pp. 1387-1394.

Neural network classification of flaws in pipes using ultrasonic waveforms

Acciani, G.;Fornarelli, G.;Guaragnella, C.
2005-01-01

Abstract

In this paper an approach to evaluate the corrosion level of non accessible pipes is presented. The method is based on the classification problem of the ultrasonic waves reflected by four classes of defects whose geometry is defined by well-known axial, angular and radial extensions. A set of optimal features constitutes the database for the final classifier. These features are chosen by time and frequency parameters extracted from simulated ultrasonic waves. Three types of classifier have been tested: the k-nn classifier, the Linear Vector Quantization and the Multi Layer Perceptron neural networks. The results show that the method achieves a good recognition rate when the classification is performed by the Multi Layer Perceptron neural network and that the various classes recognized could be used as reference class to evaluate the state of the pipe under test.
2005
Neural network classification of flaws in pipes using ultrasonic waveforms / Acciani, G.; Brunetti, G.; Fornarelli, G.; Guaragnella, C.. - In: WSEAS TRANSACTIONS ON CIRCUITS AND SYSTEMS. - ISSN 1109-2734. - 4:10(2005), pp. 1387-1394.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11589/6238
Citazioni
  • Scopus 5
  • ???jsp.display-item.citation.isi??? ND
social impact