Spectral Analysis of Experimental Ultrasonic NDT Signals

Abstract

Non Destructive Testing (NDT) plays an important role in different fields. It allows materials control or structures to check their condition, as well as detect defects without damaging inspected parts. The practical difficulty of extracting information to characterize materials led to use different methods of signal processing. Real ultrasonic signals considered in this paper are collected from Steel plate with defect and Aluminum cubic sample without defects prepared in LEND laboratory of Jijel University in Algeria. Two experiments are used: pulse-echo by contact and immersion techniques wherein ultrasonic energy is transmitted by piezoelectric transducer. The reflected signals are received by the same transducer where the energy is converted into an electrical signal. Signals are treated by three techniques adapted for nonstationary signals, and based on energy distribution in time-frequency plan; namely: Continuous Wavelet Transform (CWT), Wigner Ville (WVD) and Choi-Williams (CWD) distributions. This energy distribution allows an easy flight of time’s measurement of different echoes, and thereafter velocity’s calculation and defect’s localization. Advantages and limits of each method are as follows: WVD achieves good resolution of interfaces; however, its capacity remains limited by the appearance of non-desirable terms which may limit results readings. CWD avoids interferences phenomenon. It allows exactly extraction and clearly representation of signal components in time/frequency. Application of CWT clearly shows that temporal resolution is improved by contrast frequency resolution is degraded for high frequency terms. Also the disadvantage of this method comes from the absence of criterion of mother wavelet choice. Comparative study shows that CWD makes it possible to have a velocity closer to that given in theory. While the CWT and WVD give more accurate results regarding the defect’s position. This justifies the use of these efficient methods for non-destructive ultrasonic testing for defect localization and material characterization (Velocity and Young's modulus).

Keywords

• Ultrasonic,NDT,Wigner-Ville,Choi-Williams,Wavelet

References

1. Boashash, B. (2003). Time-Frequency Signal Analysis and Processing – A Comprehensive Reference. Elsevier Science . Bouden, T., Nibouche, M., Djerfi , F., & Dib, S. (2012). Improving Wavelet Transform for the Impact-Echo Method of Non Destructive Testing. Lecture Notes in Electrical Engineering, vol. 141. Springer-Verlag , 241-217. Chen, C. H. (2007). Ultrasonic and Advanced Methods for Non-destructive Testing and Material Characterization. University of Massachusetts , 3-33. Choi, H., & Williams, W. J. (1989). Improved time-frequency representation of multicomponent signals using exponential kernels. IEE transactions on Acoustics, Speech and Signal Processing, vol. 37, no. 6 , 862-871. Dib, S., Djerfi, F., Merdjana, H., & Bouden, T. (2017). Time Frequency Techniques for Detection and Characterization of Experimental Ultrasonic Signals. ICATS. Annaba, Algeria. Ding, J. (2007). Time frequency analysis and wavelet transform class notes. Taipei, Taiwan: National Taiwan University. Djerfi, F. (2017). Techniques avancées de traitement du signal adaptées aux contrôle non destructifs par ultrasons. PhD Thesis . Jijel, Algeria. Fuente J.V., V. L. (2002). Time-frequency analysis of ultrasonics backscattering noise for nondestructive characterisation on cement pastes. 8th ECNDT. Barcelona. Gaohui, W., Jun, S., Dezhi, Z., Huasong, G., Wulan, S., & Yan, L. (2008). Ultrasonic NDE of thin Composite plate Based on an enhanced Wigner-Ville distribution. 17th World Conference on Non destructive Testing, (pp. 25-28). Shanghai, China. Khan, N. A. (2010). Cross-term Suppression in Wigner Distribution. PhD. Disertation . Islamabad, Patistan. Malik, M. A., & Saniie, J. (1996). Performance comparison of time-frequency distributions for ultrasonic non destructive testing. IEEE Ultrason Symposium , 701-704. Qian, S., & Chen , D. (1996). Joint Time-Frequency Analysis: Methods and Applications. New York: Prentice Hall. Zhao, Y., Atlas, L. E., & Marks, R. J. (1990). The use of cone-shape kernels for generalized time-frequency representations of non stationary signals. IEEE Transactions on Acoustics, Speech and Signal Processing , 1084-1091.