Signal quality enhancement using higher order wavelets for ultrasonic TOFD signals from austenitic stainless steel welds
•DWT thresholding used to de-noise TOFD signals from Austenitic Steel welds.•Higher order mother wavelets used for the de-noising.•A comparative study performed with wavelets of Daubechies and Symlet families.•Signals de-noised using ‘db36’ compared with the SNR values obtained for ‘db4’.•Higher ord...
Gespeichert in:
Veröffentlicht in: | Ultrasonics 2013-09, Vol.53 (7), p.1288-1292 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •DWT thresholding used to de-noise TOFD signals from Austenitic Steel welds.•Higher order mother wavelets used for the de-noising.•A comparative study performed with wavelets of Daubechies and Symlet families.•Signals de-noised using ‘db36’ compared with the SNR values obtained for ‘db4’.•Higher order wavelets yield greater SNR values compared to the lower order wavelets.
Time of flight diffraction (TOFD) technique is a well-developed ultrasonic non-destructive testing (NDT) method and has been applied successfully for accurate sizing of defects in metallic materials. This technique was developed in early 1970s as a means for accurate sizing and positioning of cracks in nuclear components became very popular in the late 1990s and is today being widely used in various industries for weld inspection. One of the main advantages of TOFD is that, apart from fast technique, it provides higher probability of detection for linear defects. Since TOFD is based on diffraction of sound waves from the extremities of the defect compared to reflection from planar faces as in pulse echo and phased array, the resultant signal would be quite weak and signal to noise ratio (SNR) low. In many cases the defect signal is submerged in this noise making it difficult for detection, positioning and sizing. Several signal processing methods such as digital filtering, Split Spectrum Processing (SSP), Hilbert Transform and Correlation techniques have been developed in order to suppress unwanted noise and enhance the quality of the defect signal which can thus be used for characterization of defects and the material. Wavelet Transform based thresholding techniques have been applied largely for de-noising of ultrasonic signals. However in this paper, higher order wavelets are used for analyzing the de-noising performance for TOFD signals obtained from Austenitic Stainless Steel welds. It is observed that higher order wavelets give greater SNR improvement compared to the lower order wavelets. |
---|---|
ISSN: | 0041-624X 1874-9968 |
DOI: | 10.1016/j.ultras.2013.03.013 |