Continuous Wavelet Transform Analysis for the Enhancement of Signal-to-noise Ratio in Coarse Grain Austenitic Stainless Steel

The time frequency-based techniques, short-time fourier transform (SIFT) and continuous wavelet transform (CWT), have been used for understanding the frequency dependent scattering and enhancement of signal-to-noise ratio (SNR) in high scattering austenitic stainless steel with grain size in the range of 30 to 210 mu m. SIFT and CWT analysis indicated that the higher frequency content of back-wall echoes (BWEs) decreases rapidly with the propagation distance; however, those of backscatter decrease slowly. Furthermore, the frequency of the BWEs decreases dramatically with an increase in grain size, and the higher frequency content of the back-wall is always lower than the backscatter; this shows good promise for enhancement of SNR by time frequency analysis. A CWT-based methodology that was found to correlate SNR improvement with wavelet scales is proposed for processing of ultrasonic signals in austenitic stainless steel specimens with different grain sizes. The pseudo frequencies have been calculated for the wavelet scales. A comparative study has also been made to correlate peak frequencies at BWE locations obtained using SIFT with pseudo frequencies calculated by the proposed methodology using CWT. The improvement in SNR with wavelet scales approach has been used for the detection of discontinuities whose signals are merged in high backscatter noise in the coarse grain austenitic stainless steel.