Signal Processing Techniques for Ultrasonic Waves to Measure Stresses in Oil Pipelines

Ensuring the structural integrity of oil pipelines is vital to prevent environmental damage and economic losses. In that sense, it is important to know the magnitude of the stress in the pipe, which must be done using non-destructive techniques. Measuring stress using ultrasonic longitudinal critically refracted waves (LCR) has been applied to pipelines with very promising results. The technique is based on the acoustoelastic theory that relates speed variation of elastic waves traveling in the material with its state of strain. Nevertheless, the signals acquired from piezoelectric transducers in such application may show high levels of noise coming mainly from material sources (grain boundaries, irregularities). The noise makes the measurement of wave velocity difficult, resulting in imprecise evaluations of the stress in the pipeline. The aim of this study is to evaluate techniques for filtering digital signals of LCR waves propagating in an oil pipe fabricated with API 5L X70 steel. We analyzed the signal-to-noise ratio (SNR) of digitalized acquired signals in four circumstances: without treatment; signals treated with successive averages; treated with FIR (Finite Impulse Response) and IIR (Infinite Impulse Response) digital filters, and with the Discrete Wavelet Transform (DWT). The results show that the signals treated with DWT present better SNR compared to the other techniques.Copyright © 2013 by ASME