Measurement and analysis of wave propagation in water-filled steel pipeline using iterative quadratic maximum likelihood algorithm

Acoustic wave propagation (up to 50 kHz) within a water-filled steel pipeline is studied using laboratory experiments. The experiments were carried out in a 6 m length of cylindrical stainless steel pipeline using acoustic transducers to acquire signals from 100 locations uniformly spaced along the longitudinal axis of the pipe. By applying the iterative quadratic maximum likelihood algorithm (IQML) to the experimental results, parameters such as wave numbers, attenuations and mode amplitudes were accurately extracted for individual modes from the measurement data. We found that the IQML algorithm could extract these parameters more accurately in situations where the measurement data had low signal to noise ratio as compared to other algorithms such as Prony’s method. A very good match was obtained between the experimental results and predictions from an analytical waveguide model for the wave number dispersion curves, attenuations and acoustic power characteristics of the axisymmetric and non-axisymmetric modes. Additional physical explanations of the propagation phenomena in the pipeline waveguide were obtained using the experimental results and analytical model.Acoustic wave propagation (up to 50 kHz) within a water-filled steel pipeline is studied using laboratory experiments. The experiments were carried out in a 6 m length of cylindrical stainless steel pipeline using acoustic transducers to acquire signals from 100 locations uniformly spaced along the longitudinal axis of the pipe. By applying the iterative quadratic maximum likelihood algorithm (IQML) to the experimental results, parameters such as wave numbers, attenuations and mode amplitudes were accurately extracted for individual modes from the measurement data. We found that the IQML algorithm could extract these parameters more accurately in situations where the measurement data had low signal to noise ratio as compared to other algorithms such as Prony’s method. A very good match was obtained between the experimental results and predictions from an analytical waveguide model for the wave number dispersion curves, attenuations and acoustic power characteristics of the axisymmetric and non-axisymmetri...