High-resolution inversion for dispersion characteristics of acoustic logging waveforms

Abstract

Abstract In acoustic logging, most published studies of extracting array data characteristics are difficult to avoid the interference from the false modes. The false modes contribute little to the acoustic waveforms but makes the true modes hard to identify. We introduce a threshold of the energy spectrum into the pole-calculation method to remove the false modes, thereby eliminating their interferences to the true modes. To avoid the separation of multiple crossings of pole distribution, we directly acquire the slowness dispersion of multiple modes by superposing the single-mode results calculated from the weighted spectral semblance method. We processed the slowness-frequency snapshot and design an extreme-point extraction formula to generate the multi-mode scatterplot, which provide a precise slowness dispersion result. In terms of four synthetic cases of acoustic logging, it has demonstrated that the proposed algorithm can provide a high-resolution slowness dispersion profile for multiple modes. Even for acoustic logging while drilling with strong tool wave components, the aliases and interferences of false modes do not exist. Finally, a test proves that the proposed algorithm also has a good anti-noise ability.