A Data-Driven Approach to Extract Shear And Compressional Slowness From Dispersive Waveform Data

摘要: A data-driven approached is developed to extract shear and compressional slowness from dispersive sonic logging waveforms, or waveforms in which the formation arrival interfered by tool of similar but different slowness. It relies on extracting dispersion curve, slownessfrequency coherence map, for a range frequencies slownesses. The curve then converted into an objective function such that true associated with anomaly function. Specifically, can be probability density histrogram slownesses extracted curve. coincides one local maxima histogram, automatically determined optimization algorithm. Alternatively, also derived summing map across selected frequency range. finding maximum minimum nth derivative function, depending upon character data. In either case, plotted as color depths. This better QC than widely used semblancebased it reflects populations waveform. method has been extensively tested validated using synthetic data, wireline LWD data 20 wells six geographically-diverse regions.