ABSTRACT

This paper presents the principle and demonstrates the advantages of using a Phase Velocity Processing technique (PVP) applied to the full wave acoustic data recorded by a logging-while-drilling tool. Phase Velocity Processing is particularly well suited for handling the characteristics of LWD data that may degrade the results of Semblance method. Among them are: large tool-mode content, high signal attenuation across the receiver array, and strong road noise. Another advantage of applying PVP is its high vertical resolution: due to the common receiver mode of operation vertical resolution of the calculated slowness is equal to the offset between adjacent receivers. By comparison, the Semblance method delivers results averaged across the entire receiver array span. The PVP algorithm employs series of Fast Fourier Transform procedures to (a) filter the data using Heisenberg's method, (b) define time domain distribution of the phase arrivals using a Hilbert transform, and finally (c) perform the Phase Velocity analysis. Large number of phase arrival points provides efficient mechanism to calculate standard deviation of the results, quality indicators are added to show how well the algorithm was able to calculate both compressional and shear slownesses. Processing package includes filters operating in the frequency, time, slowness or space domains to minimize competing acoustic modes, especially the tool arrivals that are frequently present in LWD waveform data. Examples demonstrate improvement of the slowness calculation performed with the PVP over the results delivered by the Semblance method. The results obtained in attenuating formations, across the bed boundaries, and in thin beds are presented. The ability of the new filtering technique to identify and suppress the tool mode arrivals is also demonstrated.

This content is only available via PDF.
You can access this article if you purchase or spend a download.