A Method to Extract Hydrocarbon Reservoir Parameters from Wave Modes Excited by Acoustic Sources in Deviated Wells Surrounded by Anisotropic Formations, 14-R9778

Printer Friendly Version

Principal Investigators
Jorge O. Parra
Dawn C. Domaschk

Inclusive Dates:  01/01/08 – 06/30/09

Background - Because of an absence of large hydrocarbon reservoirs on shore and in shallow waters, the oil and gas industry is exploring deep ocean waters. The major basins contain thick shale formations, so understanding the physical properties of these rocks is critical at the borehole and seismic scales. Because most of the wells drilled from a platform are deviated, the full waveform monopole/dipole sonic logs are affected by the shale anisotropy and the angle of deviation of the well. This precludes estimating the vertical P-wave and S-wave velocities as a function of depth at the borehole scale to be tied with the surface seismic data. The shales are heterogeneous and anisotropic because of both the mineral distribution (intrinsic) and cracks and fractures. In addition, the characterizing shale as a source rock is a current topic of interest for the major service companies, universities and oil companies. The key is to know how to estimate the physical properties of this rock from seismic data for further interpretation. The anisotropic characteristics of shale or any type of rock are currently estimated from well log data acquired in vertical wells, so new processing techniques are needed to extract rock physical and fluid properties from deviated and horizontal wells. To evaluate and understand whether such objectives can be accomplished, the dispersion and attenuation spectra of the modes excited by a sonic tool in a deviated well must be understood. This can be achieved by modeling the full waveform sonic response of a point source (dipole/monopole) in a fluid-filled borehole, and then analyzing the dispersion and attenuation characteristics of the different modes that are excited by the source. The presence and characteristics of these modes depend on whether the formation is hard or soft and whether the induced anisotropy is caused by fractures, cracks, minerals, or stresses. SwRI researchers have developed the Transformed Boundary Integral Equation (TBIE) method for simulating acoustic waves propagating in deviated wells; this method overcomes the limitations of some existing approaches. This numerical method has showed that cross dipole data is very sensitive to borehole deviation. An objective of this work is to develop processing techniques to extract dispersion curves from cross-dipole well log data acquired in deviated wells. A second objective is develop and apply the TBIE modeling software to generate synthetic data to verify cross-dipole processing algorithms.

Approach - The approach was four-fold:

  1. Implement a processing algorithm to extract dispersion and attenuation spectral curves from full waveform monopole/dipole sonic logs.
  2. Implement a root finding method to extract the poles from the resonance equation for a point source in a deviated fluid-filled borehole surrounded by an anisotropic formation. Use the results to calculate phase velocity and attenuation of multiple wave modes excited by an acoustic source in a borehole.
  3. Simulate cross-dipole full waveform micro-seismograms for representative fracture-induced anisotropy and stress-induced anisotropy formations by including intrinsic attenuation. Use these cross-dipole synthetic data to test the proposed processing algorithm described in (1).
  4. Test the proposed methodology by computing theoretical dispersion/attenuation spectral curves with the method addressed in (2), using the same rock physical properties applied to generate the cross-dipole seismograms in (3). Compare these curves with those extracted from cross-dipole waveforms and apply the method to interpret real data.

Accomplishments - The processing algorithm and the modeling software were completed and applied to generate synthetic full waveform cross-dipole logs for testing an inversion algorithm that was developed during another project, 14-R9700. As a result, one paper was published in the February 2009 issue of World Oil, and a second paper was presented at the Society of Petrophysics and Well Log Analysts (SPWLA) meeting held in The Woodlands, Texas. This paper was published in the SPWLA Symposium Transactions. As a result of SwRI's promotional efforts, BP Oil Company has provided a data set from a North Sea reservoir. SwRI researchers are currently working with this data set that is allowing them to adapt the program to process data recorded in complex environments. Once this software is demonstrated with this data set SwRI will pursue funding opportunities with BP and other oil companies.

Figure 1. Spectra and semblance (dispersion) plots for a deviated well intercepting a shale formation at 60 degrees.


Figure 2. Spectra and semblance (dispersion) plots for a deviated well intercepting a shale formation at 60 degrees.

2009 Program Home