Multi-scale analysis for azimuthal anisotropy of reservoir rocks from cross-dipole logs

We present an algorithm for the inversion of formation azimuthal anisotropy from cross-dipole logs employing a wavelet filter. The characteristics of principal flexural waves can be more clearly identified by wavelet transforms of four-component flexural waveforms and multi-scale analysis in the frequency–time domain. Wavelet coefficients of the desired wave components are then extracted by a scheme called time–frequency information mass extraction (TFIME). The flexural wave can be reconstructed to filter out major interferences. We also introduce a multi-objective function into the simulated annealing optimization to further reduce the ambiguity in determining formation anisotropy parameters. Field examples are presented to demonstrate the practical application of the method.