| Popis: |
1. SUMMARY We describe a hybrid optimization method of inversion which combines a local method with a genetic algorithm, and apply it to determine the anisotropic parameters from multi-component seismic waveform data. In the inversion we use a fitness function which measures the agreement between the synthetic and observed data. The synthetic seismograms from a point source in horizontally stratified anisotropic media are calculated using the reflectivity technique. In the initial stage of this hybrid optimization method, the global space-sampling properties of the genetic algorithm are used to direct the search to the region close in the global solution. This solution i s further improved using a local search method. The numerical experiments presented in this paper illustrate the feasibility of the approach. |
