How to improve resolution and reduce multi-solution are always the key problems in seismic inversion. There are many reasons resulting in these problems, such as band-limited characteristic of seismic data and noise. In this paper, we proposed the fixed-point seismic inversion method to relieve the multi-solution problem and improve the resolution. Concretely, a point that returns to the same value after a finite iteration of a function is called a periodic point, and a fixed point is a periodic point whose period is equal to one. Thus, the parameters of subsurface can be recovered more accurately. Based on this theory, we can find that in case of seismic data has no noise or a little noise, the exact inversion results can be obtained, even with moderated noise in seismic data, the optimal inversion results can also be obtained by the proposed method. Finally, the synthetic models and seismic data are processed by the proposed method. From the synthetic model and field data examples we can see that after a finite iteration, the P-wave impedance information can be recovered well. Besides, comparing with conventional seismic inversion method, we can see that the resolution and fidelity of layer boundary information are higher than that of conventional seismic inversion method. And the stability and practicability of the proposed method are verified by synthetic model and seismic data tests.
Presentation Date: Tuesday, October 13, 2020
Session Start Time: 8:30 AM
Presentation Time: 10:10 AM
Location: 362D
Presentation Type: Oral
