ABSTRACT

Cycling skipping and local minima are serious problems in full waveform inversion (FWI). If FWI misfit functions involve local minima and starting model is far away from the true model, it is difficult to achieve the global minimum with local gradient-based optimization methods. In this study, we propose a misfit function based on adaptive matching filter (AMF), which enables us to measure time varing phase differences between observations and predictions without using local windows. 1D and 2D numerical examples illustrate that the AMF misfit behaves as a smooth, quadratic function with a broad basin of attaction. 2D inversion examples show that the AMF misfit enables us to build good starting models for FWI if low frequency signals are absent in observations.

Presentation Date: Wednesday, October 19, 2016

Start Time: 8:50:00 AM

Location: 143/149

Presentation Type: ORAL

Keywords:
local minima, adaptive matching filter, prediction, least-square waveform misfit, application, true model, misfit, misfit function, waveform inversion, seg seg international exposition
Subjects:
Reservoir Characterization, Seismic processing and interpretation
This content is only available via PDF.
2016. Society of Exploration Geophysicists
You can access this article if you purchase or spend a download.