Full Waveform Inversion Using an Energy-Based Objective Function with Efficient Calculation of the Gradient
Y. Choi and T. Alkhalifah
Event name: 79th EAGE Conference and Exhibition 2017
Session: FWI B
Publication date: 12 June 2017
Info: Extended abstract, PDF ( 1.73Mb )
Price: € 20
Full waveform inversion (FWI) using an energy-based objective function has the potential to provide long wavelength model information even without low frequency in the data. However, without the back-propagation method (adjoint-state method), its implementation is impractical for the model size of general seismic survey. We derive the gradient of the energy-based objective function using the back-propagation method to make its FWI feasible. We also raise the energy signal to the power of a small positive number to properly handle the energy signal imbalance as a function of offset. Examples demonstrate that the proposed FWI algorithm provides a convergent long wavelength structure model even without low-frequency information, which can be used as a good starting model for the subsequent conventional FWI.