Application of Elastic Full-Waveform Inversion on Blackfoot Data
MetadataShow full item record
Estimating a high resolution image of the subsurface has been always a challenge in the oilfield exploration. With advances in computational power, Full Waveform Inversion (FWI) has been proven as an efficient imaging tool of the subsurface since the first paper of Tarantola A. (1984). FWI consist basi- cally of two major steps: Forward modeling and backward propagation of data residuals. Modelling of the seismic wave-field requires a numerical solution of the partial differential equations like finite difference method, finite element method and spectral elment method. In this research, We implement finite difference method for modeling the seismic data. The used seismic model- ing methodology is described in details in chapter 2. Numerical optimization methods like quasi-Newton, Conjugate gradient and Steepest Descent have been used for fitting the synthetics with data to improve the accuracy of the solution (Chapter 3). We implemented the method using a real recorded seismic data from the Blackfoot oilfield in Alberta, Canada. The major challenge we faced in this research is to provide an accurate initial model to optimize better results. We applied FWI on the borehole seismic data and a new model was estimated from the borehole seismic data to be used as a starting model for inverting surface data. Chapter 4 shows, in details, the inversion of borehole seismic data while inversion of surface seismic data is described, in details, in chapter 5. This methodology helps to provide an accurate solution and better results. The resultant velocity models obtained from the application FWI on surface seismic data were used for Elastic Reverse Time Migration (ERTM) on the 2D seismic section and it was capable of developing a clearer image.
Master's thesis in Petroleum Geosciences Engineering