Abstract:

Different methods of seismic inversion, including deterministic and stochastic approaches, have been developed in the oil & gas industry with the same aim of better characterizing petroleum reservoirs. An essential question involving this field is whether the different seismic inversion results from different algorithms can be compared with each other and moreover what would be a reliable method to do it quantitatively. Facies volumes resulted from the seismic inversion process are multi-dimensional problems that require the choice of multiple parameters to set the appropriate inversion. In addition, the multiple-parameters pertaining to different algorithms make the comparison between the results even more challenging.
This study proposes using a distance function to measure global dissimilarities between any two facies realizations and translate the dissimilarity matrix results into a plot using multi-dimensional scaling (MDS). The MDS plot will permit us to measure the dissimilarities between different methods encountered in the geophysical field of reservoir characterization. In this case, the output volumes are categorical variables of a sand and shale lithology definition. The distance function in this project was chosen to be L1-norm. The distance function is still subjective and there is room for more research in this field. Therefore, other types of distance functions might also be useful to show dissimilarities of facies realizations. However, this is not the scope of this work. The seismic inversion algorithms assessed in this study comprise the Simultaneous Model Based Inversion, Deterministic Sparse Spike Inversion, and Geostatistical Inversion. The first two approaches are using Bayesian Inference for facies prediction and the last one is a combination of Bayesian Inference and Markov Chain Monte Carlo algorithm.

Press the Back button in your browser.

Copyright 2016, Leticia Acquaviva Luebbert: Please note that the reports and theses are copyright to their original authors. Authors have given written permission for their work to be made available here. Readers who download reports from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the author, Leticia Acquaviva Luebbert.