Integrated strategy for porosity mapping using genetic inversion on heterogeneous reservoir

Abstract

Seismic reservoir characterization is the stat-of-art in using various source of data. Generally, seismic data, due to their low resolution are randomly used in the final steps of reservoir characterization. However, large coverage of 3D seismic data, compared to well data, make it possible to be applicable for distribution of characters through the whole reservoir. In this regard, seismic data should be inverted to illustrate desired characters through the media. Conventionally, seismic inversion is performed using well logs which have defects in its derivation steps, such as wavelet extraction and its propagation through media. The proposed strategy to resolve such deficiencies, is the genetic inversion. However, genetic inversion has its own deficiency in accuracy. In this study, we propose an integrated strategy for using various source of data in an iterative manner for resolving this obstacle. The proposed strategy, uses combined related attribute to evaluate initial acoustic impedance inverted model by genetic inversion. The model then would be updated to satisfy well data. The proposed strategy was applied on a heterogenous reservoir from south west of Iran. Three seismic attributes were integrated to produce a unique attribute for initial model evaluation. The final model then was evaluated by well data and compared with the conventional method of seismic inversion. Result of the proposed strategy in genetic inversion depicted improvement in final acoustic impedance model and porosity distribution model.