Petro-Elastic Parameters Effects on History Matching Procedures

Abstract

Rock physics models are the quantitative link between the seismic information and reservoir properties. Thus, several authors have been used this link to develop methodologies aiming to integrate seismic derived information and production history matching. Nevertheless, there are significant uncertainties related to petro-elastic parameters definition and its effects on typical history matching results. This paper evaluates this problem through the application of two history matching procedures. The empirical allowed range of petro-elastic parameters regarding fluid and rock behavior, such as temperature, salinity and porosity, mineral and dry rock modulus, respectively, are combined to define different data sets. Each set defines petro-elastic models to be coupled to a numerical reservoir model providing synthetic impedance distributions. These sets of impedance are then used in two different integrated history matching techniques. First, each set is combined with production data defining a global objective function to be minimized and provide a new horizontal permeability distribution resulting in an updated model. Secondly, a constrained inversion procedure is applied to convert impedance into saturation and pressure distributions. These maps indicate the optimization regions and the quantitative seismic information to improve the global objective function minimization accuracy aiming to derive a new permeability distribution. Thus, permeability and bottom-hole pressure from the updated reservoir models are compared to evaluate the several combinations of petro-elastic parameters effects. The major goals of this paper were to quantify the petro-elastic parameters definition effects on history matching procedures, indicating which of them is more sensitive to their variations. Furthermore, it was possible to assess their individual impact in the history matching results and highlight the importance of a careful definition of these parameters and its coherent integration with fluid flow models. This approach also contributes to quantitative integration aspects of oil reservoir development and management.