Reservoir Fluid Replacement Modeling of Acoustic Impedance Module by Machine Learning.

Abstract

This study focuses on the modeling of acoustic impedance in reservoir fluids during the process of fluid replacement. Acoustic impedance is a crucial parameter for characterizing subsurface formations and plays a vital role in various applications, including reservoir characterization and hydrocarbon exploration. Understanding the changes in acoustic impedance during fluid replacement can provide valuable insights into the fluid distribution and reservoir properties. Integration of machine learning (ML) techniques with reservoir fluid replacement modeling to enhance the prediction of acoustic impedance, is a critical step in reservoir characterization. In this study, fluid dynamics and reservoir behavior in conventional and non-conventional reservoir systems are meticulously analyzed using two models: a Conventional Reservoir (Model 1) and a Shally/Non-Conventional Reservoir (Model 2). Three different scenarios (Cases 1, 2, 3) involving brine, oil, and gas are considered. Notable variations in initial and final compressional and shear velocities and combined bulk modulus values are observed in both models. In each case, constancy in a specific fluid parameter is maintained, despite changes in other parameters, which implies the existence of an intrinsic feedback mechanism striving for equilibrium within the system. A reduction in both compressional and shear velocities, alongside an increase in the bulk modulus, is noticed, indicating a distinctive system response to fluid extraction. Changes in the bulk modulus suggest an adaptation of the reservoir system to uphold its structural integrity, despite the decrease in fluid volume.In the non-conventional model, a decline from initial to final values in compressional and shear moduli is witnessed, implying an increased potential for deformation and a decreased ability of the reservoir to withstand stresses. In contrast, an ascension in the combined bulk modulus values indicates the reservoir's resilience in withstanding external pressure forces.Additionally, the study reveals the differential effects of various fluids on acoustic impedance, with a reduction induced by the gas being particularly noteworthy. A complex interplay of fluid states leading to significant influence on hydrocarbon density is also unraveled, which offers key insights into the reservoir’s potential. The importance of understanding the intricate relationships between temperature, pressure, and entropy in defining fluid phases within reservoir systems is emphasized.Finally, the significant role of lithological factors, often surpassing the effects of fluids on reservoir behavior, is highlighted. The comprehensive understanding of reservoir dynamics brought forward by this study could serve as a guide for the development of more efficient extraction strategies and sustainable reservoir management practices.