Reservoir Connectivity and Compartmentalization Inference Using Drainage Volume and Dynamic Data

Abstract

Abstract In hydrocarbon reservoirs, reservoir heterogeneity and fluid production/injection result in unique reservoir energy signature (waves/pulses) and determine its shape and propagation. Reservoir engineers uses this propagation of the pressure waves or pules to determine many key reservoir properties (e.g., drainage volumes, reservoir energy, rock properties, decline analysis, etc.) to help in evaluating different field development strategies. The objective of this paper is to illustrate applications of Fast Marching Method (FMM) in assessing reservoir performance, identifying reservoir patterns and anomalies from production/injection data, and predicting the reservoir response when considering modeling uncertainty for model calibration. The proposed hybrid approach in this work is a physics-constrained data-driven approach. It uses the diffusive time-of-flight (DTOF), this represents the propagation time of pressure disturbance/wave from a source or a sink, from which the drainage volumes can be obtained as it is the case in traditional well testing. The DTOF is calculated from the 3D diffusivity equation after the transformation to a 1D equation. The high frequency diffusivity solution can be casted in the form of the Eikonal equation to allow for an analytical computation of the DTOF, which is solved via the FMM. Using the DTOF calculated production and injection rates will help us inferring faults existence and their transmissibility, fracture networks (existence, location, orientation and direction, faults’ transmissibility, fractures’ conductivity, and inter-well connectivity network.). The fundamental concept is to formulate a solution of the diffusivity equation that describes the transient flow. In this work, several synthetic models were used to benchmark. The work demonstrates how the DTOF was used to: generate pressure maps for reservoir monitoring, predicts the operational constraints (e.g., bottom-hole pressure) drainage volumes, and predict new wells’ performance. FMM results approximately matches in terms of well performance compared to simulation results; the DTOF gives a great insight about the pressure drop in the reservoir during the early- and mid-stages of the simulation. For a relatively short time intervals, FMM proved to be computationally efficient with a much shorter turnaround time to solve the problem, and closely matching the results obtained from numerical reservoir simulation. The physics-constrained data-driven using the DTOF was able to identify the pressure drop for the whole reservoir and to predict the bottom-hole pressure for the wells. Using the DTOF, it is possible to infer major geological features such as faults, fracture networks and regional heterogeneity. Fast Marching Method is an efficient method for solving the diffusivity equation for the DTOF to quickly give engineers an insight into the reservoir pressure (energy) and contacted reservoir volumes in order to maintain evergreen reservoir models.