Incorporating 4D Seismic Steam Chamber Location Information into Assisted History Matching for A SAGD Simulation

Abstract

Modern assisted history matching tools allow an engineer to specify the uncertain parameters in a simulation dataset then perform an optimization to minimize the difference between observed field data and the corresponding simulation outputs. During this optimization, multiple simulations are run, with the uncertain parameters varying within the ranges specified by the engineer. The difference between the observed field data and the simulation output is measured by an objective function. Standard objective functions have been reported in the literature for the difference between observed and simulated production and injection rates, and for measurements in time and space done at observation wells. In this work we incorporate an additional objective function term that measures the difference between the observed and simulated steam chamber location and shape. In addition, the differences in 3D volumes were visualized, which lead to a better physical understanding of what parameters should be adjusted during a history match. For a steam injection process like Steam Assisted Gravity Drainage (SAGD), 4D seismic may be used to determine where a steam chamber is located in a reservoir at a point in time. The objective function developed in this work measures the difference between an observed chamber's shape and location in the reservoir and the corresponding shape and location for the chamber indicated in the simulation output. The objective function is a binary mismatch function, checking each simulation grid block to see if the seismic chamber and the simulation chamber agree or disagree, and calculating the ratio of the total volume of disagreements over the total volume of agreements. This steam chamber mismatch function was included in an assisted history match performed on a well pair from Suncor Energy's SAGD project at Firebag. The inclusion of this additional information added additional constraints to the simulation model, leading to a conceptually more dependable history match and a better geological and dynamic characterization of the reservoir.

History Matching of Electromagnetically Heated Reservoirs Incorporating Full-Wavefield Seismic and Electromagnetic Imaging

“…The parameters m, n, and k are empirically defined constants. For determining the empirical constants, we have used the parameters provided by Archie (1942) and Hill and Milburn (1956) for a sandstone rock formation. The conductivity for the injected water C w is given by the industry-proved IJWC equation (Report and Dresser 1982)…”

Integration of Artificial Intelligence and Production Data Analysis for Shale Heterogeneity Characterization in SAGD Reservoirs

“…In their previous study, Mirzabozorg et al (2012) argued that saturation and temperature data, in addition to production data, are needed for accurate future predictions. Hiebert et al (2013) incorporated the observed shape and location of steam chamber during history matching of a heterogeneous SAGD reservoir. A comprehensive history-matching routine consisting of geomodel construction and flow simulation for the Long Lake SAGD project were presented in and Feizabadi et al (2014).…”