Integrated Structural Reconstruction and History Matching Using Ensemble Filter and Low-frequency Electromagnetic Data

In this work, we describe a new integrated structural reconstruction approach based on the level set technique and ensemble Kalman filter-based history matching1 using water saturation distribution from low-frequency electromagnetic techniques (Maxwell’s equations) 2 and initial stochastic realisations of permeability distribution. In this work, we discuss the use of the ensemble Kalman filter combined with the Level Set method3 to solve the severely ill-posed problem of parameter and shape reconstruction in history matching of 3D reservoirs. The developed algorithm utilizes Sequential Gaussian Simulation for the creation of the initial geostatistical guesses for permeability which will then be utilized in the ensemble Kalman filter for honoring dynamic production data. The production data here includes water and oil production rates and water saturation distribution from low frequency EM reconstruction. This integrated methodology allows to obtain a better match of the produtcion data honouring the water monitoring obtained from low frequency EM data and also considering the initial model uncertatinty based on geostatistical techniques. The methodology is applied to a 3D syntethic oil reservoir model, the simulator (Schlumberger’s ECLIPSE E100) has been employed for the forward modelling and our in house Fortran and matlab codes has been employed for the History Matching and Electromagnetic reconstruction.