Improved Geosteering by Integrating in Real Time Images From Multiple Depths of Investigation and Inversion of Azimuthal Resistivity Signals

Abstract

Summary Optimal field development often entails placing the wells in prescribed locations within the reservoir. An error of a few meters in height above the oil/water contact or with respect to the roof may result in leaving behind a significant portion of the producible reserves. Driven by this key requirement, new technologies continue to emerge to help geologists, drillers, and reservoir engineers geosteer the wells. In recent years, two types of logging-whiledrilling (LWD) information have been used. On one hand, wellbore imaging can determine when a well path has left the reservoir and the angle of exit. On the other hand, traditional axisymmetrical resistivity logs help to quantify the distance to an approaching boundary through inversion, but fail to tell its azimuth. A newly deployed azimuthal deep resistivity instrument recognizes an approaching geological event before it intersects the well while continually imaging it at multiple depths of investigation. Of particular interest is the azimuth of approach with respect to the well path, advising in real time the most favorable change of direction. In addition, a series of transverse electromagnetic measurements specific to azimuthal resistivity, called geosignals, are presented. Geosignals help to quantify the distance and the rate of approach with great accuracy, before the actual intersection could occur. With this real-time information, geosteering engineers can remain at prescribed distances from important boundaries, including oil/water contacts and overlying shale roofs. Modeling and actual logs demonstrate that the new LWD instrument performs at its best when the reservoir is overlaid by shale. Modeling suggests, however, that suboptimal performance occurs in reservoirs with very resistive caprock, such as anhydrite.

Borehole resistivity simulations of oil-water transition zones with a 1.5D numerical solver

“…By substituting (16) and (18) into the variational formulation (20), and using (21) and (22), and the orthogonality of Bessel functions, for each Hankel mode ξ q , we obtain the following variational formulation:…”

“…Geophysical resistivity measurements are used to map the subsurface, explore hydrocarbon reservoirs, and maximize the production of the existing ones. We categorize existing resistivity measurements as: (1) on surface, such as those obtained using controlled source electromagnetic (CSEM) [1][2][3][4][5] and Magnetotellurics (MT) [6,7]; and (2) borehole resistivity measurements, for example, those acquired using logging-while-drilling (LWD) devices [8][9][10][11][12][13][14][15][16][17][18][19]. LWD devices are useful both for reservoir characterization [20][21][22] and geosteering purposes [23,24], which is the act of adjusting the tool direction to travel through a specific zone.…”

“…Recently, deep and extra-deep azimuthal logging devices have been introduced as a new type of LWD instruments [14,15]. In addition to map the subsurface, they help us to select the well trajectory properly in the hydrocarbon reservoir in order to increase the productivity of the well [14, [16][17][18]. There exist several differences between conventional LWD devices and deep azimuthal devices, e.g., the number of transmitters and receivers, and the spacings between them, which are significantly larger in deep azimuthal configurations.…”

Optimization of Well Trajectory Under Uncertainty for Proactive Geosteering

“…The data that are available for estimation include the depth of the top reservoir surface at the well entrance (marked by the black cross in Fig. 4a) and the directional-resistivity measurement-while-drilling (MWD) data (Bittar et al 2009;Chemali et al 2010). Three resistivity measurements with different directions and depths of investigation are assimilated every 6 m. The steering angle is optimized after assimilation of five sets of resistivity measurements (so every 30 m), on the basis of the most-up-to-date model realizations.…”

“…When uncertainty is quantified by a set of realizations, as in the EnKF, the solution of robust optimization is the one that minimizes the average cost function evaluated on this set of realizations (Chen et al 2009;Van Essen et al 2009). The continuous procedure of model updating and optimization for geosteering is very similar to the closedloop-optimization workflow for reservoir management, in which reservoir-simulation models are history matched to be consistent with production data and production strategy is optimized on the basis of the updated models (Sarma et al 2006;Jansen et al 2009;Lorentzen et al 2009;Chen et al 2009Chen et al , 2010.…”