Abstract

We present a heuristic method of delineating reservoir regions with favorable production potential for well placement. The method uses basic concepts of proxy formulation for well productivity to relate the production potential of a well placed in any region in the reservoir, to the basic properties of that region (petrophysical, dynamic, geometrical). These properties may be readily obtained from any history-matched or data-conditioned reservoir model. Therefore, a productivity potential map for the reservoir can be easily generated. Also, the relevant properties relate to parameters measurable in an LWD operation. We use this technique to simulate the impact of alternate steering strategies on well path and productivity. This requires a fine-scale model of the region designated for drilling (i.e., model prior to upscaling). The influence of additional LWD measurements on reservoir trajectory and expected well performance can therefore be assessed with this technique (e.g., impact of NMR-derived permeability). We also examine how this method may be applied when multiple geological scenarios are conceivable for the area targeted for drilling. We apply a simple technique for model aggregation that proves effective in identifying drainage areas that are optimum or near-optimum in an overall sense. Therefore, a well placed in the region identified by the aggregate model shows a performance with respect to the underlying geological models that is best or not far from it.

Introduction

The objective of this work is to apply the basic concepts of reservoir engineering to the problem of well placement planning, and assess the feasibility of planning well location and steering strategy to achieve maximum production potential. This work is relevant to application of advanced measurement and drilling technologies, LWD and rotary steerable systems.

We divide the problem into two parts, one concerning the selection of drilling target, the other of steering strategy. However, we seek a formulation that creates a natural linkage between these two parts.

We propose a function in terms of the combined attributes of each reservoir gridblock (petrophysical, dynamic, and geometrical) and observe that this function shows good correlation to the production potential of wells intersecting those gridblock. We examine this for the case of horizontal wells in heterogeneous water-drive reservoirs. The parametric form of this function (called productivity proxy function) is abstracted from Darcy's law, but adjustments are made to improve the correlation with flow simulation results, which also depends on the operating plan for the well (time horizon, operating constraints). The key adjustment is the incorporation of the effect of neighboring gridblocks through a thickness-averaging scheme, such that the connectivity of each region to the underlying drive mechanism is taken into account.

One advantage of the proxy approach is that it allows a quick identification of reservoir regions that show good production potential, using information readily available from any history-matched or conditioned reservoir model. Combination of this technique with more exhaustive well placement techniques1–2 can therefore be beneficial.

Another advantage of the proxy approach is that the parametric group is expressed largely in terms of parameters measurable in an LWD operation. Therefore, it is possible, at least in the LWD planning phase, to use this technique to assess the impact of additional measurements on well path and productivity (porosity, saturation, permeability, pore pressure). Application of this technique to real-time steering is also worthy of consideration. Figure-1 shows a basic workflow for planning, operation and evaluation of drilling operations.

This content is only available via PDF.
You can access this article if you purchase or spend a download.