SPE Members

Abstract

This paper describes the recovery performance of horizontal wells in mixed pattern waterfloods. It extends the present streamtube modeling techniques to horizontal wells by approximating the horizontal wellbore with a finite, distributed line source in two-dimensions. Together with the conventional (vertical well) solution, art integral expression for the stream function results. Streamlines extracted from this expression were used to implement waterflood recovery performance calculations.

Recovery performance at breakthrough and post breakthrough for mixed five-spot patterns were analyzed using horizontal wellbore lengths of 25%, 50%, 75% and 100% of well spacing. Results demonstrate that recovery at breakthrough is a decreasing function of wellbore length when the injection and production wells are different kinds (vertical, horizontal). Three mobility ratios (M = 0.3, 1.0, 10.0) were used to study how favorable and adverse mobility ratio fluids affect recovery performance of horizontal waterfloods.

Introduction

Horizontal wells are probably the most interesting issues in the oil industry today. Intuitively, at least, it seems plausible that horizontal wells may offer certain advantages over conventional (vertical) wells such as intersecting more fractures, improving sweep efficiency and accelerating oil recovery. Most of the work reported in the literature is in the areas of horizontal well testing and horizontal well productivity. Usually, in these studies, the horizontal well is placed parallel to one of the coordinate axes in a rectangular reservoir. Then either an analytic or a numerical solution is sought to the governing partial differential equations and pertinent boundary conditions.

Broman et al. recently reported one field application of horizontal wells as peripheral waterflood producers and peripheral gas injectors in the Prudhoe Bay Unit (PBU). It was reported that the main production mechanism in the PBU was governed by a large expanding gas cap. The peripheral areas, however, received limited pressure support from the gas cap and consequently, sections were waterflooded in a pressure maintenance program. It was further reported that the success of these operations have significantly encouraged future potential applications of horizontal wells to include additional waterflood development as injectors and/or producers.

One numerical simulation study has also been recently reported. The authors addressed the issue whether horizontal wells will offer significant increase in oil recovery at an economic limit because of improved sweep efficiency. They conducted comparative simulation studies of horizontal and vertical wells in a host of different scenarios. It was concluded that only in few cases does a horizontal well show an advantage in ultimate oil recovery especially at some specified water-cut or gas-oil ratio. However, significant differences in ultimate recovery were reported when the economic limit was based on a minimum oil rate.

Broman et al. did not report if their field implementation was a result of a simulation study or field experience. Clairidge and Olu, on the other hand, considered, as one of their seven example "typical reservoir situations", a steam-drive followed by a water-drive in a horizontal layered reservoir. The paper gave an excellent overview treatment of sweep efficiency comparisms of horizontal and vertical wells for the scenarios considered; however, application of horizontal wells in waterflooding was not its primary focus.

There is no other published literature on the situability of horizontal wells for waterflooding. Many investigators, however, have studied applications of horizontal wells in steamflooding. Among these are laboratory investigations, theoretical studies, numerical simulation studies, and field pilot implementation. Butler and Petela presented a theoretical pilot implementation. Butler and Petela presented a theoretical analysis for a heavy oil recovery process which involved downward vertical steamflooding between parallel horizontal wells. They used the rate of advance of the steam zone along the central steam surface in the reservoir to estimate a theoretical breakthrough time. Huang and Hight reported a simulation study which evaluated steamflood processes with horizontal wells.

P. 353

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