Scaling Criteria And Model Experiments For Horizontal Wells

Abstract

Abstract Horizontal wells are gaining increased prominence in primary recovery and steam injection, yet experimental and theoretical modelling of horizontal wells is still inadequate. This paper explores these two areas. Equations were developed for three-phase and non-isothermal flow in the vicinity of a horizontal well, as well as flow inside the well. These were used to obtain scaling criteria for designing laboratory experiments. The question of skin factor due to perforated casing is also considered. An interesting idea offered is the use of variable diameter horizontal wells. A series of steam injection experiments was carried out in a model, with a concentric well. Temperature distributions were determined and used to interpret oil recoveries and pressure behaviour. Oil recovery performance for different experiments was evaluated to determine the effectiveness of different types of horizontal wells, and the impact of perforated casing. Introduction It has been reported extensively(1,2) that horizontal wells offer prospects of improved performance over conventional vertical wells, primarily due to the larger contact area between the formation and the horizontal well. This is especially true for thin oil formations where the reservoir contact area for a horizontal well can be hundreds of times that for a fully-penetrating vertical well, Furthermore, horizontal wells, when strategically placed, can reduce water/gas coning and result in improved cumulative recovery in view of the small draw downs(3). Horizontal wells are also believed to have a much better chance than vertical wells of intersecting systems of vertical and horizontal fractures in an oil-bearing formation, and thus lead to higher production rates(4). In steam injection recovery of heavy oils and oil sands horizontal wells especially when combined with steam-assisted gravity drainage could lead to high production rates, more uniform distribution of steam front leading to improved sweep efficiency and cumulative oil recovery(5,6). One aspect that has so far received little attention in discussions of horizontal well performance concerns the flow in the vicinity of, and inside a horizontal well. Often, it has been assumed that a horizontal well has infinite conductivity and that the flow in the well is predominantly laminar - meaning the pressure drop across [he well is not significant. Dikken(7) showed that this assumption is not necessarily true for most practical situations, and that pressure drops across horizontal wells can significantly affect the performance. This paper attempts to contribute to an understanding of this aspect of horizontal wells by presenting mathematical models of flow in the vicinity of, and inside a horizontal wellbore. These models were used to obtain scaling criteria for a series of scaled model steam injection experiments. Data, including well pressure drop and fluid production rates in the presence of a perforated casing are also reported. Literature Review Butler, McNab, and Lo(5) presented the first theoretical and experimental study of steam-assisted gravity drainage and horizontal wells to recover bitumen from oil sands. Equations were derived to calculate oil drainage rate and lateral position of steam-oil interface as functions of time and height. A subsequent study by Butler and co-workers(H) introduced improvements to the original model.