Summary Simple alterations to the completion can be used to enhance the performance of long horizontal wells producing from high-permeability formations. Production performance and operational feasibility of the investigated completion methods are compared. The methods should be considered for horizontal wells where frictional pressure losses along the perforated section restrict production performance. Introduction Inflow from the reservoir to a horizontal well in high-permeability formations can be severely restricted by the frictional pressure loss along the horizontal well (Fig. 1). For long horizontal wells with high production rates, considerable pressure loss is experienced along the horizontal part of the well. The liner pressure increases exponentially from the heel end toward the toe end of the well, and the drawdown and influx to the liner decrease accordingly. The effect of a declining drawdown along the well becomes important in high-permeability reservoirs where the pressure loss through the reservoir may be of the same magnitude as the pressure loss along the horizontal wellbore. The upstream part close to the toe of the well contributes less to the production than the downstream part close to the heel, and a severe reduction in well productivity is experienced. In thin oil zones, the drawdown will often be restricted because of gas and water coning, and maximum production rates may be limited as a result of coning problems rather than lack of downstream transport capacity. By reducing the effect from pressure loss along the wellbore, higher production rates can be maintained without causing gas or water coning problems to become more severe. In this study, optimization of completion design is based on three basic principles:reduction of the frictional pressure loss along the perforated part of the well,redistribution of the frictional pressure loss along the perforated part of the well by changing the flow direction in parts of the liner, andcreation of an optimal sandface pressure profile by introducing inflow control along the wellbore.

Summary An 11-month horizontal well production test was performed offshore Norway inthe giant Troll gas field to prove performed offshore Norway in the giant Trollgas field to prove possible thin oil zone reserves. The oil column in the testarea possible thin oil zone reserves. The oil column in the test area is only 22 m, and the 500-m horizontal well was positioned 4 m from the water zone and18 m from the gas zone. The well, tied into production and testing ship Petrojarl 1, was put on production in production and testing ship Petrojarl 1, was put on production in Jan. 1990, The initial oil rate from the horizontalwell was more than four times higher than that of a vertical well in the samearea, Test results show that horizontal wells represent a viable technology foreconomic oil production from the thin oil zone in the Troll field. Introduction The Troll field, located below 300 m of water offshore Norway (Fig. 1), contains 0- to 26-m-thick oil rims sandwiched between a large gas cap and anactive aquifer. In the westernmost fault block of the field, the Troll West oilprovince, the oil zone is the thickest, 22 to 26 m. The oil in place (OIP) ofthis province is estimated to be 121 × 10(6) stock-tank m3. The oil in the Troll West oil province is located in high-quality sands withpermeabilities from 3,000 to 10,000 md. The oil production is limited by gasconing, resulting in rapidly production is limited by gas coning, resulting inrapidly decreasing oil rates. Developments by vertical wells therefore haveconsistently exhibited marginal economy. Horizontal wells are known to improve well productivity and to reduce water-and gas-coning problems. The Helder field, offshore The Netherlands, wasredeveloped in 1987–88 with 10 horizontal wells. Results show improvedvolumetric sweep, reduced water coning, and productivities up to 20 timeshigher than for vertical wells. In the Prudhoe Bay field in Alaska, manyhorizontal wells have been drilled. A 5-mm-long horizontal well in this fieldyielded productivities 1.5 to 3 times that of vertical wells. The wells alsoare successful in reducing gas coning. In the Troll field, pretest simulation studies indicated that developmentbased on 500-m horizontal wells could offer an economically attractive oildevelopment. Because of the 300-m water depth, highly unconsolidated reservoirrock, and a thin oil column, however, development with horizontal wells wasconsidered a high-risk project. To confirm the horizontal-well potential and long-term behavior, and therebyreduce the risk involved, a decision was made in June 1989 to perform an8-to-12-month production test with a 500-m horizontal well in the oil zone ofthe Troll West oil province. province. The well, operated by Norsk Hydro A.S., was completed in Dec. 1989, and production from the test ship began in Jan.1990. When the test concluded after 11 months, the well had produced 1 043 000stock-tank m3 of oil and the water cut had stabilized at about 35%. Background The Troll field covers 700 km2. The estimated gas in place (GIP) and OIP are1670 × 10(9) std m3 and 615 × 10(6) stocktank m3, respectively (Fig. 2). Twomain north/south faults divide the field into three provinces: Troll West oilprovince, Troll West gas province, and Troll East (Fig. 2). The Troll West oilprovince is province, and Troll East (Fig. 2). The Troll West oil province isinto the Southern and the Northern oil provinces (Figs. 3 and 4). Concession Block 31/2 (Fig. 2), operated by A/S Norske Shell, was awarded in1979. Blocks 31/3, 31/5, and 31/6 were awarded in 1983 with shared operatorshipbetween Norsk Hydro A.S., Den norske Stats Oljeselskap A.S., and Saga Petroleum A. S. Phase 1 of the Troll development, as proved in 1986, calls for gas Phase 1 of the Troll development, as proved in 1986, calls for gas production to startfrom Troll East in 1996. Before the horizontal test production to start from Troll East in 1996. Before the horizontal test well, 26 exploration wells hadbeen drilled within the four blocks; hydrocarbons were encountered in 22 wells. Two-dimensional seismic surveys Cover all four blocks. To select the horizontalwell location accurately, a 3D seismic survey was carried out during the summerof 1989. Reservoir Description The Troll field is contained within three easterly tilted fault blocks(Figs. 2 and 3). The reservoir interval is of Middle to Upper Jurassic Age, andthe reservoir sediments consist of clean, medium to coarse sand interbeddedwith micaceous and silty, very fine to fine sand. These sediments are generallyvery poorly consolidated. A typical feature of the reservoir sediments in the Troll field is thenumerous calcite-cemented layers occurring in all lithologies in all wells. There are two classes of calcite-cemented layers: extensive (up to severalkilometers) wide sheets found at the boundaries of the geological zones and theless extensive (1 to 100 m) calcite-cemented layers located within thegeological zones. JPT p. 914

The Norwegian North Sea Brage field is currently on decline from its plateau oil rate of 19 400 std m3/d. This paper summarizes five years of production experience from the two highly different reservoirs that have been developed so far. The fluvial, Lower Jurassic Statfjord formation consists of 1000 to 4000 md sandstone reservoir, with excellent vertical and lateral communication properties. A recovery of approximately 64% is expected from waterflooding the highly undersaturated 36 API oil. The shallow marine Fensfjord formation, of Middle Jurassic Age, is a stratified sandstone reservoir with average zone permeabilities in the range of 1 to 200 md. Calcite layers and high permeability strikes, with up to 5000 md permeability, amplify the heterogeneity and complexity of this reservoir. The oil has similar properties as in Statfjord, but a recovery of only about 32% is expected. The principal recovery mechanism is aquifer support aided by water injection from the flanks. The initial Fensfjord development was based on vertical wells with perforated cemented liner. Early production experience proved sand production to be a major problem. To avoid sand production, two wells were hydraulically fractured with proppants. This concept has later been left in favour of horizontal wells with sand screen, sliding sleeves and external casing packers (ECP's). New water injection wells are horizontal. In these wells the horizontal section is divided into two parts: the toe part is completed with a sand screen and accessed through tubing; the heel part is a perforated liner and accessed through the annulus. In this way the two parts can be controlled as two individual injectors. An additional benefit is the possibility of producing oil from the toe part of the well while at the same time injecting in the annulus section. A water-alternating-gas (WAG) injection pilot was established in the Fensfjord reservoir in 1994. Following a successful one-year pilot, with emphasize on performance monitoring through the use of tracers and mapping of high permeability streaks, the WAG process has been gradually expanded to include six injectors. To avoid sulphate scale, injection of seawater has been replaced by injection of formation water produced from Tertiary sands in the shallow Utsira formation. P. 103

The importance of friction pressure loss within the completed section of horizontal wells, under some reservoir conditions, has been analyzed. The phenomena were studied both analytically and experimentally with sub sequent implementation in a commercially available 3-D,3-phase reservoir simulation model. Based on simulation studies, criteria for the selection of appropriated imensions for horizontal wells under different reservoir conditions are outlined. A field example from the Troll Field offshore Norway where these results were applied during the planning of a horizontal test well, is presented. Introduction The Troll field, located in 300m water depth offshore Norway (Fig. 1), contains 0–26 m thick oil rims sandwiched between a large gas cap and active aquifers. In the westernmost fault block of the field, the Troll WestOil Province, the oil zone is at its thickest; 22–26 m. The oil in place of this province is estimated at 121x10 6 Sm3.The reservoir contains several highly permeable, unconsolidated sands separated by low permeability micaceous sands. Impermeable calcite cemented horizons with imperfectly known spatial disposition and extensions occur within the high permeability sands. The oil production from Troll may be limited by gas coning, resulting in a limited available drawdown for the well. Since the oil rim is only 22–326 m thick, the drawdown can only be in the range of 1.5 to 2 bar in order to avoid the production of free gas-cap gas. The recoverable oil in the Troll West Oil Province is located in high-quality sands with permeabilities ranging from3000 to 10 000 md, so high liquid rates were expected(Fig. 2). With the combination of limited drawdown and high rates, wellbore friction pressure loss was considered to be of importance [1].The subject has been studied analytically by Dikken [1]. However, implementation of the effect of horizontal well friction in simulators with subsequent application for well planning and design purposes has previously not been reported. Horizontal well friction was given due attention during the planning of a 500 m horizontal well in the Troll field. The well was drilled and completed in late1989, and has been on production since January 10th 1990. The cumulative oil production as of early June1990 was 500 000 Sm [2]. Background The Troll field covers an area of 700km2 . The estimated gas and oil in place is 1670x10 9 Sm3 and 615x10 6 Sm3respectively (Fig. 3). Two main north-south bound faults divide the field into three provinces: Troll West Oil Province, Troll West Gas Province and Troll East(Fig. 3).Concession Block 31/2 (Fig. 3), operated by A/SNorske Shell, was awarded in 1979. Blocks 31/3, 31/5, and 31/6 were awarded in 1983 with shared operatorship between Norsk Hydro a.s, Den norske stats oljeselskapa.s, and Saga Petroleum a.s.

This paper presents the results of two long-tenn horizontal well tests in the Troll field. Planning and evaluation of the tests and simulation of and consequences on field development are summarized. The two tests demonstrated significant production potential from 12-to 22-m oil columns and verified the pretest assumption that one horizontal well could replace four vertical wells.