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

Oil production from thin oil zones sandwiched between a gas cap and an active aquifer is often limited to uneconomic rates due to water and gas coning. Horizontal wells seem to provide a promising solution to the coning problem. In order to avoid premature gas breakthrough, the horizontal wells should normally be positioned as far away from the GOC as possible. This assumption has been examined further; wells completed below the WOC have been compared to wells in the oil zone. This strategy relies upon oil coning into the well completions through the water zone; the so called "inverse coning" process.In this study, production from the 12 m thick oil zone in the Troll West Gas Province has been simulated with a 500 m horizontal well completed in the water zone below the WOC. The effects of various rock and fluid parameters have been evaluated, and a correlation for time to gas breakthrough has been developed based upon the simulation results.The calculated gain in oil production with the well below the WOC ranged from 15% to over 50% in this study, depending on the reservoir and operating conditions. Based upon cumulative oil production considerations, an optimal completion depth of 3-4 m below the WOC was determined. Cone behaviour was shown to be dominated by permeability level, fluid mobilities, completion depth and liquid production rate. The primary disadvantage with these wells is the sustained increased water cut.References and illustrations at end of paper

This paper was selectad for presentation by the OTC Program Committee following review of information mntained in an abatract submitted by tha author(s). Contents of tha paper, as presented, hava not been reviewed by tha Offshore Technology Conference and are subject to correot~n by tha author(s). The mater[~, = wsent~d, does not neces=rilY reffeã nv oosition of the OfLshoreTochnoloov Conference or Ikeofficers. Permission to copy la restricted to an abstract of not mora than 300 words. Illustrations may not be copied. The abstract shb~ld contain conaplcuoua acknowf&dgmant of where and by whom the PaPer ia presented. ABSTRACT The 22-26m thick oil zone in the Troll West Oi[ Province, offshore Norway, is planned to be developed by 18800m long horizontal wells. The viability of such wells has been proven by long-term testing. To obtain economic oil production the wells will be completed in the lower part of the oil rim close to the oil-water contact. The high water production which is expected can, to some extent, be controlled by gas reinfection. A flexible gas injection scheme will therefore be implemented. Allowing wells to produce free gas together with the oil (super

A number of unconventional techniques were used for constructing an integrated three-dimensional (3D), three-phase numerical reservoir model of the huge Troll field. The selected techniques included comer-point geometry (CPG), non-neighbor connections between grid cells, local grid refinement, improved vertical equilibrium (VE) description, and oilwell coning functions. By combining these techniques, an efficient model, capable of handling several complex reservoir problems simultaneously, was developed. This model became a flexible tool Jor reservoir management planning.