Horizontal wells are an industry recognized technology for developing thin oil columns. However, structural and/or heterogeneity in reservoir properties present challenges that effect horizontal inflow draining and initial wellbore cleanup. Strong bottom-water drives, coupled with inflow heterogeneity, can result in localized coning and early water breakthrough.

This article describes the deployment of precise real-time distance-to-boundary (DTB) mapping while drilling to optimize well placement and address uncertainties in the reservoir model. It also describes the application of nozzle-based inflow control device (ICD) technology in the downhole completion system, enabling inflow balancing to delay water breakthrough and reduce water-cut later in the production cycle.

The H oil field

The H oil field is located in the Pearl River Mouth basin in the South China Sea. (Images courtesy of Schlumberger)

The H oil field is located in the Pearl River Mouth basin in the South China Sea, approximately 119 miles (190 km) southeast of Hong Kong. It is a continental margin sedimentary basin formed during the rifting of the South China Sea in the late Mesozoic to early Tertiary. The main pay zones were deposited during the early Miocene as delta front bar and stacked fluvial-deltaic channel sands with thin shale layers. Oil accumulations with low gas saturation and no gas cap are mostly found in four-way-dip closures, which are associated with basement highs.

Oil was discovered in the area by Agip, Chevron, and Texaco in 1989. New exploration and appraisal wells drilled between 2002 and 2005 resulted in another significant discovery and the commercialization of two previously discovered fields. The field is currently operated by the CACT oil consortium, in which Agip and Chevron each have a 16.33% share and 32.66% respectively. China National Offshore Oil Corp. (CNOOC) is the majority shareholder with 51%.

Production history

Production profile for well A-1, April to May 2009.

The H field was put online in November 1991 with 20 wells producing from 12 multistacked reservoirs. After more than 15 years of high-rate oil production, oil-water contacts (OWCs) had risen, leaving thin 16-ft (5-m) or less remaining attic oil columns and causing high water production. Consequently, since 2006, sidetrack drilling programs were launched to improve drainage contact in the sand layers while producing at lower drawdown.

In some early horizontal wells, uneven sweep due to reservoir heterogeneity caused localized coning and high water-cut. To address these challenges, an integrated solution using precise real-time DTB mapping technology for geosteering with a rotary steerable system, coupled with downhole inflow control technology using nozzle-based ICDs integrated in the sand screen completions design, was implemented. Two wells were drilled and completed in 2008 using these technologies, representing the first ever installations of nozzle-based ICDs in China.

Distance-to-boundary mapping technology

The DTB technology is a deep azimuthal electromagnetic resistivity logging-while-drilling (LWD) system called the PeriScope bed boundary mapper. The tool makes 360° deep directional measurements to provide orientation of formation boundaries as far as 21 ft (6.4 m) from the borehole, using a combination of tilted coil technology and multiple frequencies and spacings. During drilling operations, LWD measurements are transmitted in real time to the surface. These unique symmetrized directional measurements, with maximum sensitivity to formation or fluid boundaries, make it possible to map boundaries in real time, independent of anisotropy and dip.

ICD

LWD measurements (top) and planned well trajectory and actual trajectory based on DTB technology (middle and bottom). Blue dots in the middle panel show the position of the top reservoir boundary as identified by the deep azimuthal electromagnetic resistivity.

The ICDs are designed to re-distribute downhole pressure in order to normalize fluid inflow along the producing interval of a horizontal well. This pressure drop re-distribution is designed to compensate for high drawdown at the heel sections of the well along with inflow variations due to permeability heterogeneity that may otherwise lead to localized coning and premature water breakthrough. Sizes and the numbers of nozzles are varied and fine-tuned on site based on LWD derived properties. In high permeability inflow zones, nozzles are designed to self-regulate to increase backpressure and suppress significant inflow surge to encourage a balanced inflow from the less permeable zones.

Well-A1

The first of the two case wells was geosteered close to the top of a layer within the multistacked reservoirs. It contains a 23-ft (7-m) attic oil column trapped in a gently dipping anticlinal structure with expected lateral variations in permeability. Predrill modeling and simulation predicted that the balancing of inflow enabled by ICDs would provide an estimated incremental 700 b/d of oil while reducing water cut from 76% to 62% after four months of production.

Reservoir properties derived from LWD, such as porosity-derived permeability and deep resistivity-derived saturation, were used for real-time fine-tuning of the ICD nozzle size and numbers configuration. It takes only 16 to 24 hours for an ICD design engineer to update the ICD model and refine the nozzle configuration. The ability to adjust ICD nozzle configurations at the well site is a significant advantage as the actual variations in lateral permeability always differ from predictions.

Well-test data for the first month of production from well A-1 indicated 2,000 b/d of oil with zero water-cut, higher flowing tubing pressure, and a better productivity index compared to previous wells that were drilled and completed with standalone sand screens. After 200 days of production, water-cut stabilized at about 40%.

Well-A2

Modeled production profile for well A-2 with and without ICDs and DTB well placement technology.

The second candidate well was a horizontal sidetrack designed to recover remaining reserves in a thin laminated sand layer after the casing of the original wellbore collapsed. The estimated net thickness of this reservoir was approximated at 16 ft (5 m). Close proximity to the OWC required the integration of precise placement of the trajectory within 1.6 ft (0.5 m) from the top of the sand and ICD downhole flow balancing to defer early water breakthrough. Hence, the decision was made to employ DTB technology for an optimum steered trajectory.

The DTB well placement technology accurately delineated — in real time — the top of the structure along the horizontal section of well A-2 while overcoming structural uncertainties and achieved 100% net to gross clean sand penetration over its horizontal length of 974 ft (297 m). It enabled successful placement of the horizontal section of well A-2 within 1.6 ft (0.5 m) from the top of the thin sand reservoir target, even though the actual top target came in 7.5 ft (2 m) shallower than expected.

Additionally, the 1,312-ft (400-m) horizontal production interval that was originally planned was not achieved due to an unexpected dipping of the top of reservoir. This was mapped with DTB technology that provided the asset team with the confident decision to prematurely stop drilling operations, avoiding any possibility of drilling closer to the OWC or out of the pay zone.

Since the oil layer was very thin, the impact of ICDs in preventing early water breakthrough was less effective than in well A-1. Well A-2 began production with a low water-cut of about 1%; however, water breakthrough occurred within less than a month. Water-cut increased to about 85%, but then stabilized despite increases of total fluid production from 3,000 b/d to 4,000 b/d. This indicated the effectiveness of the nozzles in conforming to the Bernoulli flow principle, and consequently, to self-regulate at higher production rates. Compared to standalone sand screens, ICD simulations in well A-2 illustrated an incremental 200 b/d of oil for water-cut in the range of 80%.

Conclusions

The production results from the first wells in China to be completed with nozzle ICDs indicate good justification for their application in thin oil column layered reservoirs. Previous wells drilled in similar reservoirs, but without ICDs, experienced earlier water breakthrough. Well A-2, drilled using advanced DTB well placement technology, demonstrated the ability of real-time geosteering to keep the trajectory within a very confined target window while mapping the reservoir. In addition to improving reserves and productivity through more accurate well placement, DTB technology can save rig time by avoiding the need for additional sidetracks. The installed ICD segments provided downhole flow regulation and influx balancing throughout the lateral well section, helping to optimize oil production and reserve recovery. Based on the success of these first two wells, CACT continues to employ DTB well placement technology and ICD flow control systems in more of its wells in the H oil field.