Introduction

The study on remaining oil in high water cut oilfield is a fundamental for oilfield development and enhanced oil recovery, which is a worldwide highly comprehensive research problem and hot issue. Today, advanced technologies of horizontal drilling and application of advanced well completion such as ICD turn the development of unswept oil into a reality.

Extreme water production is one of the main reasons which contribute in decrease of productivity of wells. Rising water cut influences inflow performance and tubing performance curve negatively. Excessive water production enhances the expenditure of lifting of fluids to surface, treatment of water, and disposal. Production of water is also related to scale formation at different production system components (Al Hasani et al. 2008).

Determining good well locations and completions of production and injection wells are important to optimally develop and produce a reservoir (Forouzanfar and Reynolds 2013). Dosunmu and Osisanya (2015) show that horizontal well productivity is significantly affected by oil viscosity, horizontal permeability, and well diameter. Al Qahtani et al. (2015) showed a noble idea of taking full advantage of long horizontal well if it is produced from both ends, i.e. from heel and toe.

Oil recovery is increased by appropriate selection of ICD, which also stops flow of unwanted production of water and gas. Results obtained from fields and simulation studies show that production of oil, gas, and water is very much dependent on the type of ICD and number of ICD used (Mojaddam Zadeh et al. 2012).

Well placement in a bottom water drive reservoir poses several challenges. The real oil fields are complex environments due to heterogeneities, presence of geologic discontinuities (e.g., faults, fractures, and very high and low permeability zones), and geologic uncertainties. Moreover, coning tendency in vertical well leads to early water breakthrough, which can be avoided using horizontal well placement techniques with optimized length, rate and offset. Further smart horizontal wells are also being used extensively for maximizing recovery. A proper methodology needs to be developed for avoiding early water breakthrough and increase water-free production.

Objective of the study is to carry out sensitivity study on offset, withdrawal rate, horizontal length, compartment length and nozzle sizes in three-dimensional numerical simulation model.

Methodology

In this work, methodology adopted for determination of optimum variants for vertical well, horizontal well and smart horizontal well is given below.

  • Static modeling.

  • History matching.

  • Sensitivity analysis.

    • Vertical well (offset from woc, withdrawal rate).

    • Horizontal well (offset from woc, withdrawal rate, horizontal length).

    • Smart horizontal well (compartment length, nozzle size).

  • Economics and recovery factor.

Static modeling

Pillar gridding process in petrel is used to make the static model. The original model was kept intact with minor modifications. A number of NE–SW trending normal faults of limited lateral continuity intersects the structure at the Barail Arenaceous level. Figure 1 shows the fault pattern of the reservoir.

Fig. 1
figure 1

Fault pattern of the reservoir

Full size image

Geological grid was made with cell size 75 m × 75 m using grid rotation of 40°. Once the grid was made, structure and properties were upscaled and the total number of cells in the model is now 1,035,960. The total number of cells present in the geomodel prior to upscaling is 8,171,520 which were further upscaled to 100 × 100.

Five different rock types (RT) have been kept in the model (RT 1–5). RT 1 is the best rock type where as RT 5 is bad rock type in the study. Rock-type statistics expressed in terms of porosity, permeability and oil saturation is presented in Table 1. Porosity, permeability and reservoir thickness distribution of the geocellular model (GCM) have been shown in Figs. 2, 3 and 4, respectively.

Table 1 Statistics for different rock type
Full size table
Fig. 2
figure 2

Porosity distribution

Full size image
Fig. 3
figure 3

Permeability distribution

Full size image
Fig. 4
figure 4

Reservoir thickness distribution

Full size image

History match

Observed data (oil, gas, water rates and pressure data) for 46 wells for duration of 16 years was computer generated with an industry standard simulator. Outcome of the history match on reservoir level is illustrated in Fig. 5. Well-wise flowing/static pressure data, gas–oil ratio and water cut were used as parameters for history matching. From Fig. 5, it is evident that GOR, pressure, water rate, oil rate have been matched reasonably well with the observed data by the industry standard simulator at field scale.

Fig. 5
figure 5

History match results

Full size image

Performance prediction

The oil flow rate, vertical and horizontal permeability, oil viscosity, oil formation volume factor, oil–water mobility ratio and the perforation interval length are the major factors which affect the breakthrough time of water and gas (Wagenhofer and Hatzignatiou 1996). Evaluation of some of these parameters has been carried out to see their effect on production behavior. Performance prediction of the simulation case has been carried out in three parts, which is listed below:

  • vertical well (offset from WOC, withdrawal rate).

  • Horizontal well (offset from WOC, withdrawal rate and horizontal well length).

  • Smart horizontal well (compartment length, nozzle sizes).

Sensitivity study of vertical well—offset and withdrawal rate

The history-matched model was used to perform sensitivity studies on offset from WOC and withdrawal rate. All variants are been mentioned in Table 2. From all combinations mentioned in Table 2, 15 cases were constructed, details of which is mentioned in Table 3.

Table 2 Sensitivity parameters for vertical well
Full size table
Table 3 Cases examined for vertical well
Full size table

All sensitivity studies in prediction mode were simulated using production rate control mode. Constraint on production at well scale was specified as per the capacity of surface handling equipments. Apart from above constraints, wells were subjected to some additional constraints which is shown in Table 4.

Table 4 Well level constraints for prediction runs
Full size table

Sensitivity study of horizontal well—offset, withdrawal rate and horizontal length

The history-matched model was used to perform sensitivity studies on offset from WOC, horizontal well length and withdrawal rate. All variants are mentioned in Table 5. From these grouping 48 cases were prepared, details of which is given in Table 6.

Table 5 Sensitivity parameters for horizontal well
Full size table
Table 6 List of cases for horizontal well
Full size table

All sensitivity studies in prediction mode were simulated using production rate control mode. Constraint on production at well scale was specified as per the capacity of surface handling equipments. Apart from above constraints wells were subjected to some additional constraints which is shown in Table 4.

Sensitivity study of inflow control device—compartment length and nozzle sizes

A horizontal well does not contribute equally throughout its length. Heel contribution of production is more than from toe because of frictional pressure drop along wellbore length, which leads to non-uniform influx throughout the wellbore length. This effect also known as heel–toe effect which leads to reduced recovery of oil because of early water/gas breakthrough and uneven area swept area near producing well. To resolve this issue, inflow control device (ICD) has been widely used in producing wells for controlling and optimizing individual well or overall reservoir performance. The sole purpose of ICDs is to make equal inflow along the entire wellbore length despite location and permeability variation. Application of ICDs allows the entire horizontal length of the wellbore to add to the total production and thereby optimize hydrocarbon recovery.

Careful choice of ICD increases the oil recovery and constrains unwanted production of water and gas. IT has been observed from actual field experience and outcome from the extensive simulation studies that the oil, water and gas production are extremely reliant on the type and arrangement of ICD (Mojaddam et al. 2012). A suite of optimized ICD modeling workflows and evaluation tools were developed to address these issues using a plug-in architecture in an industry-leading exploration and production platform (Schlumberger PETREL 2015) and reservoir simulator (Schlumberger 2015) Historically, the most common method for designing ICD completions has been to use steady-state or pseudo steady-state (i.e. static) analytical modeling using data at one or more points in time (Ellis et al. 2009).

ICD design has been done using equal length compartment and variable nozzle sizes in each ICD.

Equal length compartments with uniform nozzle sizes—sensitivity on packer length

A design with equally spaced compartments and uniform nozzles sizes is preferred if there is significant uncertainty in the reservoir characterization or risk in getting the completion to TD. Different packer length (Table 7) has been selected for simulation runs using ICD well completion.

Table 7 Compartment length as an input to ICD cases
Full size table

Equal length compartments with variable nozzle sizes—sensitivity on nozzle sizes

Equally, spaced compartments with variable nozzle sizes designs are the preferred option if the reservoir is homogeneous with significant heel-to-toe pressure decline along the wellbore.

High-rate horizontal wells often suffer from the so-called heel–toe effect. This is caused by the frictional pressure gradient along the well, which means that the heel has a significantly lower pressure than the toe and so produces more fluid. This can lead to premature breakthrough of water or gas. A total of 35 cases have been examined for this case, details of all the cases are given in Table 8.

Table 8 Details of all the cases examined for variable nozzle sizes
Full size table

Equal length compartments with variable nozzle sizes—uncertainty on nozzle sizes

After determining most sensitive parameter from previous section, uncertainty analysis is carried out on variable A, B E and F. Experimental design setup was done for the four variables using full factorial design which gives a total of 16 cases to be examined, additionally one central point has also been added to make the no. of simulation runs 17. Details of all the cases are given in Table 9.

Table 9 List of uncertain variables for cases examined
Full size table

Results and discussion

Vertical well sensitivity results

Results of sensitivity of vertical well are plotted in Fig. 6. It can be observed that cumulative production increases with increasing assigned rate from 50 to 150 m3/day. Cumulative oil production increases with increase in offset from WOC. Incremental production is maximum by increasing the assigned rate from 50 to 100 m3/day.

Fig. 6
figure 6

Sensitivity results of vertical well

Full size image

In bottom water reservoirs with active aquifer, proper planning and completion of the well(s) in vertical direction will require the consideration of fractional well penetration to achieve optimum critical oil production rate and breakthrough time (Okon et al. 2018).

Cumulative oil production increases with increase in offset from WOC and production rate. There are inverse relationships between water cut at end of prediction and cumulative oil produced, as well as between water cut at end of prediction and offset from WOC (Kumar et al. 2017a, b).

Controlling parameters for economical and optimal production rate from vertical well in a bottom water drive reservoir are optimum perforation interval and systematic approach (Gamal et al. 2016).

Horizontal well sensitivity results

Sensitivity results of horizontal well are plotted in Fig. 7. It can be observed that cumulative oil production increases by assigning higher withdrawal rates (200,250,300 m3/day). For maximum recovery, horizontal well length should vary between 600 and 800 m; additional increment in horizontal well length does not increase in oil production. Increasing horizontal well length from 600 to 1200 m does not increase the oil production cumulative linearly; maximum value is obtained for 800 m horizontal length in 36 cases out of 48 cases.

Fig. 7
figure 7

Sensitivity results of horizontal well

Full size image

In bottom water reservoirs with active aquifer, placement of horizontal well should consider stand-off to drainage width as well as horizontal well length to drainage width to achieve optimum critical production rate and breakthrough time (Okan et al. 2018).

Offset from WOC, horizontal length, and withdrawal rate have an impact on cumulative oil production in case of horizontal well. Maximum impact on cumulative oil production is given by offset from WOC followed by withdrawal rate and last on horizontal well length (Kumar et al. 2017a, b).

ICD sensitivity results

Sensitivity on compartment length

To maximize cumulative oil production from ICD, it is pertinent to determine compartment length/number of nozzles and nozzle size in each ICD. Initially sensitivity on compartment length has been carried out and it is observed from Fig. 8 that, 150 m compartment length or seven nozzle ICD give maximum cumulative oil production.

Fig. 8
figure 8

Sensitivity results of compartment length

Full size image

Sensitivity on nozzle sizes

To determine nozzle size for each ICD for maximizing cumulative oil production, it is essential to determine which nozzle ICD is having maximum impact on recovery. A total of 39 cases have been run and results are plotted in form of a Tornado plot (Fig. 9) where variable which is most sensitive to recovery efficiency is on top and least sensitivity is on bottom.

Fig. 9
figure 9

Tornado plot

Full size image

Equal length compartments with variable nozzle sizes—uncertainty on nozzle sizes

Nozzles B, F, E and A have maximum impact on recovery and hence a detailed uncertainty analysis has been run. A comparison of oil production cumulative is done in Fig. 10 for all the cases examined. Figure 11 compares the production profile of minimum (2548_800_300_150_50) and maximum recovery (2548_800_300_150_39) efficiency.

Fig. 10
figure 10

Comparison of oil production cumulative for different nozzle sizes

Full size image
Fig. 11
figure 11

Production profile comparison of minimum and maximum recovery

Full size image

The number of ICDs, the number of nozzles in each ICD, and the size of nozzles are optimized to maximize the oil production for different reservoir condition. This optimal design, which consisted of five compartments in the horizontal section with 14 ICD joints, proved to be more effective in delaying water breakthrough and increasing recovery (Maalouf et al. 2017).

Economics and recovery factor

Economics and recovery efficiency for various cases has been compared in form of a table which is presented in Table 10. Table 11 lists all the assumptions in working out the economics.

Table 10 Results for various sensitivity cases
Full size table
Table 11 Economic assumptions
Full size table

Conclusions

Withdrawal rate and offset from WOC have an effect on oil production for vertical well. From sensitivity study on offset and withdrawal rate, it was observed that offset has more impact on cumulative oil production than withdrawal rate. Highest recovery of 0.63 MMm3 cumulative oil is obtained using offset of 20 m and initial withdrawal rate of 150 m3/day.

Horizontal well length, offset from WOC and withdrawal rate have an effect on oil production for horizontal well. Highest impact on oil production is given by offset from WOC; second big effect is given by withdrawal rate and finally horizontal well length effects oil production. Highest recovery of 0.95 MMm3 cumulative oil is obtained using optimal horizontal well length varying between 600 and 800 m offset of 20 m and initial withdrawal rate of 300 m3/day. Above results when compared with producing well number 49, cumulative oil production of 0.152 MMm3 is achieved during the entire production history with offset of 10 m, horizontal well length of 400 m and initial withdrawal rate at 200 m3/day.

Compartment length and nozzle size have an effect on oil production for smart horizontal well (ICD). Maximum effect on oil production is given by nozzle size in each ICD followed by compartment length. Seven ICDs and compartment length of 150 m with variable nozzle sizes give maximum recovery with cumulative oil production of 1.08 MMm3.