APPLIED RESERVOIR SIMULATION PROJECTS REPORTS PDF

Preview

Summary

REPORTS OF GROUP PROJECTS ON APPLIED RESERVOIR SIMULATION BY JOHN IDOKO JOB Student ID: 40462 Msc. Student, Petroleum Engineering Department, AUST Abuja Submitted to: Prof. David O. Ogbe Head of Petroleum Engineering Department, AUST Abuja Email: [email protected] Phone: +2348058643856, +23480337927...

Description

REPORTS OF GROUP PROJECTS ON

APPLIED RESERVOIR SIMULATION

BY

JOHN IDOKO JOB Student ID: 40462 Msc. Student, Petroleum Engineering Department, AUST Abuja

Submitted to:

Prof. David O. Ogbe Head of Petroleum Engineering Department, AUST Abuja Email: [email protected] Phone: +2348058643856, +2348033792781

June, 2017.

Contents WORKSHOP NO. 1: RUNNING ECLIPSE COMMERCIAL SIMULATOR ....................................................................................... 2 METHODOLOGY .................................................................................................................................................................. 2 Case 1; Predictive Runs Without the Effect of Gas Resolution on the Oil ...................................................................... 2 Case 2; Investigating the Effect of Gas Resolution on the Oil ......................................................................................... 2 RESULTS PRESENTATION ..................................................................................................................................................... 2 RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS ...................................................................................................... 3 PROBLEMS ENCOUNTERED AND SOLUTION APPROACH .................................................................................................... 4 WORKSHOP NO. 2: PLACING WELLS IN THE SIMULATOR ....................................................................................................... 5 METHODOLOGY .................................................................................................................................................................. 5 RESULTS PRESENTATION ..................................................................................................................................................... 5 RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS ...................................................................................................... 6 PROBLEMS ENCOUNTERED AND SOLUTIONS APPROACH .................................................................................................. 6 WORKSHOP NO. 3: GRIDDING ................................................................................................................................................ 7 METHODOLOGY .................................................................................................................................................................. 7 PART I .............................................................................................................................................................................. 7 PART 2 ............................................................................................................................................................................. 7 PART 1 RESULTS PRESENTATION ........................................................................................................................................ 7 PART 2 RESULTS PRESENTATION ........................................................................................................................................ 8 RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS ...................................................................................................... 8 MAJOR OBSERVATIONS AND CONCLUSION.................................................................................................................... 9 WORKSHOP NO. 4: IMPES VS. IMPLICIT FORMULATIONS .................................................................................................... 10 METHODOLOGY (BRIEF) .................................................................................................................................................... 10 RESULTS PRESENTATION ................................................................................................................................................... 10 RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS .................................................................................................... 10 WORKSHOP NO. 5: MODELING AQUIFERS............................................................................................................................ 11 METHODOLOGY ................................................................................................................................................................ 11 RESULTS PRESENTATION ................................................................................................................................................... 11 RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS .................................................................................................... 11 REFERENCES .......................................................................................................................................................................... 12

1

WORKSHOP NO. 1: RUNNING ECLIPSE COMMERCIAL SIMULATOR METHODOLOGY A 10 × 10 × 3 grid system, black oil model was used to simulate two cases of gas injection with an injector located in Block 1,1 and completed at cell 1,1,1; and an oil producer passing through block 10,10 and completed at cell 10,10.3. Case 1; Predictive Runs Without the Effect of Gas Resolution on the Oil The keyword ‘DRSDT’, was equated to zero in the SCHEDULE section of the input file. This implies that there is no gas resolution in the oil. Results from the runs are given in Figure 1 and Table 1 below. Case 2; Investigating the Effect of Gas Resolution on the Oil The keyword ‘DRSDT’, was nullified (commented out) in the SCHEDULE section of the input file. This allows for gas resolution in the oil. Results from the runs are given in Figure 2 below and Table 1 below.

Problem 3: Comparison of the results of Case 1 and Case 2. Figures 3 to 6 compares the performance of the reservoir, under production, when there is gas resolution in the oil (Case 1) and when there is no gas resolution in the oil (Case 2).

RESULTS PRESENTATION

Figure 1: Results of predictive runs without the effect gas resolution in oil (CASE 1)

Figure 2: Results of predictive runs with consideration for the effect of gas resolution in oil (CASE 2)

2

Figure 3: Comparison of Field Oil Production Rates for Case 1(GREEN) and Case 2 (BLUE).

Figure 4: Comparison of Produced GORs for Case 1(GREEN) and Case 2 (RED)

Figure 5: Comparison of Injector Well BHPs for Case 1(RED) and Case 2 (GREEN)

Figure 6: Comparison of Producer Well BHPs for Case 1(RED) and Case 2 (GREEN)

TABLE 1; ANNUAL PRESSURES OF THE BLOCKS WHERE THE PRODUCER AND INJECTOR ARE LOCATED CASE 1 (No Gas Resolution in the oil) CASE 2 (Considering Gas Resolution in the oil) Pressure at

Pressure at

Pressure at

Pressure at

PRODUCER

INJECTOR

PRODUCER

INJECTOR

BLOCK (psia)

BLOCK (psia)

BLOCK (psia)

BLOCK (psia)

0

4800

4783.1021

4800

4783.1021

1

4562.8037

6336.7627

4435.1

6237.38

2

5476.1211

6835.8672

5155.49

6749.4

3

4494.4424

6470.3086

5922.63

7293.81

4

4056.1846

5718.5371

5168.27

7229.25

5

3870.52

5102.793

4290.81

6004.47

6

3614.6599

4889.6079

3900.44

5279.47

7

3516.5417

4708.8442

3663.84

4842.43

8

3359.7646

4429.5708

3504.8

4555.51

9

3299.9624

4325.6465

3392.14

4347.06

10

3257.5439

4172.5049

3306.78

4177.07

Abandonment

3225.7126

4172.5049

3306.78

4177.07

Time (Year)

RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS With the Eclipse keyword, DRSDT, equated to zero as in case 1, no gas liberation from the oil is taking place. Whereas in case 2, when it is commented out, gas is allowed to liberate out of the oil phase so that the reservoir fluid in this case is more volatile with varying dissolved gas-oil ratios and free gas concentrations. This difference in the amount of gas in the system is responsible for the trends in the production rates (FOPR), Bottom hole pressures (BHP) and produced gas oil ratios (Rp) as discussed below. Figures 1 and 2 confirms that production trends in both cases are similar, with both displaying constant FOPR of 2MSTB/D for the first 2.3years and constant Rp for the first 2 years and well BHPs assuming the same shapes. 3

However, the presence of more gas in the Case 2 creates additional reservoir energy resulting in a longer plateau production rate as shown in Figures 3 and higher production rate profile during the decline stage. As expected, higher gas in volume in the system causes an increase in the bottom hole pressure profiles for both the injector and producer for the second case than the first. A closer observation of Figures 5 and 6 reveals that the BHP profiles of the injector well is significantly higher than that of the producer well. Ranging from 7200psi at inception to an average of 4500psi for the injector well and 2400psi to 1000psi for the producer well. This is also analogous to the pressure behaviour of the injector and producer well blocks (BPR) presented in Table 1 above; as it can be observed, pressure values are higher in the injector well block than they are in the producer well block for both cases. These higher pressures in the trajectory of the injector well than those of the producer well is understandable and attributable to the setting of a high surface injection pressure. Of course, for gas to flow into the reservoir, the injection pressure must be set higher than the reservoir pressure as fluid flow in the direction of less pressure differential. Also, for the reservoir pressure must be higher than the BHP of the producer, for fluid to flow into the wellbore. Therefore, the pressure declines from the wellhead of the injector to the well head of the producer. In conclusion, resolution of gas in the oil decreases the rate of decline in the oil production rate, reduces the gas production with respect to oil (Rp) and reduces the rate of decline in the production and injection wells bottom hole pressures. Numerical dispersion effects are found in the GOR and FOPR estimates as a result of the coarse grid dimensions used.

PROBLEMS ENCOUNTERED AND SOLUTION APPROACH •

•

Unfamiliarity with the software keywords created time wastage in modification of the input file. The ECLIPSE manual, internet searches and literatures helped us through. Incorrect placement of parameters at the appropriate locations and wrong ECLIPSE keywords usage resulted in several error messages during the simulation runs. This was corrected by following the error prompts to relocate the parameters and with the help from our resourceful instructor

4

WORKSHOP NO. 2: PLACING WELLS IN THE SIMULATOR METHODOLOGY Prediction runs were made for three cases: (1) Two Vertical Wells: One Injector and One Producer; (2) Add a new vertical producer to the system and (3) Replace the second producer with a 2000-ft long horizontal well from centre of block 9,9,21; on a modified version of the same input data of Workshop 1 (SPE_1_ODEH

IMPES), no gas – resolution option. The modifications include refining the original 10 × 10 × 3 grid to a 20 × 20 × 30 grid system with the first 10 new vertical grid layers having the same properties as the original first layer and the injection well, still in block (1,1), now perforated in layers (1-10); The production well is also now in block (20,20), but is now perforated in layers (21-30).

RESULTS PRESENTATION Figures 7 through 9 show comparisons of oil rates, Produced Gas-Oil-Ratios and Well BHPs for the three cases described above while Figures 10 and 11 compares oil production rates and the GORs with those of the old grid system.

Figure 7: Comparison of Field Oil Production rates for Case 1(GREEN), Case 2 (BLUE) and Case 3 (PURPLE).

Figure 8: Comparison of Produced GORs for Case 1(RED), Case 2 (GREEN) and Case 3 (BLUE).

Figure 9: Comparison of BHPs for the injector: Case 1(GREEN), Case 2 (RED) and Case 3 (BLUE); and producer: Case 1(YELLOW), Case 2 (LIGHT BLUE) and Case 3 (PURPLE). 5

Figure 10: Comparison of Produced GOR of the old grid (PURPLE) and the three cases of the new grid: Case 1(RED), Case 2 (GREEN) and Case 3 (BLUE).

Figure 11: Comparison of Field Oil Production of the old grid (PURPLE) and the three cases of the new grid: Case 1(GREEN), Case 2 (BLUE) and Case 3 (RED).

RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS Figures 7 and 8 displays the FOPR and GOR for the three scenarios described above. It reveals that on the average, the use of two vertical wells produced the least production rates while the use of a horizontal well produced the highest average oil production rate over the simulation run time. Figures 10 and 11 compares the performances of the different grid systems. Figure 11 shows that the oil production rates are higher in the new, 20 X 20 X 30 grid for the three cases (GREEN, BLUE and RED) than the old, 10 X 10 X 3 grid system (PURPLE). As expected, an increase in oil flow rate indicates a reduction in the produced GOR as shown in Figure 10. This is because of the reduction in the relative permeability to gas as oil relative permeability rises with higher oil flow rates. This is why the GOR curve of the old grid system (Purple, Figure 10) is higher than those of the new grid system. The GOR curve of the old grid (Figure 10, purple) shows numerical dispersion effects after the 3rd year. Also, its FOPR curve shows the effect of numerical dispersion after the 3rd year of production. These were, however, not present in the curves for the new 20 X 20 X 20 grid for any of the cases or parameters. In conclusion, the effect of Numerical dispersion was reduced by the use of more refined grid system. Also, field productivity is usually increased by the use of horizontal wells were there is good vertical connectivity and in oil rim reservoirs.

PROBLEMS ENCOUNTERED AND SOLUTIONS APPROACH •

When drilling more producers (vertical and horizontal), it was observed that the producers were connected together as one producer. We were able to resolve this by giving both producers distinctly different names and effecting the appropriate number of wells per group and total number of wells 6

•

The sizes of each layer in the z direction, ∆z, had to be adjusted alongside that of ∆x and ∆y, to maintain the thickness of the reservoir.

WORKSHOP NO. 3: GRIDDING METHODOLOGY PART I • A five-spot water flood situation was considered and modification of the grid density •

performed to run the simulations for four different grid refinements. The permeability values of the edge blocks were cut in half to account for the outside portion of the model which does not exist. Also, the porosity of the edge blocks also cut in half

•

•

to maintain the total pore volume of the model. The following grid refinements were used; 5X5X6, 8X8X6, 10X10X6, 20X20X6. Plot of the Cumulative oil produced, production rates, Cumulative water injected and Water cut were made as a function of time and presented below.

PART 2 The simulations described in Part I was repeated with sizes of the edge blocks fixed at 40m X 40m for all four grid refinements. For the 4th case, the 18X18X6 model was run. Results are presented below.

PART 1 RESULTS PRESENTATION Figure 12 through 14 shows the results obtained for the four grid refinements mentioned above.

Figure 12: Well oil production rates versus time

Figure 13: Cumulative oil production versus time

7

Figure 14: Well water cut versus time

PART 2 RESULTS PRESENTATION

Figure 15: Well oil production rates versus time (Pt 2)

Figure 16: Cumulative oil production versus time (Pt 2)

Figure 17: Well water cut versus time (Pt 2)

RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS Figure 12 shows that increasing the grid refinement increases the plateau production period or time to commencement of decline. The 10 X 10 X 6 grid system has the longest plateau production period of 3 years. In Figure 13, Cumulative oil production was constant from year 0 to 2 for all grid sizes. From year 2 to 4, the 20X20X6 grid showed the highest cumulative oil production and can thus be concluded that cumulative oil production increases with increasing the grid refinement size. But a change in the production profile causing the 5x5x6 grid to have the highest and 20X20X6 the lowest could be as a result of a sealing fault being encountered. Year 7 and above showed the 20X20X6 having the highest and 5X5X6 having the lowest cumulative production. Figure

14

reveals

that

increasing

the

grid

refinement

delays

water

breakthrough time. This means that the larger the grid size, the longer it takes for water to breakthrough.

8

MAJOR OBSERVATIONS AND CONCLUSION From Figures 4.2 and 4.6 (the plot of cumulative oil production for Part 1 and Part 2), the cumulative oil production for both the 20X20X6 and the 18X18X6 grid models was approximately 4.7MMSTB. Similarly, well water cuts for the 20X20X6 and the 18X18X6 grid models are approximately 0.68STB/D (Figures 14 and 17). In conclusion, from the figures 12 to 17, the 20X20X6 and the 18X18X6 grid models show similar behaviour. Hence, 20X20X6 is the optimum number of grid blocks since the solutions are not changing. YOU DO NOT NEED TO RUN THE 20X20X6 case here. Why? There was no need to run the 20X20X6 case because the 18X18X6 model showed similar behaviour to it. Both have a breakthrough time of approximately 6 years. Hence, running the 20X20X6 model would result in waste of computation time and resources.

9

WORKSHOP NO. 4: IMPES VS. IMPLICIT FORMULATIONS METHODOLOGY (BRIEF) The SUMMARY sections of the given input data files were modified to provide estimations of Cumulative oil recovery,

cumulative water injected, Gas-oil ratios, Water cut and Wells Block Pressures for both the Fully IMPLICIT and IMPES formulations.

RESULTS PRESENTATION Figures 18 &19 below compares the results from the Full IMPLICIT formulation and the IMPES formulation.

Figure 18: Plots of production parameters for both the FULL IMPLICIT formulation: Cum. Production (Blue), Cum. Water injected (Purple), GOR (Yellow), Block Pressure (Light Blue); and the IMPES formulation: Cum. Production (Green), Cum. Water injected (Black, Dotted), GOR (Purple, Dotted), Block Pressure (Red)

Figure 19: Plots of Well Block Pressures for both the FULL IMPLICIT formulation: Injector (RED), Producer (Green); and the IMPES formulation: Injector (BLUE), Producer (PURPLE)

RESULTS ANALYSIS, DISCUSSION AND OBSERVATIONS From the graphs, FULLY IMPLICIT formulation can be said to be more stable than IMPES formulations. The plot for the cumulative water injected shows that the amount of water injected ...