SPE 91789 Variation of Fracture Opening with Pressure -- Key to Well Productivity in Naturally Fractured Reservoirs

Y. Du, SPE, New Mexico Institute of Mining and Technology, and L. Guan, SPE, Texas A&M University

Copyright 2004, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2004 SPE International Petroleum Conference in Mexico held in Puebla, Mexico, 89 November 2004. This paper was selected for presentation by an SPE Program Committee following review of information contained in a proposal submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract Naturally fractured reservoirs usually developed many scales of fractures, for example, the super-high-permeability flow in big fractures, mid-permeability flow in small fractures and very-low-permeability flow in matrix rock. This results in the subsurface multi-scale flow during the reservoir development. This multi-scale flow makes it tougher to analyze the well productivity than that of a conventional reservoir. In this paper, muti-rate well-testing data were used to calculate the production capacity at different reservoir pressures in naturally fractured reservoirs. The method is exemplified with two wells in a case reservoir. The calculation results are very similar to field production data. We conclude that oil production rate in these kind of reservoirs is not only impacted by both twophase flow and non-Darcy flow as in conventional reservoirs, but also impacted by the variations of fracture opening. This conclusion can be used to better define the optimal rate for the reservoir under different reservoir pressures, thus further improve our forecast accuracy and overall field development results. Introduction Although many works have been done and published on productivities of artificially fractured well, the literature is limited on the wells in naturally fractured reservoirs. Many authors [1-3] (McGuire, W.J. and Siora, V.J, Prats, M. and Levine, J.S; van Poollen, H.K., Tkrsley, J.M. and Saunders, C.D) have investigated the effect of fracture length and conductivity on post-fracture well productivity for vertically fractured reservoirs assuming pseudo steady-state or steadystate flow in the reservoir and fracture height equal to the reservoir thickness. Later, work by Tinsley[4] et al investigated the relationship between fracture height and well productivity for cases where the fracture height was equal to or less than the reservoir thickness. All of these works employed two-

dimensional mathematical or physical analog models to predict post thickness fracture well productivity. However, the literatures are limited for naturally fracture wells. Due to the decline of reservoir pressure, the fracture opening in the reservoir, especially near the wellbore area, will decrease, which will result in the loss of permeability near wellbore. This permeability loss will impact the well productivity and should be taken into careful consideration when designing the field development. In this paper, muti-rate (flow after flow) well-testing data were used to calculate the production capacity at different reservoir pressures in naturally fractured reservoirs. The method is exemplified with two wells in a case reservoir (named T reservoir). According to Yan, X[5], the case reservoir is a naturally fractured reservoir. Permeability and porosity of its rock matrix are very low and the rock matrix does not serve as reservoir rock. There developed many scales of fractures, for example, the super-high-permeability flow in big fractures, mid-permeability flow in small fractures and very-lowpermeability flow in matrix rock. The fractures are the major flow media. Subsurface oil is under saturated, low GOR, and relative high-density black oil. The pressure difference between initial reservoir pressure and bubble point pressure is higher than 20 MPa. Phenomenon for naturally fractured reservoir well Multi-rate well testing data indicate that production rate is not proportional to pressure drawdown near wellbore. Even in some small pressure range, the well productivity becomes smaller while the pressure draw downs increase. After a through analysis of this reservoir, the following reasons were put forward to account for this particular phenomenon [6]. (1). The bottom hole pressure is lower than the bubble point pressure. The gas starts to come out from oil in reservoir condition. Two-phase flow exists in reservoir. The relative permeability of oil decreases and oil viscosity becomes larger, which reduces the oil mobility. (2). NonDarcy flow near wellbore is caused by the potential high production rate. (3). Mechanical reasons such as particles movement due to the high-speed flow in the porous, which reduces the reservoir permeability. (4). The opening of the fracture becomes smaller. Due to the pressure drawdown near well bore, the pressure difference between overburden pressure and porous pressure increases, which decreases the opening of fracture and lower the fracture permeability. In the

www.petroman.ir

SPE 91789

following parts, we will dispute some of the reasons and indicate that the opening of the fracture becomes smaller is one of the main reasons for this unique phenomenon. The case carbonate T reservoir has a low bubble point pressure and flowing bottom hole pressure near well bore are much higher than the bubble point pressure of the oil. Thus the possibility of reason one is very small. Our experience on carbonate reservoir production indicates that for the naturally fractured reservoir, during the early production period, the fracture opening near the wellbore is large. With the increase of production drawdown near wellbore, the fracture opening near the wellbore will close somewhat. This fracture close will affect the permeability near the wellbore, which results in the decrease of well productivity. This happens in most of the world carbonate reservoir with fractures. Table 1 lists the well productivity variation with reservoir pressure draw down in Renqiu Field, a carbonate field in north China. It shows the well productivity decreases when the reservoir pressure drawdown increases.
Table 1 Well productivity (Jo) and reservoir pressure draw down (DPr) from Renqiu Field, China [7]

than field measurements, while that for 8 mm choke is larger than field measurements. When we use 8 mm choke data to predict oil rate at 5, 6 and 7 mm choke, all the prediction for 5 mm, 6mm and 7 mm choke are smaller than field measurements. This indicates that some other effects exist except two-phase flow.
Figure 1. Forecasting results by Vogel equation
200 Choke=5 mm 180 160 140 Qo (m 3 /d) 120 100 80 60 40 Choke=6 mm Choke=7 mm Choke=8 mm Well test measurements

Well Name Ren 4 Ren 5 Ren 26 Ren 31

DPr (MPa) 0.1 0.6 0.7 0.9

Jo (t/(d.Mpa)) 316 126 26923 9542

DPr (MPa) 0.8 1.2 1.2 1.5

Jo (t/(d.Mpa)) 228 71 16200 5453

37

41

45 Pw f (MPa)

49

53

In this paper, we used the flow-after-flow well testing data to analyze the well productivity decrease due to the pressure drawdown resulting from decreasing fracture opening.
Table 2 Flowing pressure and calculated oil rate by Vogel method (well T47)
Pwf Mpa 51.61 48.02 43.22 38.42 Choke mm 5 6 7 8 AOF 3 m /d 351.1 335.4 295.2 255.3 Cal 1 3 m /d * 73 106.6 148.2 186 Cal 2 3 m /d 70 * 101.8 141.5 177.6 Cal 3 3 m /d 61.4 89.6 * 124.6 156.4 Cal 4 3 m /d 53.1 77.5 107.7 * 135.2 Measured 3 m /d 73 101.8 124.6 135.2

Note: * indicates the data when forecasts were made using Vogel equation. For a given choke size, its corresponding Pwf and Qo was obtained from the test data. Using these data, Vogel equation was then used to predict oil rate at other choke sizes. Table 2 shows the flowing pressure and calculated oil rate by Vogel method using well T47 test data. Figure 1 plots all the results. When we use 5 mm choke data to predict oil rate at 6, 7 and 8 mm choke, all the prediction are larger than field measurements. When we use 6 mm choke data to predict oil rate at 5, 7 and 8 mm choke, the prediction for 5 mm choke is smaller than field measurements, while the predictions for 7 and 8 mm choke are larger than field measurements. When we use 7 mm choke data to predict oil rate at 5, 6 and 8 mm choke, the prediction for 5 mm and 6 mm choke are smaller

For the four tested choke sizes, the Vogel equation prediction results using 6 mm choke have a smaller error referencing the other 3 choke measurements and predictions. It may indicate that the fracture closing pressure is close to the wellbore flowing pressure at 6mm choke conditions. In fact, this guesstimation is verified by the discussion later. (2). Non-Darcy flow near wellbore because of the high production rate. The Reynolds number of the flow near this well was estimated by assuming that all the production fluids are produced from fractures. The flow cross-area is estimated using the fracture size (Cross section area of the fracture opening) from core description and drilling break data in the drilling report. The Reynolds number near wellbore of this well is between 1700~2100. According to Bakers experiments, the lowest Reynolds number for turbulent flow should be greater than 4000~ 8000 in radial inward flow [8]. In addition, according to flow-after-flow test in naturally fractured reservoir in Iran, the additional pressure difference caused by turbulent flow is only several to tens Psia (0.01~0.2 MPa)[9]. However, the pressure difference here is several MPas. The turbulent flow effects, therefore, should not be so significant. Lab core analysis indicates that the fracture is the main flow avenue in this reservoir. The reservoir rock is not sensitive to flow speed. In addition, the flow speed near the wellbore is not very high and the possibility of particle movements may not be super-high. In summary, one of the main reasons resulting in productivity decrease is the opening decrease of the fracture near wellbore. We realized that the variation of fracture opening with pressure is the key to well productivity in this naturally fractured reservoir. In naturally fractured reservoirs,

www.petroman.ir

SPE 91789

some natural fractures near wellbore are open under the initial condition. The somewhat closure of the fracture will lead to the decrease of permeability near wellbore and further the decrease of productivity. This closure of the fracture is controlled by two parameters: fracture closure pressure and fracture closure degree. The fracture closure degree indicates the degree of permeability loss due to the variation of fracture opening. Case studies In this paper, we used well production rates and production pressure drawdown near wellbore relationship drawn from muti-rate well test to analyze the fracture variation near wellbore and further use this fracture property to predict well production rate at different reservoir pressure (well IPR). Here two sets of well data were taken from T-reservoir as examples. For each example, the muti-rate well test measurements were first matched by trying different combinations of fracture closure pressures and fracture closure degrees. After this matching, the fracture parameters such as fracture closure degree and closure pressure were obtained, which was then used to predict IPRs. Determination of the closure parameters Under Steady State flow, if we assume there exists an area with radius of Rfr near wellbore, in this area the fractures are closed and formation permeability is Kc. Whereas, in other area away from wellbore farther than Rfr, the reservoir permeability is still original permeability Ki. If the reservoir pressure is Pe and wellbore flowing pressure is Pwf, then the production rate:

T401 matching When n=1, Pst=54.0 MPa and Jo=46 m3/(d.MPa). Assume Dko=0.9, 0.95 and 0.98, according to Eq. 1, we calculated the oil rate at different flowing bottom pressures. They are plotted in figure 3. When Dko=0.95, the calculated results matches the well testing measurements very well. Thus we get the Dko=0.95 and fracture closure pressure 54.0 MPa for well T401. In other words, when the reservoir pressure is lower than 54 MPa, the productivity of this well will decrease dramatically.

Measured

Qo (m3/d) Pwf (MPa)

Figure 2. T47 IPR matching results

Q = JO[1 Dko( Pwf < Pst < Pe

P st P wf P e P wf

Eq.1 )n ](P ......... e P wf ).........

Q Oil production rate (m3/d), Jo Initial oil productivity (m3/(d.MPa), Dko Fracture closure degree (fraction), Pwf Well-bore flowing pressure (MPa), Pst Pressure at which fracture close (MPa), n Index of fracture closure (generally 1.0 ~ 2.0), Kc Permeability of the part reservoir where fracture closed (md), Ki Initial reservoir permeability (md), Pe Current reservoir pressure (MPa), Rfr Radius of the part reservoir where fracture closed. T47 matching When n=1, Pst=47.0 MPa and Jo=8.4 m3/(d.MPa). Assume Dko=0.3, 0.4 and 0.5, according to Eq. 1, we calculated the oil rate at different flowing bottom pressures. They are plotted in Figure 2. When Dko=0.4, the calculated results matches the well testing measurements very well. Thus we get the Dko=0.4 and fracture closure pressure 47.0 MPa for well T47. In other words, when the reservoir pressure is lower than 47 MPa, the productivity of this well will be only 60 percent of the original value.

Measured

From the above calculation, the fracture closure degree of T47 is 0.4, while that from T401 is 0.95. The fracture close degrees of the two wells are quite different. This difference agrees with different fracture condition of the two wells. This is because of the dimension/size of fractures near the wellbore is different. The fracture near well T47 is large, while that of

www.petroman.ir

Qo (m3/d) Pwf (MPa)

Figure 3. T401 IPR matching results

SPE 91789

T401 is small. (Width of 0.1~0.3 mm and length of 10~20 cm.). For the well with small fractures, the minor variation of the fracture width will cause relative significant permeability loss. However, for T47 well, the fracture near well bore is large (0.5m~3.0m) and rich. This characteristic causes the non-significant permeability variation. This indicates that in different location in the reservoir, the closure degree may be different.

Well IPR forecasting Based on the above matching, we obtained the parameters of fracture closure such as fracture closure pressure and degree and n. These parameters were then used to predict future IPR under different reservoir pressures. Figure 6 shows the predict results of T47 well and Figure 7 shows that of T401 well. These predictions were used to determine the optimum production rates and reasonable pressure drawdowns near wellbore of the wells in natural fractured reservoirs. Later actual production rates of these two wells show this method is effective.
Pr=60 MPa Pr=58 MPa Pr=56 MPa Pr=54 MPa

Qo (m 3/d)

Figure 4. Schematic show of the fractures near T401 wellbore

Pwf (MPa)
Figure 6. T47 IPR forecasting results

Pr=59 MPa Pr=55 MPa Pr=51 MPa Pr=47 MPa

Qo (m3/d)

Figure 5. Schematic show of the fractures near T47 wellbore

Figure 4 is the schematic show of the fractures near T401 wellbore in T reservoir. As shown by the arrows, the dark channels indicate the fractures. It shows that the subsurface fractures near well T401 are relative small. However, as shown by Figure 5, the subsurface fractures near well T47 are relative big. The arrow points to the big fracture.

Pwf (MPa)
Figure 7. T401 IPR forecasting results

www.petroman.ir

SPE 91789

Conclusion As most of the naturally fractured reservoirs in the world, the variation of fracture opening is one of the main reasons for the productivity decrease with the increasing of pressure draw down. This paper describes an efficient approach to predict the future IPR by flow-after-flow test data. Production and pressure drawdown data from the flow-after-flow test can be used to analyze the fracture closure characteristics. This method has been successfully applied to the T reservoir. The fracture closure pressure calculated from this method is between bubble point pressure and static reservoir pressure of T reservoir. This is consistent with rule of thumb.

Fracture closure is a non-repeatable process. Figure 8 shows the permeability variation when the reservoir pressure first decreases then increases back to original pressure [7]. Clearly, for all the samples, the reservoir permeability is lower than its original value when the reservoir pressure first decreases and then increases back to the original pressure value. The difference indicates the non-cycle process. The lower the original permeability, the larger the difference. Therefore, in order to obtain a longer production plateau period, reasonable pressure drawdown should be determined in the early period of development. Nomenclature Dko Fracture closure degree (fraction) Jo Initial Oil productivity (m3/(d.MPa) Kc Permeability of the part reservoir where fracture closed Ki Initial reservoir permeability n Index of fracture closure (generally 1.0 ~ 2.0), Pe Current reservoir pressure Pwf Well-bore flowing pressure Pst Pressure at which fracture close Q oil production rate Rfr Radius of the part reservoir where fracture closed. References
1. McGuire, W.J. and Siora, V.J: The Effect of Vertical Fractures on Well Productivity, J. Pet. Tech. , Oct. 1960, P72. 2. Prats, M. and Levine, J.S: Effect of Vertical Fractures on Reservoir Behavior Results on Oil and Gas Flow, SPE 593, 1963 SPE Rocky Mountain Joint Meeting, Denver, May 23-24, 1963. 3. Van Poollen, H.K., Tkrsley, J.M. and Saunders, C.D: Hydraulic Fracturing Fracture Flow Capacity vs Well Productivity, SPE 890-G, 32nd Annual Fall Meeting of SPE, Dallas, October 6-9, 1957. 4. Tinsley, J. M., J. R. Williams Jr., R. L. Tiner, and W. T. Malone: Vertical Fracture Height Its Effect on Steady-State Production Increase, JPT, May 1969, pp 633-638. 5. Yan X.: Reservoir Characterization of T Field, Not published, 1998. 6. Du, Y. et al: Development Plan of T Field, Not published, 2001. 7. Bai, S.: Development of Carbonate Buried Hill Oil Reservoirs, Chinese Petroleum Industry Press, Beijing, China, 1996 8. Zhao, S.: Carbonate Oil Reservoir in Renqiu, China, Chinese Petroleum Industry Press, Beijing China, 1997 9. Saidi, A.M. et al: Reservoir Engineering in Fractured Reservoir, Institute of E&P, NorthChina Petroleum Field, 1993.

Figure 8. Permeability variation with the reservoir pressure decrease then increase

This fracture closure degrees of these two wells indicate that the closure degree maybe different at different reservoir locations. This is because of the dimension of fractures near the wellbore is different. For example, the fracture closure degree of T47 is 0.4 while that from T401 is 0.95. This maybe because that the fracture near well T47 is large, while that of T401 is small. (Width 0.1~0.3 mm and length 10~20 cm.). For the well with small fractures, the minor variation of the fracture width will cause relative significant permeability loss. However, for T47 well, the fracture near well bore is much larger (0.5m~3.0m), richer and longer. This characteristic causes the non-significant permeability variation of T47 well in this reservoir.

www.petroman.ir