IPTC-12303-MS-P Application of Horizontal Wells with Multiple Hydraulic Fractures_图文
IPTC 12303 Multi-Stage Fracturing Stimulations Improve Well Performance in Tight Oil Reservoirs of the Changqing Oilfield
X. Li, SPE, H. Wei, B. Chen, SPE, PetroChina Changqing Oilfield Co., X. Liu, SPE, Pinnacle Technologies, Inc., C. Wang, and X. Zhao, PetroChina Changqing Oilfield Co.
Copyright 2008, International Petroleum Technology Conference This paper was prepared for presentation at the International Petroleum Technology Conference held in Kuala Lumpur, Malaysia, 3–5 December 2008. This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at IPTC are subject to publication review by Sponsor Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International Petroleum Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-952-9435.
Abstract The oil reservoirs in the Changqing Oilfield consist of stacked sand/shale deposits, and the majority of the target zones have a permeability of 0.05-0.3mD. All oil wells completed in tight reservoirs from the Changqing Oilfield require fracture stimulation to achieve commercial production and to improve well productivity. Fracture stimulation is also very common for water injection wells to enhance injectivity. Multiple payzones in the region were either stimulated with standalone treatments that were isolated and pumped separately or with single-stage treatments to simultaneously treat several intervals using a kind of limited-entry technique. However, with the continuing development in these fields, larger pay intervals (gross pay thickness up to 75 meters) with poorer reservoir quality were encountered. Post-fracture production results indicated that the above stimulation techniques were not always effective in these tight reservoirs. As a result, a new staging technique was evolved and employed to stimulate these tighter reservoirs with better economic effectiveness. With this technique, two or three pay intervals were first perforated and dual-stage fracture treatments were then performed: one stage at a time without using any mechanical isolation. The idea was two-fold: a) using the stress increase around the fracture created by the first injection to divert the subsequent treatment to un-stimulated interval(s), and b) using the proppant banks created from the first propped treatment to minimize the proppant settling to the bottom portion of the fracture created by the second propped treatment. Three-month average post-treatment production data indicated that the dual-stage treatments outperformed conventional single-stage treatments in some reservoir environments. Candidate selection was a key to the success of this technique, and reservoir conditions that favor these types of treatments are as follows: multiple pays, lower permeability, lower net pressure, and less stress contrast between different payzones. Radioactive tracer logging and microseismic fracture mapping were conducted to evaluate fracture height growth. This paper presents a case history of fracture performance evaluations. Reservoir data, fracture modeling and post-fracture production data were compiled and employed to demonstrate the benefits of this fracture treatment technique in some tight oil reservoirs. Reservoir Overview The Changqing Oilfield is made up by many individual reservoirs that stretch across the giant Ordos Basin. As shown in Figure 1, the Ordos basin is located in north central China. The oil payzones in these reservoirs can be found at depths from 500 to 2,200 meters. The case studies here involved dual-stage fracturing stimulations in two oil reservoirs in the Changqing Oilfield, denoted as Reservoir S and Reservoir Y in this paper. The maps for Reservoirs S and Y are shown in Figures 2 and Figure 3, respectively. Both of the two reservoirs of interest consist of stacked sand/shale deposits and exhibit similar reservoir properties. A typical well log from one of the reservoirs is shown in Figure 4. Oil is produced from the same target formation in the two reservoirs. The target formation often consists of several pay intervals in these two reservoirs. Depending on well locations, the pay intervals in the target formation are buried at depths between 1,900 and 2,000 meters, with net pay up to 50 meters and gross pay up to 75 meters. These reservoirs are very tight with porosity values between 11.5 – 13.5%, oil saturation values between 48.5 – 51.5%, and effective permeabilities ranging typically from 0.05 to 0.3 mD. The reservoirs are also very heterogeneous with reservoir quality varying significantly from zone to zone and from location to location. Due to the nature of high water saturation and the availability of mobile formation water in these reservoirs, water is usually produced along with oil production. However, there are no water-oil contact lines because there are no water zones or aquifers below the oil reservoirs. Crude oil in these reservoirs has an average specific gravity of 0.85
(or an API gravity of 35) with an average solution gas to oil ratio of 78. The reservoir energy comes mainly from the expansion of solution gas dissolved in the crude oil. Table 1 summarizes the reservoir and fluid properties for these two reservoirs.
Figure 1: the location of the Ordos basin in China
Figure 2: the map of Reservoir S
Figure 3: the map of Reservoir Y
Figure 4: a typical well log from the reservoirs of interest
Table 1: Reservoir and Fluid Properties Porosity (%) Water Saturation (%) Effective Permeability (mD) Pore Pressure (MPa) Temperature (?C) 11.5 – 13.5 48.5 – 51.5 0.05 – 0.3 14.0 – 14.9 62.0 – 65.0 Oil Specific Gravity at surface Oil Viscosity (cp) at reservoir Oil Volume Factor Reservoir Compressibility Solution Gas to Oil Ratio 0.85 0.93 1.33 1.9E-3 78.0
Stimulation History Hydraulic fracturing is the key to achieving economic success for many low-permeability oil and gas reservoirs. Multi-stage hydraulic fracturing is most often used to stimulate multiple pay intervals. This study involves a case history of fracturing stimulations of multiple pay intervals in two tight oil reservoirs in the Changqing Oilfield. Multiple payzones in the area were traditionally stimulated using either separate-stage treatments with each stage isolated and pumped separately or single-stage treatments to simultaneously treat several intervals. However, with the continuing development in the area, reservoirs with larger pay intervals (net pay thickness up to 50 meters) and poorer reservoir quality were encountered. Post-fracture production data for wells stimulated with the above two stimulation techniques were compiled and evaluated. Table 2 shows the comparison of post-treatment production data for a number of wells stimulated by the separate-stage and single-stage treatments in Reservoir S. All the wells selected in Reservoir S had similar reservoir properties, but 6 wells were stimulated with the separate-stage fracture treatments while 15 wells were stimulated with the single-stage fracture treatments. The production rates in 3 months after the treatments indicated that wells stimulated with either of the above two techniques
performed similarly: the average liquid (oil and water) production rate from the wells stimulated with the separate-stage treatments was slightly better while the oil production rate was slightly worse than that with the single-stage treatments. Table 3 shows the comparison of post-treatment production responses for a number of wells selected in Reservoir Y. The three-month post-treatment production data for Reservoir Y indicated that wells stimulated with the single-stage treatments performed slightly better than those stimulated with the separate-stage treatments. Note that most oil wells in these reservoirs also produced formation water. It is obvious that more time and costs were required to perform the standalone, separate-stage treatments than the single-stage treatments. Since the separate-stage treatments could not provide any better results than the single-stage treatments in these reservoirs, the continuing practice of this technique is placed in question. Toward this end, further investigations were carried out to seek for a more effective stimulation method for these reservoirs.
Table 2: Three-month production comparison for Reservoir S Stimulation No. of Liquid Prod Oil Prod Water Cut 3 3 Type Wells Rate (m /d) Rate (m /d) (%) Separate-Stage 6 4.01 2.97 26.0 Single-Stage 15 3.91 3.18 18.6 Table 3: Three-month production comparison for Reservoir Y Stimulation No. of Liquid Prod Oil Prod Water Cut 3 3 (%) Type Wells Rate (m /d) Rate (m /d) Separate-Stage 11 4.80 4.15 13.6 Single-Stage 4 5.33 4.52 15.2
Understanding fracture growth behavior is the first step in evaluating zonal coverage and fracture performance. Radioactive tracing was used as a diagnostic tool to evaluate fracture growth from a single-stage treatment for a well in Reservoir Y. In this example, the tracers were embedded with some proppant samples that were pumped along with the regular proppant stages during the treatment. Radioactive tracing and spectral gamma ray logging have been employed for many years1 to determine the extent of vertical fracture height growth associated with a propped fracturing treatment or an acid fracturing treatment. Based on log interpretation results, the target pay was located at the upper portion of a sand body. The entire sand body was buried at depths from 2099.1 to 2138.0 meters. A perforation interval of 7.8 meters at the top section of the target zone with better reservoir quality was selected, and it was located at depths from 2099.1 to 2106.9 meters. The propped treatment consisted of 65 tons of proppant and 140 m3 of fluids, and the treatment was pumped at an average rate of 2.2 m3/min. As shown in Figure 5, the results from the radioactive tracing showed a propped fracture height of 9.0 meters only, located at depths from 2099 to 2108 meters. Based on the radioactive tracing results, the propped fracture did not cover the entire pay interval. The propped fracture height obtained from the radioactive tracing was basically located around the same depth where the perforation interval was located. The perforation interval for this case was located at the upper portion of the sand body. Based on the formation lithology shown in Figure 5, it would be easier for the fracture to initiate from the perforations and then to grow downwards within the sand body than to grow upwards into the shale barrier above the sand body. In very low-permeability reservoirs like this one, it usually takes hours after shut in for the fracture to close on the proppant. During that period, the proppant inside the fracture gradually settles down at the bottom portion of the fracture as the fracture fluid degrades. Conventional wisdom suggests that most of the proppant settles down at the bottom portion of the hydraulic fracture. Therefore, the radioactive tracing results in this case might not be conclusive about the true nature of the fracture height growth.
Figure 5: radioactive tracing results in a well from Reservoir Y showing propped fracture height
Dual-Stage Fracturing Although the radioactive tracing results in the previous section might not be conclusive, there was definitely a possibility that the propped fracture from a single-stage treatment with one perforation interval could not cover the entire pay interval(s). Due to the limitation of equipment horsepower for fracturing operations in the area, the lack of higher pumping rates make single-stage treatments to stimulate payzones over several perforation intervals ineffective. The practice of the separate-stage treatments to stimulate one payzone interval at a time is much more expensive and time consuming. A more effective stimulation method is thus required to treat reservoirs with larger pays and poorer reservoir quality. It is clear that treating several perforation intervals while eliminating zonal isolation between stages can improve zonal coverage and make more economic sense. With any type of multi-stage fracturing treatment, the reservoir pressure and stress around the fractures created from the previous stages are always elevated due to both the leaking off of fracturing fluids into the reservoir and the closing of the fractures on proppants. The elevated stress fields created by the previous stages will affect the initiation and propagation of fractures from later stages. This is often referred to as stress shadow effect. The stress shadow effect on multistage fracturing treatments of horizontal wells has been discussed in literature2. In the Mounds Drill Cuttings Injection project3, stress changes due to cyclic injections over the same interval led to the creation of new fractures with a different fracture orientation in each subsequent injection. As a result of searching for a more effective technique to stimulate these tight reservoirs, a new staging technique was evolved and employed. With this technique, two or three pay intervals were first perforated and dual-stage fracture treatments were then performed one stage at a time without using any mechanical isolation. The idea was two-fold: a) using the stress increase around the fracture created by the first injection to divert the subsequent treatment to un-stimulated interval(s), and b) using the proppant banks created from the first propped treatment to minimize the settling of proppants to the bottom portion of the fracture created by the second propped treatment. With the dual-stage fracturing technique employed in the Changqing Oilfield, the treatment size for the first stage is usually larger than that of the second stage. At the time of writing this paper, literature review indicated that a similar staging technique was used as early as 19964, and the authors in Reference 4 named it as the Induced Stress Diversion (ISD) technique. However, the IDS technique did not consider the proppant banking effects from the previous stage(s) as one of the mechanisms to divert the subsequent treatment. The dual-stage technique was put on trial in both Reservoirs Y and S. Post-fracture performance for wells in Reservoir Y was first evaluated: 5 wells were stimulated with the new dual-stage technique and 5 wells were stimulated with the traditional single-stage technique. Table 4 shows the reservoir properties and three-month average post-treatment production data for the selected wells in Reservoir Y. The production data indicated that the dual-stage treatments outperformed the single-stage treatments by 39%. Note that the average amount of proppants used for the dual-stage treatments was 27% more than that used for the single-stage treatments. Increased treatment sizes had some effects on the post-treatment performance. In order to evaluate the effect of the treatment sizes on the production performance, fracture modeling and post-treatment production analysis was carried for a typical well with the parameters from Reservoir Y. A lumped-parameter, 3-dimensional fracture growth model and a single-phase, 2-dimensional reservoir model were used for the analysis. Figure 6 shows the modeling results of treatment size versus the fracture half-length and 3-month cumulative oil production. Based on the modeling prediction in this figure, the treatment size increase from 77 to 98 tons (or an increase by 27%) only represents a production increase of 5%. This exercise demonstrated that the dual-stage treatments had brought additional production enhancement by 34%. Also, the average net pay thickness for the wells stimulated with the dual-stage treatments was 21% smaller. With the reservoir lithology exhibited by Reservoirs Y and S, it is difficult to accurately determine net pay thickness from the entire sand bodies, which vary a lot from well to well. Although there are all kinds of reservoir property uncertainties, the dual-stage treatments apparently deliver better production results in Reservoir Y.
Table 4: production comparison between two types of stimulation treatments in Reservoir Y Well Name Y20-15 Y19-15 Y19-14 Y17-15 Y23-15 Stimulation Type Dual-Stage Dual-Stage Dual-Stage Dual-Stage Dual-Stage Net Pay (m) 26.9 17.7 22.6 16.1 23.8 Porosity (%) 11.1 10.8 10.8 11.6 11.0 Est Perm (mD) 0.06 0.10 0.04 0.09 0.08 Oil Sat (%) 59.4 48.1 43.9 52.6 47.6 Prop Total (ton) 112.5 106.0 81.5 97.8 93.0 Prod Rate (m /d) 10.48 7.48 6.99 7.55 5.24
Y21-13 Y19-17 Y19-18 Y17-11 Y23-16
Single-Stage Single-Stage Single-Stage Single-Stage Single-Stage
38.5 21.6 30.0 20.3 24.8
12.7 12.1 12.0 12.9 11.2
0.11 0.11 0.16 0.12 0.09
60.7 63.0 59.6 56.6 56.2
65.2 97.8 68.5 89.7 64.8
4.53 7.82 5.34 6.36 3.05
Figure 6: fracture modeling and production analysis for a typical well in Reservoir Y
Post-fracture performance with the dual-stage fracturing treatments in Reservoir S was also evaluated and compared: 5 wells from each type of stimulation methods were selected. The reservoir properties, such as net pay thickness, porosity, estimated permeability and oil saturation, for both groups of wells in Reservoir Y were very similar. Table 5 shows the reservoir properties and three-month post-treatment production data for the selected wells in Reservoir S. The production data indicated that the dual-stage treatments outperformed the single-stage treatments only by 14% at the expense of increasing treatment sizes (total proppants used) by 54%. Using the same modeling results of treatment size versus 3-month cumulative oil production predicted in Figure 6, the treatment size increase from 68.5 to 105.8 tons represents a production increase of 10%. This means that the dual-stage treatments alone had brought additional production enhancement by 4% only. Due to reservoir property uncertainties, the dual-stage treatments for the wells selected and tested in Reservoir Y have not provided any benefits. The effectiveness of the dual-stage treatments in Reservoir Y requires further investigations. Candidate selection is a key to the success of this technique. The following reservoir conditions seem to favor the new dualstage treatments: multiple pay intervals in larger sand bodies, lower permeability, lower net pressure, and less stress contrast between different payzones.
Table 5: production comparison between two types of stimulation treatments in Reservoir S Well Name S392-22 S393-21 S395-21 S394-20 S395-25 Stimulation Type Dual-stage Dual-stage Dual-stage Dual-stage Dual-stage Net Pay (m) 43.3 37.6 45.7 39.2 46.4 Porosity (%) 12.0 11.7 12.2 12.0 12.0 Est Perm (mD) 0.36 0.11 0.35 0.36 0.34 Oil Sat (%) 54.7 48.0 52.4 48.0 49.1 Prop Total (ton) 114.1 97.8 121.4 97.8 97.8 Prod Rate (m /d) 8.35 6.82 6.00 5.29 4.00
S391-23 S392-24 S393-23 S396-26 S397-19
Single-Stage Single-Stage Single-Stage Single-Stage Single-Stage
35.5 38.7 42.3 47.4 44.6
11.7 12.6 12.4 12.4 12.3
0.13 0.18 0.19 0.28 0.15
47.9 52.8 50.8 53.9 48.9
65.2 81.5 57.1 81.5 57.1
5.53 3.53 7.06 5.65 4.94
Example Case Study The Y23-15 well in Reservoir Y was performed with the dual-stage fracture treatment. As shown in Figure 7, the log for this well indicated a sand body of about 50 meters (or about 60 meters in measured depth) separated by three thin shale layers. In order to effectively stimulate the well, three perforation intervals were selected and their measured depths are as follows: 2,224-2,227 m, 2,236-2,239 m, and 2,248.5-2,251.5 m. After the three intervals were perforated, dual-stage fracture
treatments were then performed by one stage at a time without using any mechanical isolation. The first stage consisted of 145 m3 of fluid and 57 tons of proppant. About 15 minutes after the first stage was completed, flow back was initiated to force the fracture to close on proppants sooner. Once the first stage fracture was believed to be closed, the second stage treatment was started. The second-stage was smaller than the first one and consisted of 93 m3 of fluid and 36 tons of proppant. The treatment data for both of the two stages are shown in Figures 8 and 9, respectively. Additional treatment data are summarized in Table 6. Microseismic fracture mapping5 was employed to map the fractures from both of the two stages. Figures 10 and 11 show the side views of the microseismic mapping results for both of the stages. The dots in the plots are the mapped microseismic events. The rectangular boxes that cover the majority of the microseismic events in the plots represent the likely areas where the hydraulic fractures were located. The microseismic events outside the box are believed to be of less confidence and are thus circled out. The following results were obtained from the microseismic mapping of the first stage: a fracture height of 65m growing about 10 meters above the pay, and the fracture was slightly asymmetric with the left-wing half-length of 140 m and the right-wing half-length of 110 m. The mapping results for the second stage indicated a fracture height of 65m growing about 15 meters above the pay, and the fracture was slightly asymmetric with the left-wing half-length of 101 m and the right-wing half-length of 125 m. Fracture modeling analysis by using the technique of net pressure matching was carried out to understand the fracture growth behavior. Diagnostic injections were not performed because it was difficult to analyze any diagnostic injections with three perforation intervals. The results of fracture closure stress and complexity from a diagnostic injection in an offset well were used. The fracture geometries predicted by fracture modeling for both of the two stages are shown in Figure 12. The outer perimeter represents the fracture geometry for the first stage while the inner perimeter represents the fracture geometry for the second stage. The fracture modeling results are summarized in Table 6. The modeling results showed that the fracture height for the second stage was only slightly smaller than that for the first stage, but the fracture length for the second stage was much shorter than the first stage. The modeled fracture geometry (hydraulic fracture length and height) for the first stage was slightly smaller than the mapped fracture geometry. Both the fracture modeling and mapping results suggest that both of the two stage treatments had covered all the pay intervals across the entire sand body. Although the treatment size for the second stage is smaller than that for the first stage by about 50%, the instantaneous shut-in pressure (ISIP) for the second stage is larger than that for the first stage by 0.5 MPa. The ISIP increase in the second stage might be caused by the stress increase and/or proppant banking/bridging from the first stage. It is worth pointing out there is an offset well, Y23-16, which was stimulated by the single-stage treatment and was very close to Y23-15. The 3-month production results summarized in Table 4 indicated that Y23-15 outperformed Y23-16 significantly. The post-treatment production for Y23-15 was predicted using the fracture geometry and conductivity values obtained from the fracture modeling. As shown in Figure 13, the predicted and actual post-treatment production data compared favorably for a period of three months after the well was put into production. The well produced about 3.8 m3/day of oil at the end of the 3-month period. Note that this was a brand new well which had no production at all prior to the stimulation treatment.
Figure 8: treatment data of Stage 1 in Well Y23-15
Figure 7: the Y23-15 well log
Figure 9: treatment data of Stage 2 in Well Y23-15 Figure 10: the side view of microseisms mapped from Stage 1 in Well Y23-15
Figure 11: the side view of microseisms mapped from Stage 2 in Well Y23-15
Figure 12: model-predicted fracture and conductivity profiles of the dual-stage treatment for Well Y23-15
Table 6: fracture modeling input/results for Well Y23-15
Modeled Oil Rate (m?/day)
Measured Oil Rate (m?/day)
Figure 13: comparison of predicted and actual post-treatment production data for Well Y23-15
Treatment Summary 3 Clean Fluid Total (m ) 3 Pad (m ) Proppant Total (kg) 3 Max Prop Conc (kg/m ) 3 Average Rate (m /min) Surface ISIP (MPa) Modeling Input/Results Young’s modulus for sand/shale (MPa) Stress gradient for sand (MPa/m) Stress gradient for shale (MPa/m) Estimated reservoir permeability (mD) Net pressure at shut-in (MPa) Hydraulic fracture half-length (m) Propped fracture half-length (m) Hydraulic fracture height (m) Propped fracture height (m) Dimensionless fracture conductivity
Stage 1 146 28 57,000 750 2.6 14.3
Stage 2 96 20 36,000 725 2.6 14.8
1.5E+4 0.0143 0.0165 0.08 4.1 110 100 59 55 7.0 4.6 88 78 56 51 7.6
Conclusions 1. For some tight oil reservoirs with multiple pays and poor reservoir quality in the Changqing Oilfield, separate-stage fracturing treatments failed to provide additional production enhancement than the single-stage treatments. 2. In search for an affective stimulation technique, a new dual-stage fracturing method was evolved and employed to stimulate multiple pay zones in tight oil reservoirs. 3. The new dual-staging technique was put on trial in two reservoirs. The trial results indicated it was very effective for one reservoir and not effective for the other reservoir. 4. Both radioactive tracing and microseismic mapping were performed to evaluate the fracture height growth behavior for two different wells in the same reservoir. The radioactive tracing results showed a propped fracture height of 9 meters for one well stimulated with the single-stage technique and the microseismic mapping results indicated a hydraulic fracture height of 65 meters for the other well stimulated with the dual-stage technique. 5. Fracture and production modeling analysis was conducted and the effects of treatment size on post-treatment production performance were evaluated. Acknowledgements The authors would like to thank PetroChina Changqing Oilfield Co. for permission to publish this work. Thanks also go to Tongxiang Cui with Pinnacle Technologies for conducting some fracturing treatment analysis in this study. References 6. Fisher, M.K., Walker, R., Dunleavy, R., Woodroof, B., and Crabb, H.: “Radioactive Tracers Facilitate Stimulation Job Evaluation”, Petroleum Engineer International, February 1995. 7. Fisher, M.K., Heinze, J.R., Harris, C.D., Davidson, B.M., Wright, C.A. and Dunn, K.P.: “Optimizing Horizontal Completion Techniques in the Barnett Shale Using Microseismic Fracture Mapping”, paper SPE 90051 presented at the SPE Annual Technical Conference and Exhibition held in Houston, Texas, 26-29 September 2004. 8. Griffin, L.G., Wright, C.A., Davis, E.J., Wolhart, S.L. and Moschovidis, Z. A.: “Surface and Downhole Tiltmeter Fracture Mapping: An Effective Tool for Monitoring Drill Cuttings Disposal,” paper SPE 63032 presented at the 2000 SPE Annual Technical Conference, Dallas, Oct. 1-4. 9. Hewett, T.W. and Spence, C.J.: “Induced Stress Division: A Novel Approach to Fracturing Multiple Pay Sands of the NBU Field, Uintah Co., Utah”, paper SPE 39945 presented at the 1998 SPE Rock Mountain Regional/Low-Permeability Reservoirs Symposium and Exhibition held in Denver, CO, April 5-8. 10.Warpinski, N.R, Wolhart, S.L. and Wright, C.A., "Analysis and Prediction of Microseismicity Induced by Hydraulic Fracturing," SPE 71649, 2001 SPE Annual Technical Conference and Exhibition, New Orleans, LA, Sept 30-Oct 3.