Summary. This paper presents a case study of four high-pressure gas wells that were perforated overbalanced in heavyweight mud and then evaluated with pressure-transient tests. The surprising result of these tests was the minimal amount of completion damage. The average skin damage calculated from the test data was 2.6, which is much lower than the 12.6 predicted from data available in the literature. The information from this case study gives the completion engineer additional data on which to base decisions when designing perforation operations in high-pressure gas wells. This case study also raises the possibility that completion damage from overbalanced perforating practices decreases as reservoir pressure increases. Introduction An engineer designing a high-pressure gas-well completion has many difficult decisions concerning the selection of the most economical perforating method. To make this selection, the engineer must consider the cost of each method, the probability of mechanical failure, and the resulting well productivity. The methods available in order of preference are shooting the well underbalanced with a casing gun, shooting underbalanced with a through-tubing gun, shooting overbalanced with a casing gun in a clean fluid, or, as a last resort, shooting overbalanced in mud. It is relatively easy to determine the cost of each method and to determine the potential risks of each method; however, it is very difficult to predict each method's influence on the well's productivity. The importance of selecting the correct method of perforating a high-pressure well can be illustrated by a review of the available options. Perforating the well underbalanced with a tubing-conveyed gun is the best option in terms of achieving the maximum productivity from the well. The advantages of this option are deep perforations, with 90' phasing, and the cleaning effect from underbalanced perforations. The major drawback to this system is the reliability problems of running tubing-conveyed guns in deep, hot, high-pressure wells. On one of Conoco's offshore platforms, 50% of the tubing-conveyed perforating guns failed to fire. The next option is perforating a well underbalanced with a through-tubing gun. Although the benefits of underbalanced perforating are obtained with through-tubing guns, the small guns often do not adequately penetrate the casing and formation. We have had several wells that would not flow when perforated with through-tubing guns but then flowed at rates in excess of 7 MMscf/D [200 × 103 std m3/d] when perforated overbalanced with casing guns. The last option, perforating the well overbalanced with mud in the hole, is the most reliable method from a mechanical standpoint and is the least expensive option. Casing guns will penetrate the high-strength casing strings, and the guns can be retrieved to ensure that all the charges fired, The disadvantage of overbalanced perforating is the potential for completion damage. Perforating in a clean brine will reduce the completion damage, but weighted brines cost several hundred dollars per barrel, require special safety precautions, and may cause severe corrosion problems. McLeod and Locke presented analytical methods to predict the productivity of the different perforating methods. The problem with any analytical method is the large number of assumptions required to complete the calculations. The critical assumptions that must be made are the reservoir permeability, kR; the number of effective perforations, np; and the perforation efficiency-length, Lp, and radius, rp. Fig. 1 diagrams perforation geometry. Typical values for these variables have been presented in the literature; however, the data are generally based on low-pressure wells. Little work has been done on determining whether these variables change when the perforating method remains constant and the formation pressure is increased. Additionally, the permeability of these overpressured sands is usually much lower than the permeability encountered in low-pressure sands. This may also be an important factor. Without these data, the engineer can only assume that these variables remain constant at all formation pressures. Pressure-Transient Data Pressure-transient data from four high-pressure gas-well completions indicate that the variables affecting perforation productivity change as the formation pressure increases. The four completions were all perforated overbalanced in heavyweight water-based drilling muds, yet had an average skin value of only 2.6. The average skin value is much lower than the value that would have been predicted in the literature. Table 1 summarizes the actual calculated vs. predicted skin values for these four wells. These data seem to indicate that the completion damage caused by overbalanced perforating is reduced as the formation pressure increases. Table 2 summarizes the pressure-transient results and completion data for these four wells. The only well that had a significant skin value was Well C, which was initially drilled in Feb. 1982 and was plugged and abandoned. The well was then re-entered and completed in June 1985. The majority of the skin is believed to be a result of formation damage that occurred while the well was plugged. The well-test data support this idea and are discussed further in the Appendix. During completion operations, Wells A and B were subjected to significant surge pressure after perforating. After the perforation runs, the tubing string was run with a downhole shut-in valve located above the packer. This valve was run in the closed position, and the tubing was filled with water. When the valve was opened, the well was subjected to a surge pressure equivalent to the reservoir pressure minus the hydrostatic head of water. The skin factor for these two wells proved to be less than those for Wells C and D, but the difference is not significant. Although this would be the preferred method of bringing a well on line, it is not very practical when a permanent completion is run. Despite the overbalanced perforating techniques, the initial skin values in the four wells indicated that good-quality completions were obtained. The completion quality can be determined by a comparison of the calculated value of the crushed-zone permeability, kdp, with values published in the literature. Locke reported that laboratory tests indicated that the crushed-zone permeability should be about 20% of the native reservoir permeability. McLeod indicated that the crushed-zone permeability should fall within 10 to 25% of the native permeability. Both authors believed that these values were for wells perforated under optimum conditions and that the crushed-zone permeability would be reduced further if a well were perforated overbalanced in mud. Laboratory and field data have indicated that the crushed-zone permeability would be as low as 2.5% of the native permeability when a well is perforated overbalanced in mud. The crushed-zone permeability for the four wells tested was determined in the following manner. SPEPE P. 33^
Summary Successful fracture treatments over large intervals are often difficult to obtain. In the Laredo Lobo gas field in Webb and Zapata Counties, TX, successful fracture treatments over large intervals are critical for economic success of the field. Conoco Inc. uses a pre fracture injectivity test along with temperature and gamma ray logs to ensure that the entire completed interval will be treated during the fracture job. If the entire completed interval is not being treated, the logs act as a road map to determine where additional perforations are required. Introduction Post fracture evaluation logs run in the Laredo Lobo field show that often only a portion of the completed interval was treated. Successful fracture jobs are difficult because the wells are completed over intervals as large as 61 m (200 ft) that often contain several individual pay sands. A typical log section is shown in Fig. 1. Successful fracture treatments over large intervals are critical for the future of Laredo field, as wells completed over small, easy-to-stimulate intervals are often uneconomical. Many techniques have been used to treat these large intervals. In the early life of the field, staged treatments with ball sealers were used widely. Currently, the use of ball sealers has been abandoned because post fracture evaluation logs have proved them ineffective. Staged treatments with temporary sand, gel, or mechanical plugs, although effective, are often impractical because of high pressures and small shale intervals that separate the pay sands. The most effective treatments involve the limited-entry technique to divert the fracture fluid across the entire interval. However, the success of the limited entry treatments has been hampered by poor perforating efficiencies and low fracture rates. The low fracture rates in the Laredo field, which are caused by tubing and pressure restrictions, require that a minimum perforation pressure drop be used in the limited-entry fracture design. To distribute the fracture fluid accurately across the completed interval with a minimum perforation pressure drop, the fracture gradients of the individual zones in the completed interval must be known. In the Laredo field, the fracture gradients of the individual zones cannot be determined until the well is fractured. If the fracture gradients of the individual zones are assumed incorrectly, the fracture job will not treat the entire completed interval. This paper proposes that a pre fracture injectivity test be run to check assumptions made in the design of a limited-entry fracture treatment. Temperature and gamma ray logs run in conjunction with the test are used to determine whether the entire completed interval is being treated and to find any mechanical problems with the well before an expensive fracture job is pumped. Theory Most fracture jobs designed to treat large intervals fail because the fracture fluid is not distributed properly over the entire completed interval. When a particular zone in the completed interval does not receive enough fluid, it screens out early in the fracture job because the necessary fracture width is not created. The limited-entry technique has been the most effective means to distribute the fracture fluid across the completed interval. In limited-entry treatments, a minimum pressure drop of 2100 kPa (300 psi) is maintained across the perforations to force fluid into all the perforations. Many limited-entry treatments fail because the fracture design includes the assumption that a single fracture will be created across the entire completed interval. In practice, several independent fractures will be created across the completed interval when sufficient barriers separate the individual pay zones. JPT P. 995^
Often in a mature oil field the question is asked: where is the remaining oil in place, ROIP, and how much oil is left? However, recently the reservoir and production engineers have added the third question to the quest for additional oil recovery. The third question deals with the uncertainty in the distribution and volume of the remaining oil in place in an under-developed part of the field. The integrated work of geologists, production engineers and computer geoscientists can lead to developing a VOlumetric model which utilizes formation properties, stratigraphy, and fluid surveillance data to determine the ROIP volume, as well as the associated uncertainty with regard to the ROIP. This presentation gives an overall view of the process which produces reservoir grids of original and current fluid in place volumes, and uncertainty factors to be utilized by reservoir and production engineers in targeting future development in the field. Also, these products can be used to enhance and optimize fluid surveillance exercises in the under-developed regions of the field. DESCRIPTIONThe reservoir engineers have difficult time utilizing the popUlar tools such as reservoir simulators to answer the above questions. One of the reasons for this is the fact that most reservoirs are enriched with geological complexity that effect fluid movement, not easily simulated in a numerical model. Another complexity arises from the multiple drive mechanisms currently interfacing with each other in a large reservoir, making Reference and illustrations at end of paper 747 the setting of the boundary conditions a handsome task. For the reasons mentioned above, an alternative approach is taken that utilizes the existing field data to predict the ROIP, with a quantified uncertainty. This approach is based on time-lapse volumetric model, which uses the fluid surveillance data gathered in the field by various means. The fluid surveillance data can be interpreted from open hole logs, cased hole logs, production history, and even interpretation of the regional production trends. METHODOLOGYThe production engineers, with the help of geologists can utilize the existing open and cased hole logs, production history, or other types of surveillance tools in wells, to determine the vertical distribution of the fluid in those wells. Depending on the type of surveillance tools, the degree of confidence varies from well to well. Even with in a single well, the degree of confidence changes from one depth to another, depending on the surveillance tool that was used to identify the fluid type in the formation. For wells with no surveillance logs or data, an interpretation is made by development geologists who relies on the offset wells data. ObViously, the degree of confidence with this type of interpretation is lower than if the well had recent surveillance logs that showed its recent fluid distribution. After the geologists and engineers have interpreted all the wells in the field, and posted uncertainty ranks to each interpretation, a field wide exercise ...
Often in a mature oil field the question is asked: where is the remaining oil in place, ROIP, and how much oil is left? However, recently the reservoir and production engineers have added the third question to the quest for additional oil recovery. The third question deals with the uncertainty in the distribution and volume of the remaining oil in place in an under-developed part of the field. The integrated work of geologists, production engineers and computer geoscientists can lead to developing a VOlumetric model which utilizes formation properties, stratigraphy, and fluid surveillance data to determine the ROIP volume, as well as the associated uncertainty with regard to the ROIP. This presentation gives an overall view of the process which produces reservoir grids of original and current fluid in place volumes, and uncertainty factors to be utilized by reservoir and production engineers in targeting future development in the field. Also, these products can be used to enhance and optimize fluid surveillance exercises in the under-developed regions of the field. DESCRIPTIONThe reservoir engineers have difficult time utilizing the popUlar tools such as reservoir simulators to answer the above questions. One of the reasons for this is the fact that most reservoirs are enriched with geological complexity that effect fluid movement, not easily simulated in a numerical model. Another complexity arises from the multiple drive mechanisms currently interfacing with each other in a large reservoir, making Reference and illustrations at end of paper 747 the setting of the boundary conditions a handsome task. For the reasons mentioned above, an alternative approach is taken that utilizes the existing field data to predict the ROIP, with a quantified uncertainty. This approach is based on time-lapse volumetric model, which uses the fluid surveillance data gathered in the field by various means. The fluid surveillance data can be interpreted from open hole logs, cased hole logs, production history, and even interpretation of the regional production trends. METHODOLOGYThe production engineers, with the help of geologists can utilize the existing open and cased hole logs, production history, or other types of surveillance tools in wells, to determine the vertical distribution of the fluid in those wells. Depending on the type of surveillance tools, the degree of confidence varies from well to well. Even with in a single well, the degree of confidence changes from one depth to another, depending on the surveillance tool that was used to identify the fluid type in the formation. For wells with no surveillance logs or data, an interpretation is made by development geologists who relies on the offset wells data. ObViously, the degree of confidence with this type of interpretation is lower than if the well had recent surveillance logs that showed its recent fluid distribution. After the geologists and engineers have interpreted all the wells in the field, and posted uncertainty ranks to each interpretation, a field wide exercise ...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.