Copyright 2006, IADC/SPE Drilling Conference
This paper was prepared for presentation at the IADC/SPE Drilling Conference held in Miami, Florida, U.S.A., 21–23 February 2006.
This paper was selected for presentation by an IADC/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 International Association of Drilling Contractors or 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 IADC, SPE, their officers, or members. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International Association of Drilling Contractors and Society of Petroleum Engineers 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, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 1.972.952.9435.
Abstract
Multilateral well control is an issue in today’s petroleum industry that has not had sufficient study to date. Many operators are drilling multilateral wells, with limited knowledge of the pressure characteristics in the multiple wellbores as well as how the pressures in the individual laterals are related. If a kick occurs in a multilateral well, the operator must be able to determine which lateral or laterals the kick is in, as well as the effect that circulation of the kick can have on the pressures in the other laterals.
This paper reports the results obtained in a study that has been conducted utilizing a Multilateral Well Control Simulator developed at Seoul National University and Texas A&M University.1-3 This simulator calculates and records the pressure profiles in all laterals, on a continuous, basis during tripping, drilling, and well kill operations. From these simulations, we can determine the effect that the changing annulus pressure profile can have on all of the laterals.
Depending on the relative pressure gradients of the different laterals with respect to each other, kicks can actually be induced in one lateral while circulating gas from another and as the gas enters the mother wellbore. For safe and efficient well control, the operator must have knowledge of the formation pressures in each lateral so that steps can be taken to prevent the changing annulus pressure from inducing additional kicks in the other laterals.
The industry now has a study and tool which can help operators drilling multi-lateral wells to better understand how pressure changes in one lateral effect the others. This tool will allow a better understanding of the pressure changes in multiple lateral sections that will occur during well kill operations.
Introduction
The goal in conventional kick circulation is to maintain the annulus pressure at the bit (assuming the bit is located at the bottom of the hole) equal to or greater than formation
pressure. This is to prevent additional influx of formation fluid into the wellbore. This is done my manipulating the drilling choke so that the sum of the choke pressure, annulus hydrostatic pressure and annulus frictional pressure are at least equal to formation pressure. This can be expressed mathematically by:
ann hsp ann f choke bhp form p p p p p ,,++==.........................Eq. 1
where,
Pressure
c Hydrostati Annulus Pressure Friction Annulus Pressure
Choke Pressure Hole Bottom Pressure Formation ,,=====ann hsp ann f choke bhp form p p p p p
Since the annulus hydrostatic pressure changes during the kill operation (due to gas expansion and changing hole geometry) the circulating drillpipe pressure is monitored to assure that the proper bottom hole pressure is maintained, and can be expressed mathematically by Eq. 2. p SPP = p bhp - Δp bit - p f,dp - p hsp,dp ……………..………..Eq. 2 where,
Pressure
c Hydrostati g Drillstrin Pressure Friction g Drillstrin Drop
Pressure Bit Pressure
Bottomhole Pressure Standpipe ,,===Δ==dp hsp bit f bit bhp SPP p p p p p
When circulating the old mud density the circulating drillpipe pressure is equal to the system pressure loss plus the stabilized shut in drillpipe pressure. This circulating pressure is commonly referred to as the Initial Circulating Pressure. As kill mud is circulated to the bit the frictional pressure in the drillstring,
the hydrostatic pressure in the drillstring and the bit pressure drop all increase resulting in a decrease in circulating drillpipe pressure. This Final Circulating Pressure is calculated by multiplying the kill rate pressure by the ratio of the kill mud density to the old mud density.
Since the kill rate pressure is the pressure loss in the entire circulating system, the measurement includes the annulus
friction pressure which results in the bottom hole pressure
IADC/SPE 99028
Analysis of Gas Kicks in Multi l ateral Wells Utilizing Computer Simulation
J. Schubert, T exas A&M U.; J. Choe, Seoul Nat l. U.; D. Dre h er, Halliburton; and H. Juvkam-Wold, T exas A&M U.
during a kill operation equal to the formation pressure plus the annulus frictional pressure – a built in safety factor.
Multilateral wells can complicate well control operations in a number of ways such as swabbing during trips to kill operations. Just how well control is affected by multiple open laterals has not been fully studied or understood. Methodology
The authors utilized a multilateral well control simulator which was originally developed by Choe and Juvkam-Wold as a conventional PC based well control simulator, and has been subsequently modified to model surge and swab during tripping operations with capabilities to simulate simple single vertical wellbores to complex multi-lateral wellbores. 1-3 In order to simulate the multilateral drilling sequence it was required to make some assumptions to input into the simulator. The reservoir was assumed to be homogenous so that the wellbore conditions in all the laterals would be the same. The permeability is 250 md, the porosity is 0.25, and the skin factor is 2.5 The mud density used in the simulator was of 10 ppg. The size of the kick has a significant impact in the pressure response of a multilateral kick. With this in mind, a kick size of 10 bbl and 50 bbl.
For this study, the authors modeled well control for a multi-lateral well with four laterals open to the mai
n wellbore. Fig. 1 depicts the wellbore that was modeled. Lateral 1 is defined as the lateral where the drillstring is located and has a vertical depth of 9,000’. Lateral 2 is at 10,000’ vertical depth, Lateral 3 is an opposing lateral to Lateral 1 at 9000’ vertical depth, and Lateral 4 is located at 8000’ vertical depth. The pressures were measured at the heel of all the laterals and at the bit.
swabbingWhen comparing pore pressure gradients of Laterals 2, 3, and 4 to Lateral 1, three situations were studied: all pore pressure gradients were equal, pore pressure gradients of Laterals 2, 3, and 4 are less than Lateral 1, and Laterals, 2, 3, and 4 had pore pressure gradients greater than Lateral 1. The pore pressure gradient in Lateral 1 was kept at 10,0 ppg while the pore pressure gradients in the remaining laterals were modeled at 9.0 ppg, 10.0 ppg, and 11.0 ppg.
Results
Case I, Base Case, Single Lateral
For Case I, we modeled a single lateral horizontal wellbore with a pore pressure of 10.0 ppg. The kick was taken with mud density of 9.5 ppg. The choke and drillpipe pressures are shown in Fig. 2. The major difference in this case and a kick for a well that has no horizontal section is in the initial shut-in conditions. When all the formation fluids remain in the horizontal section, the shut-in casing pressure
and shut in drillpipe pressure will be equal since there has been no reduction in annulus hydrostatic pressure from the low density formation fluids. Fig. 2 shows the rapid increase in choke pressure as the formation fluid is pumped out of the horizontal section where the vertical height of the kick increases rapidly.
The pressure decline schedule, on the other hand, should be modified so that final circulating pressure is reached when kill mud is circulated to the end of build in the horizontal section. If a straight line decline is used where final circulating pressure is attained when kill mud reaches the bit, excess pressure will be exerted on the annulus.
One potentially dangerous situation could occur with a kick in a horizontal wellbore must be discussed. If the circulation rate is not sufficient to remove all the gas from the horizontal section, a considerable amount of gas could enter the vertical section of the wellbore when normal circulation and drillstring rotation is restored. If this is not anticipated, a severe kick could occur as the gas reduces the annulus hydrostatic pressure. The operator cannot assume that one full circulation will remove all the kick fluids from the well. After the well is apparently dead, a comparison of the mud volume in the pit must be made between the post and pre kick levels. The volume of barite, water, and any other additives must be taken into account at this time also. In short, the operator must set the kill
rate such that all gas will be efficiently removed from the wellbore. Efficient kill rate is beyond the scope of this paper but research in this area has been conducted at Louisiana State University,4,5 Texas A&M University, 3,6,7 and others. 8,9
Case II, 4 Laterals, 2, 3, and 4 Pore Pressure Gradients Equal to Lateral 1
We will start our discussion of multilateral well control with the case where the pore pressure gradients in all the laterals are equal. In our study this was assumed to be a 10.0 ppg equivalent. For this case, we swabbed a kick while tripping out of Lateral 1 and shut the well in with a pit gain of both 10 bbl, Fig. 3, and 50 bbl, Fig. 4. Figures 3 and 4 show the pressure that is maintained in all laterals as compared to the formation pressure in each lateral. In this example the annulus pressure at the heel of Lateral 1 was maintained slightly greater than the formation pressure of Lateral 1. Fig. 1 shows the positions of all the laterals with respect to Lateral 1. The drillstring is always in Lateral 1 in this paper.
Lateral 1 and 3 are opposing laterals and located at a vertical depth of 9,000’ and a pore pressure of 4680 psi. Lateral 2 is the deepest lateral in the well with a vertical depth of 10,000’ and a pore pressure of 5200 psi. Lateral 4 is located at a vertical depth of 8,000’ with a pore pressure of 4160 psi.
The simulator was set to operate automatically with the pressure at the heel of Lateral 1 at 4714 psi –
slightly greater than formation pressure. In both Fig. 3 and 4 there is an increase in pressure in each lateral beginning at approximately 500 pump strokes. This increase then subsequent decrease in pressure represents when the gas bubble passing the Lateral 4. As can be seen from these plots, the pressure at all other times during the kill remains constant, slightly greater than the formation pressure in each respective lateral. This is what would be expected as long as the circulating drillpipe pressure is maintained properly as one would when utilizing a standard kill sheet. Looking back at Fig. 2, the choke pressure increases to offset the decrease in annulus hydrostatic pressure as the bubble exits the horizontal section as well as when it expands as it nears the surface.
A problem would occur, however if gas from the lower lateral exits the horizontal section into the inclined or vertical wellbore. When this occurs, it is similar to circulating a kick
off bottom. The standard kill sheet is designed to maintain the annulus pressure at the bit at a constant pressure.
The authors have already stated that with a horizontal wellbore, the pressure at the heel should be maintained constant at a pressure at least equal to the formation pressure in the lateral. For this case, the authors maintained the pressure at the heel of Lateral 1 at 4714 psi. If we add the hydrostatic pres
sure of 1000 feet of 10.0 ppg mud (520 psi) to the pressure maintained at the heel of Lateral 1, we get a pressure at the heel of Lateral 2 of 5234 psi, 34 psi greater than the formation pressure in Lateral 2.
Now if we assume gas has exited Lateral 2 below Lateral 1 so that the average density of the mud/gas mixture below Lateral 1 is equivalent to 9.0 ppg, the hydrostatic pressure in the bottom 1000’ of the well would be 468 psi. By adding 468 to 4714 psi, we get 5182 psi, 18 psi lower than the formation pressure in Lateral 2. This could result in additional influx, and a continued decrease in hydrostatic pressure below Lateral 1.
The obvious solution to this problem would be to temporarily plug all laterals except the one that is being drilled or worked on. Since this is not economically feasible, another solution must be utilized to determine if all the gas has, in fact, been removed from the wellbore, where this gas may be located, and how to systematically remove all gas from all the laterals.
Is there gas and where is it located?
We will discuss now how to determine if there is any gas remaining and where it may be. We will start this section assuming that a kick has been taken while drilling (as opposed to tripping pipe) in Lateral 1. Assume that the kick is taken at the bit in Lateral 1 (e.g. a higher pressure pocket is encountered).
Upon detection of the kick, shut-in the well, measure the stabilized shut-in drillpipe pressure shut-in casing pressure and pit gain. If the shut-in drillpipe and casing pressure are equal, the gas is still in the lateral.
Kill the well conventionally, as in he base case, shut-in the well and check to determine if there is any remaining pressure. If none, do not assume all the gas has been circulated from the well. Compare the post kick mud volume in the pit to the pre-kick pit volume plus the volume of any weight material, water and other additives added to the mud during the kill. If the comparison shows that there is any remaining gas, circulate at drilling rate while rotating the drillstring to assure that all the gas is removed from Lateral 1.
If there is no drillpipe pressure after the first circulation, but there is remaining casing pressure, there is likely to be gas remaining in the non-horizontal portion of the well. Circulate bottoms up again to make sure that there is no gas near the surface and shut-in again. If there still remaining pressure there is likely to be gas below Lateral 1. If this the case, raise the mud weight enough in Lateral 1 to provide a sufficient trip margin to prevent swabbing in a kick. Carefully POH from Lateral 1 (stripping through the closed annular preventer when there is casing pressure), then strip into the bottom of Lateral 2. Circulate kill mud as when killing Lateral 1. Again check for remaining pressure or gas in the well.
Now we will discuss the solution if a kick is swabbed. When swabbing is detected, stop the trip, check for flow. If no flow, carefully trip back to bottom, and prepare to circulate the kick from the well as one would for a swabbed kick in a vertical wellbore or single lateral well. After circulating bottoms up through the choke at kill rate, shut-in the well and check for any remaining pressure. Again, if there is any casing pressure remaining, there is likely kick fluid that is either in the vertical, build, or inclined section of the wellbore. Circulate bottoms up from Lateral 1 again. If there is still remaining pressure, the gas is either in the build section of Lateral 2, 3, or 4, or in the vertical portion of the wellbore below Lateral 1.
If this is the case, carefully strip out of Lateral 1, strip into Lateral 2, and circulate Lateral 2 clean, shut-in the well, and check for pressure again. If there is no casing pressure, check mud volumes as before to determine if there may still be gas in Lateral 3 or 4. It is recommended now to trip out of Lateral 2, into Lateral 3 and circulate bottoms up, then repeat for Lateral 4. Only when the operator is sure that all laterals are under control with no gas or other formation fluids in the wellbore, should the operator continue the operation that was being performed before the kick was taken.
Case III, 4 Laterals, 2, 3, and 4 Pore Pressure Gradients Less Than Lateral 1
In the case where the pore pressure gradients of Lateral 2, 3, and 4 are less than Lateral 1, inflow should not occur in these lower pressure laterals during kick circulation or due to swab effects during trips. This is not say that an influx in these laterals cannot occur, just that it is not as likely and the same steps to insure that formation fluid from all laterals be removed, as described above in Case II, prior to continuing the operations that were being conducted before the kick occurred Case IV, 4 Laterals, 2, 3, and 4 Pore Pressure Gradients Greater Than Lateral 1
In the case where the pore pressure gradient in Lateral 1 is less than the others, kicks should not occur in lateral one if the mud density is sufficient to control the highest pore pressure gradient in the well. Again, this is not to say that kicks will not occur in any or all of the laterals, the operator should not assume that if all the gas from a single lateral is removed that there is not any gas present in any of the other laterals. Steps should be taken to assure that all gas from all laterals is removed prior to continued operations.
Conclusions
From this study, there are four conclusions to be made
1.When a kick is taken in a multilateral well, it cannot
be assumed that the kick only occurred in the lateral
where the drillstring is located especially if the well
was swabbed in.
2.When circulating a kick from a lateral other than the
deepest, conventional circulation pressures are
enough to control the pressures in all laterals unless
there is gas present between the lateral containing
the drillstring and the deeper lateral.
3.Prior to returning to the operation that was being
conducted prior to the kick in a multilateral well, the
operator must take a systematic approach to
determine that there is no gas remaining in the well.
4.If gas has entered any of the laterals, this gas may
not be removed from the well with the first
circulation and can migrate into the main wellbore,
reducing the annulus hydrostatic pressure and
inducing another kick.
Acknowledgements
The authors would like to thank the U.S. Minerals Management Service and the Offshore Technology Research Center at Texas A&M University for funding this project. Nomenclature
p bhp = Bottomhole Pressure
p choke = Choke Pressure
p f,ann = Annulus Friction Pressure
p f,dp = Drillstring Frictional Pressure
p form = Formation Pressure
p hsp,ann = Annulus Hydrostatic Pressure
p hsp,dp = Drillstring Hydrostatic Pressure
p SPP = Standpipe Pressure
Δp bit = Bit Pressure Drop
References
1.Gjorv, B.: “Well Control Procedures for Extended Reach
Wells,” MS thesis, Texas A&M U., College Station, Texas
(2003).
2.Rommetveit, R., Bjorkevoll, K.S., Bach, G.F., Aas, B., Hy-
Billiot, J. et al.: “Full Scale Kick Experiments in Horizontal
Wells,” paper SPE 30525 presented at the 1995 SPE
Annual Technical Conference & Exhibition, Dallas, 22-25
October
3.Lage, A.C.V.M., Rommetveit, R., and Time, R.W.: “An
Experimental and Theoretical Study of Two-Phase Flow in
Horizontal or Slightly Deviated Fully Eccentric Annuli,”
paper SPE 62793 presented at the 2000 IADC/SPE Asia
Pacific Drilling Technology Conference, Kuala Lumpur,
Malaysia, 11- 13 September.
4.Johnson, A.B. and Cooper, S.: “Gas Migration Velocities
During Gas Kicks in Deviated Wells,” paper SPE 26331
presented at the 1993 SPE Annual Technical Conference
and Exhibition, Houston, 3-6 October.
5.Bendiksen, K., Maines, D., Moe, R., and Nuland, S.: “The
Dynamic Two-Fluid Model OLGA: Theory and
Application,” SPEPE (May 1991) 171.
6.Baca, H.E.: “Counter Current and Co-Current Gas Kick
Migration in High Angle Wells,” MS thesis, Louisiana
State U., Lafayette, Louisiana (1999).
7.Ustun, F.: “The Effect of High Liquid Flow Rates on Co-
Current and Counter-Current Gas Kick Migration in High
Angle Wells,” MS thesis, Louisiana State U., Lafayette,
Louisiana (2000).
8.Choe, J. : “Dynamic Well Control Simulation Models for
Water-Based Muds and Their Computer Applications,”
PhD dissertation, Texas A&M University, College Station,
Texas (1995).
9.Choe, J. and Juvkam-Wold, H.C.: “A Modified Two-Phase
Well-Control Model and Its Computer Applications as a
Training and Educational Tool,” SPECA (February 1997)
10.Choe, J., Schubert, J.J., and Juvkam-Wold, H.C., “Well
Control Analysis on Extended Reach and Multilateral
Trajectories,” SPE Paper No 97465, SPEDC, Volume 20,
Number 2, pp 101-108 (June, 2005).
11.Long, M.M, “Kick Circulation Analysis for Extended-
Reach and Horizontal Wells,” MS thesis, Texas A&M U.,
College Station, Texas (2004).
Figures
Fig 2 Drillpipe and Casing Pressure for 10 and 50 bbl kick in a single lateral horizontal wellbore
Fig. 1 Simulated multilateral well diagram used to simulate laterals at higher, lower and horizontally opposite configurations from the main wellbore.
10 bbl kick
500
1000
15002000
2500
3000
Pump Strokes
P r e s s u r e , p s i
Fig. 3 Pressures in the laterals during kick circulation for a 10 bbl kick
50 bbl kick
500
1000
1500
2000
2500
3000
3500
4000
4500
Pump Strokes
P r e s s u r e , p s i
Fig. 4 Pressures in the laterals during kick circulation for a 50 bbl kick
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论