Patent Grant
6957574
This invention generally relates to monitoring the structural integrity and stress on a conduit, and more particularly, to monitoring the structural integrity and stress on a well casing used in oil drilling operations.
Oil and gas production from petroleum reservoirs results in changes in the subsurface formation stress field. These changes, when large enough, can result in serious damage or even complete loss of the bore hole through major deformation of the well casing. Thus, it is desirable to monitor subsurface stress fields as they may indirectly indicate the stress experienced by a well casing during oil production. While monitoring subsurface stress fields may generally be useful in determining the stress, or strain, experienced by a well casing, direct detection of casing strain is expected to give a better understanding of the subsurface forces that lead to deformation of the well casing and would allow for more precise monitoring of well casing integrity. This will lead to development of both preventative operating measures, including early abandonment in advance of dangerous well conditions and casing deformation, as well as better casing design and improved well completion programs. Consequently, oil companies have expressed an interest in direct monitoring of strain in the casing during the life of the well.
Direct monitoring of strain on a well casing, however, is often problematic because well casing strain can be caused by a number of different stresses or modes, including tensile or compressive stresses imparted along the axis of the casing, and shear stresses imparted through twisting or forces perpendicular to the casing axis. Casing strain can occur over long stretches of casing or can be very localized, and therefore may go undetected. The high magnitudes of strain that can cause deformation of a well casing, and/or the harsh environment down hole, can also cause apparatuses traditionally used to monitor strain to cease functioning.
Methods and apparatuses currently used to monitor well casing strain do not provide a solution to problems associated with direct strain monitoring. Many prior art techniques for monitoring well casing strain involve the use conventional strain gauges or sensors of the kind that are only capable of measuring strain in one orientation or mode at any given time. Conventional strain gauges are also prone to malfunctioning and damage when subjected to the high strain levels of interest and to the harsh environment of oil wells, and may not allow for direct monitoring of casing strain. Accordingly, conventional well casing strain monitoring methods and apparatuses can fail to detect critical points of high strain in a well casing that can lead to casing deformation, or may not detect strain at isolated critical locations on a casing. Precise monitoring of well casing strain is therefore difficult with the use of conventional methods and apparatuses.
It is known in the prior art that fiber optic sensors can be useful for measuring various stresses and temperatures present in the down hole environment. In U.S. patent application Ser. No. 09/612,775, entitled “Method and Apparatus for Seismically Surveying an Earth Formation in Relation to a Borehole,” filed Jul. 10, 2000, which is incorporated herein by reference, a technique is disclosed for using fiber optic sensors to detect seismic events, and in one embodiment it is contemplated that such sensors can be coupled to the well casing to detect seismic emissions emanating from the surrounding earth strata. However, this configuration is not suited to measure casing strain per se, as it is configured and attached to firmly couple to the surrounding borehole. Accordingly, the sensors disclosed in that application will naturally pick up acoustics such as seismic signals present in the surrounding earth strata, reducing their ability to measure casing strains without interference.
Thus, there is a need for a monitoring system for detecting well casing strain that allows for detection of strain from any orientation or mode before excess casing deformation occurs, that allows for distributed strain sensing capability over very long lengths of a well casing, and that does not suffer from the foregoing shortcomings of the prior art. The present disclosure provides such a method and apparatus.
Improved methods and apparatuses for directly monitoring well casing strain and structural integrity are disclosed that allows for monitoring of potentially damaging strain from any orientation or mode and over long stretches of well casing. In a preferred embodiment, optical fiber sensors are housed within a housing and attached to the exterior surface of the casing. The sensors may be aligned parallel, perpendicular, or at an appropriate angle to the axis of the casing to detect axial, hoop, and shear stresses respectively. The sensors are preferably interferometrically interrogatable and are capable of measuring both static and dynamic strains such as those emitted from microfractures in the well casing. Analysis of microfracture-induced acoustics includes techniques for assessment of relatively high frequencies indicative of the presence of microfractures. Assessment of the timing of the arrival of such acoustics at various sensors deployed along the casing further allows for the location of strain to be pinpointed.
The foregoing and other features and aspects of the present disclosure will be best understood with reference to the following detailed description of specific embodiments of the invention, when read in conjunction with the accompanying drawings, wherein:
In the disclosure that follows, in the interest of clarity, not all features of an actual implementation of a well casing integrity monitoring system are described in this disclosure. It will of course be appreciated that in the development of any such actual implementation of the disclosed invention, as in any such project, numerous engineering and design decisions must be made to achieve the developers' specific goals, e.g., compliance with mechanical and business related constraints, which will vary from one implementation to another. While attention must necessarily be paid to proper engineering and design practices for the environment in question, it should be appreciated that development of a well casing integrity monitoring system would nevertheless be a routine undertaking for those of skill in the art given the details provided by this disclosure, even if such development efforts are complex and time-consuming.
The disclosed embodiments are useful in directly monitoring well casing strain, and particularly when then the strain reaches a level that can threaten the structural integrity of the well casing. The disclosed embodiments preferably use optical fiber sensors, which provide a large number of options for measuring the strain imposed on a well casing and which offers high reliability. Fiber optic sensors also have the additional benefit that they can be easily multiplexed along a single fiber optic cable (using time division multiplexing or wavelength division multiplexing as is well known) to allow for several sensors to be connected in series, or to be connected to other optical sensors that measure parameters other than casing strain. However, other types of strain-measuring sensors can be used if desired, such an electrical, piezoelectric, capacitive, accelerometers, etc.
It is believed that the magnitude of well casing strain of interest to detect is between about 0.01% and 10.0%, which is believed to equate to stresses ranging from about 3000 pounds per square inch (psi) to well above the yield strength for a standard steel casing. At a 10% axial strain (i.e., parallel to the casing axis), the casing would be expected to undergo significant plastic deformation and possible catastrophic failure. The disclosed fiber optic sensors, which are preferably made of optical fiber having a cladding diameter of from about 80 to 125 microns, can be subject to about 100,000 psi (i.e., 1% strain) along its length without serious risk of breaking, and hence will be able to detect high strains and potential problems up to at least the onset of plastic deformation of steel casings. Therefore, it is theorized that the disclosed fiber optic sensors can be used to detect strains in the casing of between 0.01% and 1.0%, which covers a large portion of the detectable range of interest, and possibly higher ranges when detecting shear stresses which are not aligned with the optical fiber.
The use of a housing 9 to protect the sensors outside of the casing constitutes a novel advance over the prior art disclosed in aforementioned incorporated U.S. patent application Ser. No. 09/612,775 and U.S. Pat. No. 6,435,030. The '030 patent does not disclose the use of a housing for sensors deployed on the casing. In the '775 application, fiber optic sensors attached to the casing are not confined within a rigid housing because the goal of that application is to acoustically couple the sensors to the subsurface formation to efficiently detect seismic events. However, in the present application, it is desirable to isolate the sensors from acoustics or stresses in the subsurface formation as much as possible so that the strains and acoustics in the casing are measured with minimal interference. The housing 9 helps to effectuate this goal. Sensor housing 9 is preferably welded to the exterior surface of well casing 1, and covers the entire length of optical fiber 8 that is attached to well casing 1. Sensor housing 9 is further preferably vacuumed or filled with an inert gas such as nitrogen to form an acoustically insulative gap between the housing and the sensors (which is helpful even though external borehole noise could to some extent couple through other portions of the casing 1 to the sensors). The housing 9 and cable 5 are preferably affixed to the casing before it is deployed down hole, and before application of the casing cement.
Optical fiber 8 could be a standard communications fiber, although environmental considerations may dictate the use of fibers that are for instance not sensitive to hydrogen which is often present in the well fluid. As will be explained in further detail, fiber 8 is preferably formed into or spliced to coils 7 which are each bounded by a pair of fiber Bragg gratings (FBGs) 6 to form the casing strain sensors. The use of FBGs in fiber optic sensors is well known in the art, and the reader is referred to U.S. Pat. Nos. 5,767,411, 5,892,860, 5,986,749, 6,072,567, 6,233,374, and 6,354,147, all of which are incorporated herein by reference, to better understand such applications. Each coil 7, when unwound, is preferably from approximately 10 to 100 meters in length. Coils 7 are preferably attached to the exterior surface of well casing 1 with the use of an epoxy or an adhesive film. More specifically, an epoxy film is first adhered to the exterior surface of well casing 1, and the coils 7 are placed on top of the epoxy film. The epoxy film may then be cured, or heated, to rigidly bond optical fiber to the exterior surface of well casing 1. When affixing the fiber to the casing, it may be preferably to place the fiber under some amount of tension. In this way, compression of the casing may be more easily detected by assessment of the relaxation of the tensile stress on the fiber 8.
In a preferred embodiment, sensor coils 7 are attached at more than one depth on the well casing 1 (see FIG. 5). In this regard, and as is well known, several sensor regions such as that depicted in
In the embodiment of
It is preferable that each coil 7 by bounded by a pair of FBGs 6, such that each coil's pair has a unique Bragg reflection wavelength. It is further preferable to isolate the FBGs 6 from casing strain, because without such isolation the reflection (Bragg) wavelength of the FBGs might excessively shift, which would make their detection difficult and hence compromise sensor function. In this regard, it can be useful to place an isolation pad between the FBGs 6 and the outside surface of the casing, similar to the method disclosed in U.S. Pat. No. 6,501,067, issued Dec. 31, 2002, and which is incorporated by reference in its entirety. When so configured, the coils may be multiplexed together using a wavelength division multiplexing approach. Alternatively, each coil 7 can be separated by a single FBG 6 (not shown), wherein each separating FBG has the same Bragg reflection wavelength in a time division multiplexing approach, such as is disclosed in U.S. Pat. No. 6,354,147. One skilled in the art will realize that the FBGs 6 can be fusion spliced to the coils 7 and to the fiber 8, which is preferable to reduce signal attenuation as it passes through the various coils. As the details of fusion splicing are well known, they are not repeated here. The length of the coils 7 along the axis of the casing can be easily changed, e.g., up to tens of meters, which allows for static strains along this length to be averaged, which might be suitable in some applications. If a very long strain length measurement is desire, it may not even be necessary to form a coil, and instead sensor 7 can constitute a straight line of fiber optic cable affixed to the exterior of the casing. However, care should be taken to adjust the length of the sensor, be it coiled or uncoiled, so that interferometric detection is possible if an interferometric interrogation scheme is used.
The coils 7 of
In an alternative embodiment, the FBGs 6 themselves, as opposed to the coils 7, may act as the sensors. In this embodiment (not shown), the FBGs 6 would themselves be attached to the casing at the position of the coils, and would be oriented parallel to the axis of the casing. Axial deformation of the casing will stretch or compress the FBGs 6, and the amount of deformation can be determined by assessing the shift in the Bragg reflection wavelength of the FBGs, as is well known. If such an alternative approach is used, it would be preferable that each FBG have a unique Bragg reflection wavelength to allow proper resolution of one FBG from another, i.e., in a wavelength division multiplexing approach. The FBGs 6 in this approach can be serpentined around the casing 1, in a manner similar to that disclosed in U.S. Pat. No. 6,354,147 in order to measure shear strain.
To measure all potential stress modes on the casing 1, one skilled in the art will note that a combination of axially (FIGS. 1-3), circumferentially, and angled sensors can be used, and can be housed within a common housing 9 to form an all-inclusive strain sensor station.
Although it is preferred to mount the sensors on the outside of the casing 1, the sensors will function equally well if they are mounted on the interior surface of the casing. Whether to mount the sensors on the interior or exterior surface of the casing 1 would be based on considerations such as the risk to the fiber optic cable during installation as well as the availability of a “wet connect,” which are well known, for the connecting internal sensors to the cable after completion of the casing.
The manner in which the disclosed sensors may be used to detect static strains in the casing is obvious from the foregoing descriptions. However, an additionally useful benefit comes from the ability of the disclosed sensors to detect dynamic strains in the casing, namely, those acoustics emitted from microfractures that occurs within the casing when it is placed under relatively high strains. Microfracture acoustics will generally be very sharp in duration and of relatively high frequency content, e.g., in the 10 kilohertz to 1 megahertz range. This allows such acoustics to be easily resolved when compared to other acoustics that are present downhole, such as acoustics present in the fluid being produced through the production pipe 2. These microfracture-based acoustics are likely to occur under all modes of casing loading, but with different characteristic signatures of amplitude, frequency content and rate of acoustic events. The relatively low energy release of these acoustic emissions preferably requires a strain sensor that is highly sensitive, such as the interferometric sensor arrangements disclosed above.
When detecting microfracture acoustics, axial orientation of coils 7 (
As mentioned earlier, acoustic emissions from metal structures, such as well casing 1, are distinct events that normally have a characteristic high frequency content of between about 10 kilohertz to 1 megahertz. This makes detection of these dynamic events relatively simple. First, monitoring of this frequency range would normally only be indicative of microfractures, and not other acoustics naturally present down hole. Second, that these relatively high frequency events are time limited in duration helps to further verify that microfractures in the casing are being detected. Third, as the acoustics emitted from the microfractures will travel along the casing 1, their origin can be pinpointed. These points are clarified in subsequent paragraphs.
The electronics in equipment 110 convert the reflected signals from the various sensors into data constructs indicative of the acoustic strain waves propagating in the casing and straining the sensors as a function of time, again as is well known, and this data is transferred to a signal analysis device 120. The signal analysis device 120 converts the strain data into a frequency spectrum, represented in FIG. 6. As one skilled in the art will understand, the frequency spectra of
Because the conversion of the strain induced acoustic data from the sensors into its constituent frequency components is well known to those in the signal processing arts, this conversion process is only briefly described. As is known, and assuming a suitably high optical pulse (sampling) rate, the reflected signals from the sensors in the sensor stations 100 will initially constitute data reflective of the acoustic strain waves presented to the sensor as a function of time. This acoustic strain wave versus time data is then transformed by the signal analysis device 120 to provide, for some sampled period, a spectrum of amplitude versus frequency, as is shown in FIG. 6. As is well known, this can be achieved through the use of a Fourier transform, although other transforms, and particularly those applicable to processing of discrete or digitized data constructs, may also be used. While the disclosed sensors can detect frequencies up to 1 MHz, and hence should be suitable to detect microfractures in the casing, one skilled in the art will recognize that suitably short sampling periods may be necessary to resolve a particular frequency range of interest. If necessary, the signal analysis device 120 could contain a high pass filter to filter out lower frequencies not of particular interest to the detection of microfracture acoustics.
Further confirmation of the detection of microfracture-induced acoustic emissions is possible due to the fact that such noise will travel with relatively good efficiency through the casing 1, and in this regard it is believed that such emission can travel for hundreds of meters through the casing without unacceptable levels of attenuation for detection. For example, suppose the casing experiences strain at time t=0 at location 140, thereby generating microfracture-induced acoustics. These acoustics will travel though the casing until it reaches the sensor station 100 above it (e.g., at time t=t0) and below it (at time t=t0′), where t0 and t0′ will vary depending on whether location 140 is closer to the top or bottom station, and will vary in accordance with the speed of sound within the casing. At those times, the acoustics are detected at each of these two stations pursuant to the frequency analysis technique disclosed above. If not significantly attenuated, the acoustics will then propagate to the next sensor stations. Assuming the acoustics propagate between the stations 100 at a time of Δt, they will be seen at the next stations at times t=t0+Δt and t=t0′+Δt, and so on. Accordingly, by assessing the time of arrival of the acoustics at each station, the location of the strain that is generating the microfracture acoustics, i.e., at location 140, can be determined, which might allow for inspection of this location or other corrective action. This assessment can be made before or after converting the time-based acoustic signals to frequency spectra. If time based-acoustic signals are used, well known cross correlation techniques, such as those disclosed in U.S. Pat. No. 6,354,147, can be used to compare the signals at each of the stations and to compare them to understand the relative differences in time that the acoustics arrive at each of the sensor stations.
When detecting dynamics strains such as those emitted by microfractures in the casing, the sensing elements may comprise accelerometers, such as piezoelectric accelerometers capable of detecting the frequencies of interest. In this regard, it should be noted that although the use of fiber optic sensors are preferred in conjunction with the disclosed technique, the use of such sensors is not strictly required.
As fiber optic sensors generally, and specifically the fiber optic sensors disclosed herein, are sensitive to temperature, one skilled in the art will recognize that temperature compensation schemes are preferably necessary in conjunction with the disclosed techniques and apparatuses. Such compensation can be necessary to distinguish whether sensor deformation results from stress (e.g., from compression or tension of the sensors) or from temperature (e.g., from thermal expansion of the lengths of the sensors). For example, an FBG isolated from the casing (and other) strains, e.g., can be used to detect the temperature so that the disclosed sensors can be compensated for to understand only the pressures impingent upon them. As such temperature compensation schemes for fiber optic sensors are well known, and can constitute a myriad of forms, they are not disclosed further.
It is contemplated that various substitutions, alterations, and/or modifications may be made to the disclosed embodiment without departing from the spirit and scope of the invention as defined in the appended claims and equivalents thereof.
This application contains subject matter similar to that disclosed in Ser. No. 10/441,234, entitled “Housing On The Exterior of a Well Casing for Optical Sensors,” which is filed concurrently herewith, and which is incorporated herein by reference in its entirety.
| Number | Name | Date | Kind |
|---|---|---|---|
| 2846662 | Sparks | Aug 1958 | A |
| 4360272 | Schmadel et al. | Nov 1982 | A |
| 4589285 | Savit | May 1986 | A |
| 4649529 | Avicola | Mar 1987 | A |
| 4654520 | Griffiths | Mar 1987 | A |
| 4745293 | Christensen | May 1988 | A |
| 4761073 | Meltz et al. | Aug 1988 | A |
| 4806012 | Meltz et al. | Feb 1989 | A |
| 4825424 | Lamb et al. | Apr 1989 | A |
| 4898236 | Sask | Feb 1990 | A |
| 4950883 | Glenn | Aug 1990 | A |
| 4996419 | Morey | Feb 1991 | A |
| 5012088 | Cole et al. | Apr 1991 | A |
| 5051965 | Poorman | Sep 1991 | A |
| 5111903 | Meynier | May 1992 | A |
| 5163321 | Perales | Nov 1992 | A |
| 5308973 | Odoni et al. | May 1994 | A |
| 5317383 | Berni | May 1994 | A |
| 5319435 | Melle et al. | Jun 1994 | A |
| 5327216 | Berni | Jul 1994 | A |
| 5339696 | Carignan | Aug 1994 | A |
| 5353637 | Plumb et al. | Oct 1994 | A |
| 5361130 | Kersey et al. | Nov 1994 | A |
| 5380995 | Udd et al. | Jan 1995 | A |
| 5397891 | Udd et al. | Mar 1995 | A |
| 5401956 | Dunphy et al. | Mar 1995 | A |
| 5426297 | Dunphy et al. | Jun 1995 | A |
| 5444803 | Kim et al. | Aug 1995 | A |
| 5451772 | Narendran | Sep 1995 | A |
| 5452087 | Taylor et al. | Sep 1995 | A |
| 5493113 | Dunphy et al. | Feb 1996 | A |
| 5493390 | Varasi et al. | Feb 1996 | A |
| 5495892 | Carisella | Mar 1996 | A |
| 5497233 | Meyer | Mar 1996 | A |
| 5507341 | Eslinger et al. | Apr 1996 | A |
| 5513913 | Ball et al. | May 1996 | A |
| 5529346 | Sperring | Jun 1996 | A |
| 5564504 | Carisella | Oct 1996 | A |
| 5636021 | Udd | Jun 1997 | A |
| 5675674 | Weis | Oct 1997 | A |
| 5680489 | Kersey | Oct 1997 | A |
| 5767411 | Maron | Jun 1998 | A |
| 5789669 | Flaum | Aug 1998 | A |
| 5804713 | Kluth | Sep 1998 | A |
| 5892860 | Maron et al. | Apr 1999 | A |
| 5925879 | Hay | Jul 1999 | A |
| 5986749 | Wu et al. | Nov 1999 | A |
| 5987197 | Kersey | Nov 1999 | A |
| 6016702 | Maron | Jan 2000 | A |
| 6072567 | Sapack | Jun 2000 | A |
| 6175108 | Jones et al. | Jan 2001 | B1 |
| 6233374 | Ogle et al. | May 2001 | B1 |
| 6233746 | Skinner | May 2001 | B1 |
| 6268911 | Tubel et al. | Jul 2001 | B1 |
| 6354147 | Gysling et al. | Mar 2002 | B1 |
| 6384919 | Fersht et al. | May 2002 | B1 |
| 6435030 | Gysling et al. | Aug 2002 | B1 |
| 6501067 | Jones et al. | Dec 2002 | B2 |
| 6840114 | Niezgorski et al. | Jan 2005 | B2 |
| 20020041722 | Johnson et al. | Apr 2002 | A1 |
| Number | Date | Country |
|---|---|---|
| 2 284 256 | May 1995 | GB |
| 2 354 782 | Apr 2001 | GB |
| 2 367 890 | Apr 2002 | GB |
| WO 02057805 | Jul 2002 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20040246816 A1 | Dec 2004 | US |
