NO320060B1 - Procedure for borehole grinding using reverse inertial navigation - Google Patents

Description

BACKGROUND OF THE INVENTION

Field of the invention

The present invention generally relates to methods for inspecting underground boreholes. For example, the invention includes equipment for measuring while drilling (MWD) and which is carried out to be able to determine the position and direction of progress for a tandem-coupled sub or probe near the drill bit on a drill string assembly in an oil or gas well. The invention also applies to cable-guided instruments which are designed to determine the position and direction of progress for such instruments during logging work in a borehole.

Description of Related Art

Borehole surveys are usually carried out using a drill string or cable-guided measuring sub or probe that includes several magnetometers and accelerometers arranged for each of the probe's three mutually orthogonal axes. In the case of an MWD operation, the measuring probe is arranged as part of a special drill bit weight tube which is placed a relatively short distance above the drill bit on a drill string, as indicated by the probe M in fig. 1, and which will be described in more detail below. The drill bit is used for drilling through underground formations and drills out the well under torque produced either by turning the drill string with the help of a rotary drill on the ground surface or with the help of a downhole mud motor.

The direction coordinates for the measuring sub are determined at several points along the borehole by generating magnetometer data from the earth's magnetic field as well as accelerometer data from the earth's gravitational field in relation to the mutually orthogonal axes of the measuring probe. An "h" vector from the magnetometer and a "g" vector from the accelerometer are then used to determine the borehole direction. More specifically, the h-vector is used to indicate the azimuth of the borehole, while the direction of progress in relation to magnetic north and the g-vector is used to indicate the inclination of the borehole, which means the hole's angular deviation from the vertical direction. When a sufficient number of measurements at fixed points along the borehole have been carried out, the measurement values are plotted on a map of the borehole. This mapping or inspection can be established in relation to a desired coordinate system.

Such methods are of limited importance due to the fact that the earth's magnetic field varies over time, and is affected by objects containing iron or magnetic ore in the local vicinity. Such effects lead to uncertainties and errors when determining the direction of the borehole when using a measuring sub which is dependent on the magnetometer.

Various considerations have produced a constantly desired need for more accurate and more compact techniques for borehole mapping. Modern techniques for drilling for oil and gas have e.g. produced boreholes with a smaller diameter and often require the wells to lie close together. In addition, it is not unusual for a number of wells to be drilled against different geological targets from a single wellhead or a single drilling platform. Draining relatively large deposits has also made it necessary to drill deeper and gain access to smaller target formations.

A proposed method for producing a probe with a small diameter in borehole survey equipment and for providing more accurate measurement results concerns the utilization of inertial navigation techniques. In general, inertial navigation techniques utilize an internal measurement unit (IMU) containing a set of accelerometers and a set of gyroscopes, not unlike the schematic representation of an IMU in Fig. 2, which will be described in more detail below. The accelerometers emit signals indicating the instrument package's acceleration along the three axes in a Cartesian coordinate system. The gyroscopes emit signals that represent the angular rate at which the instrument package is rotated in relation to the same Cartesian coordinate system. Variations in the magnetic field can theoretically be eliminated by adding gyroscopes to each of the probe's mutually orthogonal axes. The probe's direction of travel can then be determined from accelerometer data and gyroscope data from each of these axes. Accelerometer data is affected by the Earth's gravitational field, while gyroscope data is affected by the Earth's rotational speed relative to inertial space.

The first basic type of inertial navigation equipment is "gimbal" equipment, an example of which is shown in fig. 3. Gimbal equipment, gyroscopes and accelerometers are mounted on a fully gimbal suspension platform which is maintained in a predetermined directional orientation by means of gyro-regulated servo equipment. This arrangement effectively maintains the accelerometers in a fixed directional relationship, so that the accelerometers can emit signals relative to a coordinate system that is essentially fixed in inertial space. Continuous integration of the acceleration signals with respect to time provides signals representing velocity and position of the instrument package relative to inertial space. However, known gimbal suspension equipment has usually proved unsatisfactory due to the size of the suspensions required for the gyroscopes. Such equipment cannot easily withstand such shocks, vibrations and temperatures that one is constantly faced with when examining deep boreholes. Gyroscope operation, precession, sensitivity to g-forces and other factors also affect the accuracy of the equipment.

The other basic type of inertial navigation equipment is permanently mounted such equipment, an example of which is given in fig. 4 on the drawings. For example, US 4,812,977 provides a borehole surveying system that uses a fixed inertial navigation technique for mapping a borehole. This system attempts to correct for downhole conditions to generate a more accurate survey. In a fixed equipment, the gyroscopes and accelerometers are fixedly connected to and rotate with the instrument package and thus with the borehole survey probe. In such equipment, the accelerometers produce signals indicating the acceleration of the instrument package in a Cartesian coordinate system which is stationary in relation to the instrument package. The gyroscope output signals are computerized to transfer the measured acceleration values to a coordinate system that is stationary relative to the Earth. Once transferred to the Earth-referenced coordinate system, the acceleration signals are integrated in the same manner as in gimbal-mounted navigation equipment to provide velocity and position information.

A problem associated with MWD applications exists because the capacity of an MWD storage module is often limited by tool cost and tool length considerations. It is therefore difficult to extract a complete set of output data from an IMU during extraction of the drill string, and it is therefore difficult to carry out navigation continuously all the way from the borehole bean to the wellhead and vice versa.

Errors also build up over time during the inertial navigation calculations utilized in MW D applications. If it is thus attempted to carry out inertial navigation all the way from the wellhead to the bottom of the well and then back up to the wellhead, such errors will be added to each other and become very large together.

It is then an object of the present invention to solve one or more of the problems mentioned above.

SUMMARY OF THE INVENTION

The object stated above, as well as various other objects and advantages, is achieved by means of a method for mapping a borehole and which involves positioning a probe with an inertial measurement device in a borehole at the bottom of the well with an estimated bottom position. The examination of the borehole then takes place by pulling the probe away from the bottom of the borehole while the position of the probe is determined step by step during the withdrawal using the inertial measurement unit. Probe retraction is stopped at a survey endpoint whose true position is known. The survey process is corrected by applying the difference between the probe position at the survey endpoint, as determined by the inertial measurement unit, and the known true survey endpoint position.

In a preferred embodiment, the method additionally includes process steps which involve determining the speed of the probe during the extraction using the inertial measurement unit, comparing the determined speeds with reference speeds to derive speed errors, as well as updating the survey process while utilizing these speed errors.

The inertial measurement unit preferably comprises a body mounted inside the probe and which contains several accelerometers to emit signals representing the body's acceleration in relation to the three axes in a coordinate system for this body. The body in question also contains several gyroscopes to generate signals that represent the body's rotational angular velocity in relation to the three axes in the body's coordinate system.

The accelerometers and gyroscopes can be mounted in different ways inside this body. E.g. the accelerometers and gyroscopes can be fixed to the body. Furthermore, the accelerometers and gyroscopes can also be mounted on a three-axis inertial stabilized platform connected to the body.

Preferably, the gyroscopes are selected from a gyroscope group consisting of hemispherical resonator gyroscopes, laser ring gyroscopes, vibrating gyroscopes, fiber optic gyroscopes and tuned dry gyroscopes.

According to the present invention, it is thought that the probe can form part of a weight tube arranged on a drill string, or be part of a probe suspended on a cable in the borehole.

In one embodiment of the present invention, the inertial measurement unit indicates the position of the probe by generating acceleration signals referred to the body's coordinate system and angular velocity signals using accelerometers and gyroscopes respectively. These acceleration signals referenced to the body coordinate system and angular velocity signals are computerized to update a direction cosine matrix for transformation from the body coordinate system to the inertial space coordinate system. The acceleration signals according to the body coordinate system are translated into acceleration signals in the inertial space coordinate system using this updated body/inertial direction cosine matrix. This updated direction cosine matrix for translation from body to inertial and the acceleration signals in the inertial coordinate system is processed in an inertial navigation calculation. Velocity signals in a local/vertical coordinate system are also calculated. These calculated velocity signals in the local vertical coordinate system are then compared with the reference velocity signals to derive velocity error signals. These speed error signals are computer-processed to compensate for errors during the survey. Position signals in a local/vertical travel azimuth coordinate system are calculated using the updated body/inertial direction cosine matrix and the acceleration signals in the inertial coordinate system.

The reference speeds used to derive speed error signals can be generated from different sources. The reference speeds can e.g. is produced by temporarily reducing the speed of the probe, essentially to zero, within several time periods during the probe's extraction, so that zero speed references are created. In this case, the velocity errors are determined using the probe velocities determined by the inertial measurement unit during the various zero-velocity periods.

In the event that the probe is a probe that is suspended on a cable in the borehole, the reference speeds can alternatively be derived from signals representing a large number of cable speed readings.

The speed errors are preferably processed using a Kalman filter.

It is also preferable that a Kalman filter is used to correct the local/vertical travel azimuth position signals, so that the accuracy of the well survey is improved.

The subject matter of the present invention can also be expressed as a method which comprises setting the position of a probe with an inertial measurement unit in a borehole with the bottom of the borehole as the assumed target. A survey using inertial navigation equipment is initiated using the coordinates of this target on the bottom of the borehole indicated by the output signals of the inertial measurement unit. The probe is pulled away from the target at the bottom of the borehole and mapping data is produced from the inertial measurement unit's output signals during the pullout. The probe's velocities are also determined during the extraction using the inertial measurement unit. The determined velocities are compared to reference probe velocities to derive velocity errors. Survey data is updated by exploiting these speed errors. Probe extraction is stopped when the inertial measurement unit output signals indicate that the probe is in a position at a survey endpoint whose true coordinates are known. Survey data is then corrected using the difference between the endpoint coordinates of the probe survey indicated by the inertial measurement unit output signals and the known true endpoint coordinates of the survey process.

BRIEF DESCRIPTION OF THE DRAWINGS

So that the manner in which the above-mentioned distinctive features, methods and objects of the present invention have been achieved can be understood in detail, a more comprehensive description of the invention which has been briefly summarized above will now be given with reference to a preferred embodiment which is shown on the attachments

drawings.

However, it should be noted that the attached drawings only indicate a typical embodiment of the present invention and must therefore not be regarded as limiting the scope of the invention, as the invention also enables other equally effective embodiments.

Reference must now be made to the drawings, whereupon:

fig. 1 shows an elevation, partially in section, of a floating drilling vessel and a downhole measuring probe in an MWD equipment according to the present invention,

fig. 2 schematically shows the downhole measuring probe which particularly illustrates the (Inertial Measurement Unit (IMU);

fig. 3 is a sketch of a gimbal suspension inertial navigation equipment (INS), fig. 4 shows a permanently mounted INS,

fig. 5 is a block diagram of the indicated downhole measuring probe in FIG. 1 and 2, and which describes the probe in more detail,

fig. 6 is a perspective sketch of the IMU schematically shown in fig. 2,

fig. 7 is a partially exploded view of a hemispherical resonator gyroscope (HRG) used in one embodiment of the IMU shown in FIG. 6,

fig. 8 schematically shows coordinate systems for inertial navigation and which can be used in accordance with the present invention,

fig. 9 is a block diagram indicating a method for borehole investigation according to the present invention,

fig. 10 schematically shows a parallel displacement correction of the INS-indicated survey positions in accordance with the present invention,

fig. 11 schematically shows a skewed gyroscope reference and an orthogonal reference frame, and

fig. 12 schematically shows a reference frame for an IMU body with new names to define all Euler angles without singularity.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Fig. 1 shows a typical drilling operation where an embodiment of the present invention can be used with advantage during MWD applications. It will be well known to professionals in the field that such applications include taking measurements in an underground borehole with at least part of the drill string, including the drill bit, in a position down in the borehole during drilling, standstill and/or tripping. Ordinary experts in the field will, however, also recognize that the subject of the invention can also find application in cable-fed measuring equipment.

A drilling ship S shown to have conventional rotary drilling equipment 5 equipped with downhole equipment to perform measurements of underground formation properties during drilling. Again, experts in the field will recognize that the invention is in no way limited to applications in connection with drilling vessels, but can utilize other types of offshore drilling equipment and rigs, as well as land-based drilling rigs.

Downhole's measuring equipment is hooked onto the drill string 6, which is turned off

the rotary drill 4 on the drilling ship, and includes one or more weight pipes, such as the weight pipe F shown in fig. 1 and which is equipped with stabilizer(s). Such neck tubes are equipped with appropriate sensors to measure resistivity, porosity and other properties of underground formations. These sensors are components of electrical, nuclear or acoustic instruments that generate signals that indicate the aforementioned formation properties by means of the sensors.

Such signals are usually stored down in storage modules inside the collar F down in the borehole, but the signals can further be transmitted to the earth's surface using standard MWD telemetry devices and methods. A MWD telemetering sub D is e.g. arranged on the drill string 6 inside the bottom assembly BH A, which also includes the casing F and the drill bit B. The telemetry sub T receives signals produced by the instruments in the casing F, as well as from the measurement sub or probe M which will be described below, and remotely transmits these signals over the mud path or drill string 6 to the surface instruments 7 using the pressure sensor 21 in the pump pipe 15.

The drilling rig 5 further comprises a motor 2 which transfers torque to the rotary drill 4 which is in engagement with and turns the drive pipe 3. The drill string 6 is coupled for rotation together with the drive pipe 3, and is thus turned together with the drive pipe by the rotary drill. The drill string 6 in turn rotates the drill bit B, which brings the drill bit to drill through the underground formations 32 and thereby form the drill well 9 in them. The diameter of the drill pipes that form the drill string 6 is smaller than the average diameter of the drill well 9, which leads to the formation of an annular space 10. This annular space is then defined as the part of the drill well that lies between the outside of the drill string 6, including the bottom assembly BH A and the inside or wall of the borehole. Such an annular space is also formed by the tubular liner 18 which runs from the drillship S through at least an upper part of the borehole 9 in the seabed.

The middle parts of the drive pipe 3 and the drill pipe sections that form the drill string 6 are hollow, and the downhole subs T, M and F comprise all through passages, and all this is for the purpose of conveying drilling fluid or "mud" from the drill ship S to the drill bit B. The mud is drawn from a mud pit 13 on board the drilling ship by means of a pump 11 and is delivered to the pump pipe 15 and the injector head 8 through the drive pipe 3, the drill string 6, the subs T, M and F, and finally to the drill bit B. This mud serves to lubricate the drill bit and lead drilling chips in the borehole up to the ground surface through the annulus 10. The returned sludge is then returned to the sludge pit 13, where the drilling chips are separated and the sludge is degassed and made ready for further use.

The measuring sub or probe M is shown schematically in fig. 2 with an inertial measurement unit IMU which comprises a body 20 mounted inside the probe M and which contains several accelerometers 52, 54, 56 to each emit a signal Gx, Gy and Gz which indicates the acceleration of the body along each of the three axes in a coordinate system which is determined in relation to the body 20, as indicated by the x-, y-, and z-axes indicated in fig. 2. This coordinate system is then referred to as the "body coordinate system". The body 20 further contains several gyroscopes 62, 64, 66 to each emit a signal QXl Qy, and Qz which indicates the body's angular rate of rotation in relation to the three axes in the body coordinate system.

Those skilled in the field will further recognize that the probe M can also be equipped with several magnetometers to emit signals HXl Hy and Hz which indicate the earth's local magnetic north vector in relation to the three axes of the body coordinate system, as will be well known in the related art. The use of such magnetometers together with a special embodiment of the present invention will be described in more detail below.

The accelerometers and gyroscopes can be mounted inside the IMU body in different ways. For example, the accelerometers and gyroscopes can be mounted on a three-axis inertial stabilized platform which is connected to the probe M, as shown in fig. 3. The accelerometers and gyroscopes can also be permanently mounted on the probe body 20, as indicated in fig. 4.

The three-axis inertial stabilized platform mechanism in fig. 3 constitutes an on-board reference axis system supported by gimbal suspension, torque motors, slip rings, servos and gyroscopes. The gimbal-suspended platform isolates the inertial sensors from the probe's rotational motion. The three accelerometers Ax, Ay and Az are mounted on the inertial platform 22 which is attached to the probe M. The gyroscopes gx, gy and gz are used to maintain a fixed position of the inertial platform in relation to the inertial space. On a platform of this type, the gyroscopes do not measure anything, so they do not need to be able to measure large rotational values. The gyroscopes in fig. 3 can thus be platform gyroscopes at a deviation performance of 0.01 degrees per hour and with a relatively small dynamic range.

In contrast to the inertial platform in fig. 3, shows fig. 4 a fixed mechanism where all six inertial sensors are fixedly connected to the IMU body 20. The fixed gyroscopes actually measure the rotational movement of the probe, unlike the gyroscopes on the inertial platform. A normal permanently mounted gyroscope with a performance of 0.01 degrees/hour must then have a larger dynamic range, e.g. two orders of magnitude larger than an inertial platform gyroscope.

Those skilled in the art will recognize that both the inertial platform version and the fixed mounting version can be used in conjunction with MWD. However, each of these schemes is characterized by certain advantages and certain disadvantages, which can make choosing one or the other a logical choice under certain circumstances. The gimbal structure of the inertial platform equipment constitutes e.g. a labour-intensive, precision-made mechanism that will be expensive and delicate. Equipment with gimbal suspension will therefore not be suitable for applications where strong shocks can occur.

The fixed equipment, on the other hand, has its own limitations, but it constitutes a robust construction that is better suited to withstand shock and vibration. The fixed mechanism is thus currently preferred in connection with the present invention, particularly under MWD conditions.

Fig. 5 shows a block diagram with applied designations and which indicates a specific embodiment of the MWD probe M in fig. 1 in more detail. The MWD probe M comprises an electronics subdivision 201 comprising the power supply 202, several magnetometers 204 and a main regulator 206. The MWD probe M also comprises the IMU body 20 whose components communicate with the electronics subdivision 201 over an external data bus 208. The IMU body 20 comprises a set of three gyroscopes 62, 64, 66 and a set of three accelerometers 52, 54, 56, as mentioned above. Fig. 5 also indicates a program storage device 210.1 In the particular embodiment shown, the signals generated by the accelerometer set and the gyroscope set are usually processed by software after being collected by the acquisition electronics 212, 214 mounted on the IMU body 20. More specifically, the program storage device 210 is coded with instructions which, when carried out, will constitute a method of the same kind as the method 100 which is visualized in fig. 9 and will be described below. The program storage device can be placed down in the borehole, such as the embodiment shown in fig. 5, or on the upper side of the borehole. If it is placed uphole, data is transmitted up the borehole to equipment such as the surface instruments 7 indicated in fig. 1, for processing and storage. Alternative embodiments may choose to place the program storage device 210 downhole within the electronics sub-section 201. If located downhole, the instructions may be executed by a downhole processor, such as the main controller 206. Fig. 6 is a perspective view of the IMU body 20 and the components of this which is first mentioned above in connection with fig. 2, as it can be used in a specific embodiment of the present invention. The IMU body 20 comprises a set of three accelerometers 52, 54, 56 which are placed perpendicular to each other, and a set of three HRGs 62, 64, 66. The simplest embodiment of the IMU also places the three gyroscopes orthogonally at 90° to each other to each other. Fig. 6 shows another alternative embodiment where the orthogonal arrangement of the three gyroscopes is skewed to reduce the package size. In this embodiment, which will be discussed in more detail below in connection with fig. 7 and 11-12, the input axes of the gyroscopes are skewed 60° from each other in relation to the longitudinal axis of the MWD probe M. Each of these embodiments, as well as others, can be used in different structures of the present invention.

Preferably, the gyroscopes are selected from a gyroscope group consisting of hemispherical resonator gyroscopes, ring laser gyroscopes, vibrating gyroscopes, fiber optic gyroscopes and tuned dry gyroscopes. The hemispherical resonator gyroscope, which is also known as the HRG, has been shown by an airframe test to be particularly well suited for fixed navigation equipment, mainly due to its inherent simplicity. HRG has a very broad dynamic operating area. Unlike a ring laser gyroscope, the HRG has no low-rate blocking and does not require any vibration mechanism that would add noise to the position outputs.

Reference must now be made to fig. 7, where a special embodiment of the HRG is shown in the form of a solid-state gyroscope whose inertia-sensitive element is a closed axisymmetric casing of silicon oxide (quartz) which forms resonator 32. The resonator 32 is rotated about its axis and the oscillating mass elements on this experiences Coriolis forces that cause a standing wave to rotate or precess relative to the resonator. In addition to the resonator 32 itself, the two-part recording housing comprises a driver 36 made of quartz with fused-on electrodes for capacitive driving and reading. The driver 36 and the recorder 38 are joined by means of a glass fusion bond. The entire assembly of resonator 32, driver 36 and recorder 38 is then soldered together using a tin/silver/indium solder, inside an Invar vacuum sleeve 34.

HRG in fig. 7 is oriented so that the longitudinal axis 42 of the resonator 32 is inclined in relation to the longitudinal axis 44 of the casing 34. However, the object of the invention is not limited to such an embodiment, and the HRG can be mounted so that the longitudinal axes 42, 44 lie in line or are perpendicular at each other. The skewed design shown allows for a package of three HRGs, of the type shown in fig. 6, usually becomes more compact than an unbiased bundle of three HRGs (not shown). Compactness is usually desired in the design and operation of such instruments, and such an inclined design may then be preferable to a linear design in many applications.

Construction and operation of the particular HRG shown in fig. 7, is set forth in more detail in US Patent No. 5,712,427 entitled "Vibratory Rotation Sensor with Scanning-Tunneling-Transducer Readout", issued January 27, 1998 to Litton Systems Inc., assigned from inventor Anthony Matthews.

The MWD probe M comprises in the embodiment shown in fig. 1 at least two MWD operating modes during drilling: (i) a gyrocompass mode which is in operation whenever the drillstring 6 is stopped to make a connection or disconnection of a pipe section (approximately for every 27 meters of drill pipe), where the measurement is triggered by the cyclic operation of the mud pumps switched off, as the measurement results are stored down in the borehole in a storage module (not shown) which is mounted on the IMU body 20, or optionally stored over mud pulse telemetry. (ii) a continuous inertial navigation mode, where the MWD probe M works for continuous survey of the borehole, e.g. starting from the borehole well before the drill string is pulled out.

At a certain point during the drilling process, the instruments inside the probe M will indicate that the probe is located mainly on the target at the bottom of the borehole. At this point, the drill string 6 will usually be pulled out of the drill well 9.1 according to the present invention, the dead time inherent during the extraction of the drill string from the drill well is used for accurate mapping of the drill well. The borehole is thus examined by extracting the probe from the bottom of the borehole while determining stepwise positional changes of the probe during the extraction using the inertial measurement unit IMU inside the probe M. The probe extraction is stopped at an end point for the investigation, e.g. in the wellhead on the surface of the earth or at the mudline of the seabed where the wellhead is located, and whose true position is known. Survey data is then corrected based on the difference between the probe position at the survey end point, as determined by the inertial measurement unit, and the known true end point of the well survey.

Fig. 8 shows a variant of the coordinate systems used in the further description of the present invention. According to the present invention, information from several different coordinate systems is utilized. E.g. information is collected in a certain coordinate system, in other words within the reference frame for the probe M, and utilized in another reference system, e.g. inertial reference system met. More specifically, fig. 8 schematically typical orthogonal coordinate systems related to the stationary inertial navigation equipment discussed herein, such coordinate systems comprising:

• The inertial coordinate system: (X<1>, Y', Z')

• The earth-fixed coordinate system: (Xe, Y6, Z<e>)

• The local/vertical coordinate system: (east, north, up)

• Local/Vertical Walk Azimuth (LVWA) coordinate system: (Xa, Y3, Za)

• The body coordinate system: (X<b>, Y", Z<b>)

The inertial coordinate system is defined as a right-handed, non-rotating coordinate system in relation to a fixed star, where the X-axis runs parallel to the Earth's north pole axis. The geostationary coordinate system is defined as a right-handed coordinate system on Earth where the X-axis is directed along the Earth's polar axis, but where the coordinate system revolves around the X-axis together with the Earth and with the Earth's rate of rotation. For example, the local/vertical (<n>LV") coordinate system can be defined so that its X-axis is directed east, its Y-axis is directed north and the Z-axis is directed upward. Local/vertical travel azimuth ( "LVWA") coordinate system is defined as the local/vertical coordinate system coincides with the local/vertical coordinate system when the wandering angle a is equal to zero. Otherwise, the LVWA coordinate system can be obtained by rotating the LV coordinate system by an angle a about the upward axis. Finally, the The body coordinate system is defined so that the X-axis is in the forward direction on the IMU body along the body's axis, while, for example, the Y-axis is then to the left and the Z-axis is upwards.

As mentioned above, the probe M used to implement the present invention comprises an electronics subdivision as well as an IMU. This IMU comprises three single-axis HRGs and three single-axis accelerometers. These inertial sensors, namely the HRG units and the accelerometers, are mounted on the IMU body 20 which is immovably attached to the probe housing. By utilizing three one-axis HRG and three one-axis accelerometers as a fixed inertial measurement unit (IMU), three axes of angular and linear displacement signals can be generated within a certain time.

Reference must now be made again to fig. 1, where gyroscope and accelerometer data in an embodiment of the present invention is sent uphole by the IMU in the probe M to be processed by the surface instruments 7. The only part of the INS equipment that is located down the borehole of the MWD probe M is then the IMU . The remaining functions shown in fig. 10 is then placed on the upper side of the borehole and is performed by various parts of the surface instrument package 7. The remaining work functions are then usually defined and performed with the help of software, so that they are implemented by a processor that executes instructions on a storage device.

In another embodiment, the six linear and angular displacement signals produced by the inertial sensors are analyzed using the following calculation algorithms implemented on the downhole probe: (i) gyrocompass calculations for initial determination of the probe's position by means of analytical (coarse) facility calculation as well as fine facility calculation , possibly by using a Kalman filter, (ii) velocity/position calculations using the three-axis HRG signals which update a quaternary or a direction cosine matrix (DCM) from the probe's body frame relative to the inertial frame as well as translating step speed -theity from the three-axis accelerometers to the inertial frame, and (iii) position calculations that calculate latitude and longitude using the inertial navigation technique at local/vertical travel azimuth.

The calculations of (ii) and (iii) as well as of true vertical depth ("TVD") are advanced by a linear or non-linear estimator to compensate for system errors in combination with zero rate updates. The linear and nonlinear estimators may include linear and extended Kalman filters ("KF"), backward Kalman smoothers, and minimum model error ("MME") estimators.

Reference should now be made to the block diagram in fig. 9, where it is shown that each of the three orthogonally-oriented accelerometers and the three orthogonally-oriented gyroscopes in the fixed-mounted IMU body 20 produces a temperature-compensated signal, as indicated by the various blocks 110,112. The three accelerometers together form a three-axis accelerometer, as will be well known in the field and then indicate three speed steps for the probe M measured in the body coordinate frame. The three gyroscopes in this particular embodiment are semicircular resonator gyroscopes ("HRG"), as mentioned above, and their signals then indicate three incremental rotation angles, which are also measured in the body coordinate frame. In the embodiment shown in fig. 4 and 6, the three accelerometers and the three gyroscopes are oriented orthogonally in relation to each other. However, such an orientation is not necessary to practice the present invention. Reference must now be made again to fig. 9, where it is indicated that three temperature-compensated three-axis accelerometer signals 110 and three-axis HRG signals 112 from the permanently mounted IMU are then fed into a buffer module in block 114.

The signals 110,112 are then compensated for larger deterministic error coefficients in the accelerometers and gyroscopes using the accelerometer error compensator 116 and the gyroscope error compensator 118, respectively. Examples of "larger deterministic error coefficients" include, but are not limited to, systematic deviations, scale factor errors and skewed settings . The compensated signals 120, 122 are transmitted to a speed/position processor, as indicated at block 124.

The velocity/position processor 124 updates either a body/inertial direction cosine matrix ("DCM") or a body/inertial quaternion. One particular embodiment utilizes body/inertial DCM, and this implementation will be used for further discussion of this particular embodiment. The calculated body/inertial DCM translates the incremental velocity vector from the body coordinate system to the inertial coordinate system. The updated body/inertial DCM and inertial system incremental velocity vector are output to the LVWA inertial navigation processor as indicated at block 126. The LVWA inertial navigation processor 126 utilizes an Earth gravity model 128, which in this particular embodiment is a model from the US -Department of Defense World Geodetic Survey (WGS 72/84).

The velocity vector in the LV coordinate system (east-north-up) as determined by the LVWA inertial navigation processor 126 makes a velocity error observation when the probe M is at rest, as the reference velocity at rest is zero velocity. The reference speeds used to derive speed error signals can be produced from different sources. The reference speeds can e.g. is determined by temporarily reducing the probe's speed essentially to zero for several time periods during the probe's extraction, so that zero speed references are thereby created. In this case, the velocity errors are determined by using the determined probe velocities of the inertial measurement unit during the various zero-velocity periods.

Such zero-velocity cases occur during MWD operations when the drill string pullout is stopped to disconnect a joint or section of drill pipe from the string. In the event that the probe is a probe that is suspended in the borehole on a wireline cable, the reference speeds are alternatively produced by signals representing many different speed readings for the cable.

The velocity error observations are provided, as shown at block 130, to the Kalman filter 132. This Kalman filter 132 evaluates the systematic errors indicated and compensates for these system errors by feedback 134,136,138,140,142 into the inertial navigation system ("INS"). A Kalman equalizer, indicated at block 144, may also be utilized to further correct position outputs 146 in the zero velocity updated (ZUPT) INS corrections for the purpose of minimizing the navigation errors.

At this point, it has been inertial navigated through the borehole using ZUPT corrections, from a target at the bottom of the borehole to a known end point for the survey, such as e.g. a wellhead or a place on the seabed. As soon as the navigation through the borehole is completed, the entire borehole trajectory or mapping is numerically corrected by a parallel transport calculation. Such a calculation appears from the sketch in fig. 10, and can be described mathematically in the following way.

Let (xo, yo, zo) = known, true end-point borehole coordinates, and

(x0,y0,z0)= the end point coordinates of the borehole as they appear from the INS survey using ZUPT.

The displacement error vector from the true endpoint is then given by:

(d(x0,x0), d(y0,y0), dtø, z0)) = (x0 - x0, y0- y0, zo-z0).

The entire motion path Tzupt specified by INS using ZUPT can then be re-represented by:

The corrected movement path TCorr is then given by:

Those skilled in the art will recognize that a downhole position error can be caused by multi-shot gyrocompass settings during MWD operations. The inverse inertial navigation method according to the present invention can reduce such bottom position errors with regard to the entire borehole navigation, as the error that is written from ZUPT is mainly smaller than the errors that are caused by gyro-determined survey.

As indicated above, the present invention is conceived as a reverse inertial navigation method, where accelerometer and gyroscope measurements are taken when the drill string is at rest, such as when additional splice sections of drill pipe are removed from the drill string 6 during extraction of the drill string from the wellbore 9. The probe M can, however, experience movement even if the rig's rotary bit is at rest, namely by delayed "twisting" of the lower part of the drill string in reaction to earlier turning of the upper part of the drill string by the rotary bit. Such a twisting movement can occur with land rigs as well as in connection with floating rigs or drilling ships. In the case of floating rigs and drillships, the sea's pounding movements can also cause movement of the probe M inside the borehole 9.

US Patent No. 5,432,699, which is jointly assigned to the assignee of the present invention and the contents of which are incorporated herein by reference, describes a method and apparatus for correcting such motion-induced errors in the gyroscope signals. The 699 patent makes use of magnetometer output signals to correct the motion-induced gyroscope errors, so that the borehole trajectory can be determined more accurately during periods when drilling operations are halted.

As discussed above and shown in fig. 7, the HRG units in one embodiment of the present invention are skewed with regard to their orientations relative to the longitudinal axis of the IMU body 20. This orientation affects the way in which the HRG signals are analyzed. To illustrate this effect, algorithms for coarse (analytical) alignment and continuous position update for a tilted gyroscope IMU are given below. For common IMU units that include orthogonally mounted gyroscopes, two-axis gyroscope equipment can be used for installations in vertical or horizontal wells. At the expense of setup accuracy, fixed two-gyro IMUs can also be used in awiks wells. A fixed-mounted IMU consisting of three tilted HRG units and three orthogonally-mounted accelerometers is assumed to be used here.

Fig. 11 shows the connection between an example of a skewed gyroscope frame and an orthogonal such frame. A coordinate transformation between these two frames is then given by:

where skewed axes are X, Y and Z, and orthogonal axes are X", Y' and Z\ assumed Furthermore,

So that X is normalized by:

When the mathematical form of the coordinate transformation between the two frames is given as above, then the orthogonal coordinate frame (XYZ) will be used in the following in this description. For the sake of simplicity, it shall be assumed in this description that the accelerometers and gyroscopes share one and the same IMU body 20. Mechanical fault devices between inertial sensors will thus not be considered here.

Analytical facility. At the analytical facility of the IMU, a direction cosine matrix ("DCM") is determined which transforms the local/vertical navigation coordinate frame into a fixed platform (body) coordinate frame. The three Euler angles, namely gyroscope-tool front (roll), pitch (pitch), and azimuth (yaw) can be immediately derived from the DCM. For simplicity, the IMU shown in fig. 11 slanted. The IMU body coordinate frame, namely X, Y, Z in fig. 11, new designations such as X'=Z, Y--Y, Z-X are then given for the purpose of determining all Euler angles without singularity, as shown in fig. 12. The definition of a fixed body frame is actually arbitrary, and it should be applied correctly and incorporated into the IMU device between vertical and horizontal sections of the well to prevent singularities.

As shown below, the DCM transforms vectors in the navigation frame to vectors in the IMU body frame where the actual measurements of physical quantities are performed:

where

f*3 (f<1>) = the specific vector (that is, gravity in a stationary device) relative to the inertial space measured in the body (navigation) frame, and

rob (©") = the angular velocity vector (that is, the earth's rate of rotation in the case of a stationary device) in relation to the inertial space measured in the body (navigation) frame.

Equation (1) can now be expressed as a simple matrix equation in the following way:

for j = 1, 2, 3 and Cjk, the jk-th element of the DCM is Cb = (qi). The inverse matrix of the 3x3 matrix on the left-hand side of (2) exists if f x g><n> * 0, that is, if the gravity vector is not parallel to the Earth's rotation rate vector. Equation (2) consequently has no solution around the cjk where the two vectors are parallel, namely at the north pole and the south pole, and the device then becomes meaningless.

When f x ©n * 0, equation (2) can be solved in the form:

Determining the navigation frame as an east/north/up frame and expressing pregnancy and the earth's rotation rate vectors in this frame then gives: where n = the earth's rotation rate, and L = geographical latitude at the device. The vector product is then given by:

Insertion of (4), (5) and (6) in (3) gives:

Straight-forward calculation of the inverse matrix in equation (7) gives: Idet

By inserting (9) into (7), a set of solutions for equation (2) expressed in elements of DCM is obtained as follows:

where hb is calculated from gyroscope and accelerometer outputs as shown in (9):

Normalization of the vector components then gives:

and with (10) written in simpler form: where

Explicitly expressed by its components, DMC is now given by:

On the other hand, the body/navigation DCM C[) = (Cj| )'<1> can be expressed in terms of Euler parameters, namely implement front (roll) <J>, pitch (pitch) 0, and azimuth (yaw) <* >F in the following way:

Since DCM is an orthogonal matrix, where Cj = (Cj )" 1, comparison between (13) and (14) gives: Then is

where the X-, Y- and Z-axes for the body are defined as forward direction, left and upwards respectively (see fig. 10).

Expressions (16a)-(16c) show that the derivation of the Euler angles is directly based on measurements of acceleration and turning rate. If the measured signals using the gyroscope and accelerometer contain significant random noise, the analytical device based on (16a)-(16c) will no longer be practical. A method that uses a Kalman filter device is then usually used in such cases to achieve system redundancy.

Azimuth error analysis. In gimbal equipment, the mechanical construction of the gimbal usually places a system limitation with respect to the maximum tilt, e.g. at 70° inclination in practice. In fixed equipment, however, there is in principle no such limitation of the maximum inclination, especially not after changing the frame definition of the body, as described earlier.

In general, the azimuth uncertainty is 54<*> when the fixed IMU is pointing true north in the east/north/up navigation frame given by where Bw and Nw are the west (y) gyroscope base and arbitrary yaw angle wander, respectively, and where T is an installation time. Equation (17) is generally used to show the device's quality level in the fixed IMU system.

Tool Front Update. Update of the tool front is derived in a stationary position, which means that there is almost no positional displacement of the IMU during rotation about the body's X-axis.

First, the body/inertial DCM at t=0 is given by:

C„ (0) is then the original navigation/ground-fixed frame DCM, and X is a geographic longitude. Furthermore, C£ (0) is given by (14) over the analytical device.

Let C'b(0) = [ctjk], j,k = 1, 2, 3, and the original quaternion from the body to the inertial frame is then given by:

Now the second-order quaternion update is being pursued. Let A8£ = [AØi, A02, A03]' be a gyro reading (angular increment) at t + At in the body (XYZ) frame relative to inertial space. The second-order updated quaternion is then given by:

Expressed by the quaternary elements: After the quaternary update (21), is periodically normalized (e.g. every 4 sec-und) over

The updated (and normalized) body/inertial DCM is now expressed by the updated and normalized quaternion (22) such that:

On the other hand, the inertia-to-navigation DCM update C" = C"e Cei is given by:

Due to the previous assumption, namely no position displacement of the IMU during the tool front rotation, one then has:

f cosk 0 - sinX C<n>e(t<+>At) = C<n>e(t) = C<n>e(0) = C^O)' = - sinteinL cosL -cosXsinL (25) l, sinXcosL sinL cosXcosL) and due to the earth's rotation in relation to the inertial space during the IMU instrument front's dreinin<g>, one has:

Inserting (25) and (26) into (24), as well as using (23), then gives: in Over (27) and (14) the updated implement front angle is now given by:

From (27) and (14), the updated inclination angle and the updated azimuth angle are also available, and are then given by:

Slope update:

Azimuth update:

The above-mentioned derivation of updated tool front, inclination and azimuth takes into account the earth rotation compensation above (26).

Briefing. The HRG units measure the projection of the horizontal component of the Earth's rotation rate, and the onboard MWD accelerometers indicate the position of the gyroscopes relative to the vertical direction. A classic north-finding determination is then carried out, which gives the equipment's orientation in relation to geographical north, as shown by the above-mentioned equations. The sensed rotation speed is integrated with respect to time to produce the rotation angle variation, as the original orientation provides steering capability using the gyro tool front, as indicated by equation (16a) above.

In at least one embodiment, the IMU 20 will typically be powered by magnetometers, gyroscopes, and accelerometers, all of which are engaged in generating data relevant to the work operation in question. However, the invention is not limited to this. If e.g. power consumption is important, the gyroscopes can be switched off to save electrical power as soon as the magnetic field is within acceptable values. When the gyroscopes are turned off, the work function continues with the gravity tool front utilizing the 6 standard MWD sensors, namely three magnetometers and three accelerometers.

Mapping. The azimuth direction of the well can be determined under all conditions, either within a casing or outside it. Inside the fringing, the HRG units are used to perform a north search operation, as described under orientation, and if the well is not vertical, then the azimuth direction of the well will be determined using equation 16c. Outside the liner, the proposed equipment will be able to either continue to use the gyroscope or also to use the magnetometers. Under certain conditions, disconnecting the gyroscopes will be appropriate to save battery power and allow longer drilling sequences without tripping for replacements. Under certain other conditions, and if the boring runs are short or if the MWD equipment is trip-bin driven, it will be possible to carry out all investigations using the gyroscopes. The on-board redundancy will also allow selection of one or the other working mode in the event of a sensor failure.

In consideration of what has been stated above, it is obvious that the present invention is well suited to achieve all the purposes and features set forth above, together with other purposes and features inherent in the apparatus described here.

As will be immediately clear to those skilled in the field, the subject matter of the present invention can easily be produced in other specific forms without thereby taking place any deviation from the invention's conceptual content or essential characteristics. The present embodiment must therefore be regarded as illustrative only and not as limiting. The scope of the invention is then indicated by the subsequent patent claims or by the preceding description, since all changes that lie within the idea content and equivalence area of the patent claims are therefore intended to be covered by these.

Claims (23)

1. Procedure for examining a borehole, and which includes the following process steps: position setting of a probe (M) with an inertial measurement unit (20) in a borehole (9) at the bottom of the well, characterized in that the method further comprises: examination of the borehole by withdrawing the probe from the bottom of the borehole and including determination of incremental positions for the probe during extraction using the inertial measurement unit, stopping the probe's retraction at an end point for the investigation and if the true position is known, and correcting the survey process using the difference between the probe's post-survey endpoint position, as determined by the inertial measurement unit, and the known true survey endpoint position. 2. Procedure as stated in claim 1, characterized in that the method further comprises process steps which include: determination of the probe's speeds during the extraction using the inertial measuring unit, comparison of the determined speeds with reference speeds to derive speed errors, and updating the survey results using the speed errors. 3. Procedure as stated in claim 1, characterized in that the inertial measurement unit comprises a body mounted inside the probe and which contains several accelerometers to emit signals indicating the acceleration of the body in relation to the three axes in a coordinate system for the body, and several gyroscopes to emit signals indicating the body's angular rate of rotation in relation to the three axes in the body's coordinate system. 4. Procedure as stated in claim 3, characterized in that the accelerometers and gyroscopes are rigidly attached to the body. 5. Procedure as stated in claim 3, characterized in that the accelerometers and gyroscopes are mounted on a three-axis inertial stabilized platform connected to the body. 6. Procedure as stated in claim 3, characterized in that the gyroscope is selected from a gyroscope group consisting of hemispherical resonator gyroscopes, ring laser gyroscopes, vibration gyroscopes, fiber optic gyroscopes and tuned dry gyroscopes. 7. Procedure as stated in claim 1, characterized in that the probe forms part of a weight tube arranged on a drill string. 8. Procedure as stated in claim 1, characterized in that the probe forms part of a probe which is suspended in the borehole on a wireline cable. 9. Procedure as specified in claim 3, characterized in that the inertial measurement unit indicates the positions of the survey probe in accordance with a method comprising the following process steps: generation of acceleration signals and angular velocity signals in the body's coordinate system using accelerometers and gyroscopes respectively, data processing of the acceleration signals and velocity signals in the body's coordinate system to update a body/inertial direction cosine matrix, transfer of the acceleration signals in the body coordinate system to acceleration signals in the inertial coordinate system using the updated body/inertial direction cosine matrix, processing of the updated body/inertial direction cosine matrix and the acceleration signals in the inertial coordinate system in an inertial navigation calculation, calculation of velocity signals in a local/vertical coordinate system , comparison of the calculated velocity signals in the local/vertical coordinate system with reference velocity signals to derive velocity error signals r, processing the velocity error signals for compensation of errors in the survey process, and calculation of position signals in a local/vertical horizontal azimuth coordinate system using the updated body/inertial direction cosine matrix and the acceleration signals in the inertial coordinate system. 10. Procedure as stated in claim 2, characterized in that the reference speeds are produced by temporarily reducing the speed of the probe essentially to zero for several time periods during the extraction of the probe. 11. Procedure as specified in claim 10, characterized in that the speed errors are determined using the probe speeds derived by the inertial measurement unit during the various zero-speed periods. 12. Procedure as stated in claim 2, characterized in that the probe is a cable probe suspended in the borehole on a wireline cable, and the reference speeds are produced by signals indicating several cable speed readings. 13. Procedure as stated in claim 2, characterized in that the speed errors are processed using a Kalman filter. 14. Procedure as stated in claim 9, characterized in that a Kalman filter is used to correct the local/vertical travel azimuth position signals. 15. Procedure for examining a borehole, and which includes the following process steps: position setting of a probe (M) with an inertial measurement unit (20) in a borehole (9) with the bottom of the borehole as the assumed target, characterized in that the method further comprises: initiation of an inertial navigation survey using the coordinates of the target on the bottom of the borehole and which are indicated by the output signals of the inertial measurement unit, extraction of the probe from the target on the bottom of the borehole and derivation of survey data from the output signals of the inertial measurement unit during extraction, determination of the probe's velocities during the extraction using the inertial measurement unit, comparing the determined velocities with the reference probe velocity to derive velocity errors, updating the survey process using the velocity errors, stopping the probe extraction when the inertial measurement unit outputs indicate that the probe has reached an endpoint position for the survey, and whose true coordinates are known, and correcting the survey process using the difference between the probe's derived survey endpoint coordinates, as indicated by the inertial measurement unit outputs, with the known true endpoint coordinates naters for the survey. 16. Procedure as stated in claim 15, characterized by that the reference probe speed is produced by temporarily reducing the speed of the probe to substantially zero for several periods of time during the extraction of the probe, and the speed errors are determined using the determined probe speeds by means of the inertial measurement unit during the various zero speed periods. 17. Procedure as stated in claim 15, characterized in that the inertial measurement unit comprises a body mounted inside the probe and which contains: several accelerometers to emit signals that indicate the body's acceleration in relation to a three-axis coordinate system for the body, and several gyroscopes to emit signals that indicate the body's angular rate of rotation in relation to the three axes in the body's coordinate system. 18. Procedure as specified in claim 17, characterized in that the accelerometers and gyroscopes are rigidly attached to the body. 19. Procedure as stated in claim 17, characterized in that the gyroscopes are selected from a gyroscope group consisting of hemisphere resonator gyroscopes, ring laser gyroscopes, vibration gyroscopes, fiber optic gyroscopes and tuned dry gyroscopes. 20. Procedure as stated in claim 15, characterized in that the probe is a weight tube arranged on a drill string. 21. Procedure as specified in claim 17, characterized in that the inertial measurement unit indicates positions for the investigation probe in accordance with a method comprising the following process steps: generation of acceleration signals and angular velocity signals in the body's coordinate system using accelerometers and gyroscopes respectively, processing of acceleration signals and angular velocity signals in the body's coordinate system to update a body/ inertial direction cosine matrix, transformation of acceleration signals in the body coordinate system into acceleration signals in an inertial coordinate system using the updated body/inertial direction cosine matrix, processing of the updated body/inertial direction cosine matrix and the acceleration signals in the inertial coordinate system in an inertial navigation calculation, calculation of velocity signals in a local/vertical coordinate system, comparison of the calculated velocity signals in the local/vertical coordinate system with reference velocity signals to derive velocity errors signals, processing the velocity error signals to compensate for errors during the survey process, and calculating position signals in a local/vertical travel azimuth coordinate system using the updated body/inertial direction cosine matrix and the acceleration signals in the inertial coordinate system. 22. Procedure as stated in claim 15, characterized in that the speed errors are processed using a Kalman filter. 23. Procedure as stated in claim 21, characterized in that a Kalman filter is used to correct position signals in the local/vertical travel azimuth system.