SEG/San Antonio 2007 Annual Meeting

Transcription

1 Characterization and remoal of errors due to magnetic anomalies in directional drilling Nathan Hancock * and Yaoguo Li Center for Graity, Electrical, and Magnetic studies, Department of Geophysics, Colorado School of Mines Summary Directional drilling has eoled oer the last few decades to become standard operational practice in the oil and gas exploration industry. One method of steering utilizes a technique known as magnetic Measurement While Drilling (MWD). Vector measurements of geomagnetic fields along a well-bore are used to determine it orientation. This method is adantageous in that it is inexpensie and efficient. Howeer, it can suffer from the errors caused by magnetic anomalies such as those produced by telluric currents and magnetic geologic units. We hae deeloped method to quantify the error associated with magnetic geologic units, and correct the drilling process for this error. Introduction In an effort to reduce costs and enhance efficiencies, it is necessary for single platforms to drill numerous directional wells to reach multiple targets, which can exist in lateral excess of fie kilometers from the platform (Clark and Clarke 2001). Accurate steering of directional well requires that the determination of dip and azimuth of the well be within 0.1º and 0.05º, respectiely. Although the required directional accuracy can be achieed using gyroscopic sureying instruments, such sureys are expensie because they require drilling to stop and the entire drill string to be remoed. An inexpensie method is to use magnetic data from Measurements While Drilling (MWD). In this method, the orientation of the drill bit, well-bore is determined from ector measurements of the geomagnetic field using a threecomponent fluxgate magnetometer that is incased in a nonmagnetic portion of the drill string, near the drill bit. The orientation of the drill bit, and subsequently that of the wellbore, relatie to the direction of the geomagnetic field is determined from the magnetic measurements. Assuming a known direction of geomagnetic field, this knowledge yields the true orientation of the well-bore in the geographic reference frame. Continuous magnetic MWD therefore allows one to continually adjust the drilling direction (Clark and Clarke, 2001). The accuracy of steering is thus limited by how accurately the Earth s magnetic field is measured. A measurement error of 50 nt in field strength is considered to be the upper limit beyond which accurate drilling is no longer achieable. Howeer, erroneous measurements of the Earth s magnetic field are unaoidable when magnetic sources are present. Seeral scenarios can create strong anomalies and, therefore, inaccuracies in the measured magnetic field ector for steering. Temporal ariations due to electrical currents flowing in the upper atmosphere and magnetosphere gie rise to telluric noise. Research performed by T. Clark and E. Clarke (2001) suggests that Interpolated In-Field Referencing (IIFR) techniques, which would keep track of fluctuations in the Earth s magnetic field in real time and correct the MWD data accordingly, could proide a method for drilling more accurately in areas strongly affected by this phenomenon. Problems can also arise from the general oil producing enironment, which frequently requires new wells to be drilled part-way in existing casings, or in close proximity to other wells that are cased. Significant inaccuracies also occur when the drilling enironment contains magnetic strata or ized magnetic geologic units. This scenario presents significant risk to directional wells, which must be drilled with MWD measurements in accuracies of 0.1º in declination, 0.05º in inclination, and within 50 nt in total field strength in order to hit targets that are 100 meters thick or less (Ngugi, 2002). As such, the characterization and correction for this type of magnetic noise is necessary in directional drilling applications. In this paper, we first deelop a MWD geo-steering algorithm based on three-component magnetic measurements. We then use this algorithm to simulate the errors in directional drilling in the presence of different random and coherent errors in the MWD data. MWD geo-steering method The magnetic geo-steering method utilizes three-component measurements of the magnetic field to recoer the dip and azimuth of the well-bore. The orientation is used together with the well length to calculate the well position in 3D space. The drill then steers the drill towards the next target point along the planned well path. The ability to determine the borehole orientation thus enables the drill rig operator to steer the drill dynamically, and follow a planned borehole path towards a target. This process, and associated possible errors, is illustrated in Figure 1. Thus, the critical part of geosteering is the determination of the dip and azimuth. We describe briefly our approach to determine these two quantities based on ector magnetic measurements. The dip and azimuth is defined in a global geographic coordinate system. We adopt a right-hand coordinate with x- axis pointing to north, y-axis pointing to easting, and z-axis pointing ertically downward. At any particular point along the well-bore, the three-component magnetic sensor, affixed 663

2 to the drill bit, yields the measurement of geomagnetic fields in a coordinate system. Assuming that the borehole dip θ is measured from horizontal surface down and azimuth φ is measured from the north, a commonly used conention has the z-axis pointing downward along borehole and the x-axis pointing perpendicular to the borehole in the direction of the azimuth. The y-axis completes the right-handed coordinate system and is 90 clockwise from the azimuth and perpendicular to the borehole. Based upon the aboe definition, and global field components are related to each other by a rotation matrix: B = R ϕ,θ B, (1) ( ) global Errors in geo-steering using magnetic MWD where R ( ϕ,θ ) is gien by (Li and Oldenburg, 2000), cosϕ R( ϕ, θ ) = sinϕ cosϕ cosθ sinϕ cosϕ sinϕ cosθ 0 The borehole orientation is then determined by finding the dip and azimuth that would reproduce the measured magnetic field components. This inoles a bi-ariant leastsquare minimization problem that minimizes the misfit between the measured magnetic field and that predicted by a gien orientation of well-bore defined by the dip and azimuth. We hae chosen to use a downhill simplex method to carry out the minimization (Gill et al., 2004). (2) Figure 1. Illustration of geo-steering algorithm. The points P i and the dotted lines are the planned. C 1 represent a calculated position at a gien point based on the recoered dip and azimuth from MWD measurement. A correction is then calculated to target towards the next planned point on the well path. Howeer, if the errors in dip and azimuth caused the actual position to be at A 1 instead of C 1, then the actual target would be at T 1, which deiates from the planned well position. It is important to note that any additional magnetic fields, other then the global field, will cause deiations between the measured field components and those predicted from the aboe rotation. Such additional fields could be from artificial sources or coherent magnetic geologic strata. The resultant alues for the dip and azimuth of the well-bore would be erroneous, and could lead to steering errors. Numerical simulations of steering errors With the aboe algorithm, we hae carried out a set of simulations to understand the magnitude of steering errors produced by two types of magnetic field errors. The first is random noise and the second is the coherent perturbation produced by geologic units. To simulate the effect of random noise, we hae perturbed the different components of measured magnetic field using a Gaussian random noise. We hae obsered that errors in all three components result in highly irregular well-bores and an error in a single component leads to deiation to one direction. Figure 2 shows the result of perturbing the ertical component by 10%. The deiation of final well position from the planned path is unacceptably large. Figure 2. Synthetic directional well example. Borehole is drilled due east (90º) at a 45º degree angle with respect to the surface. Surface global magnetic field is characterized by an inclination of 45º, declination of 90º (same orientation of the borehole), the blue asterisks represent the apparent location of the drill bit as calculated from the obsered magnetic field, and the red solid line represents the actual drill path taken. For the geologic units, we hae examined both sheet-like features representing laterally extensie units such as a basalt layer and spheres representing ized compact causatie bodies. For this simulation, we define the geosteering error as the difference between the final position of the drill bit with 664

3 and without the presence of a magnetic body. The final error resulting from the magnetic body s influence is inherently a function of its distance from the borehole, size, and magnetization strength. We hae carried out systematic simulation to understand how this error aries with source parameters. We reproduce the latter study by using a spherical source for breity here. More extensie results will be discussed in the presentation. The sphere is represented by a single dipole, whose magnetic field is gien by, Errors in geo-steering using magnetic MWD µ 0 1 B = m 4π r r, (3) where µ is the permeability of free space, 0 m is the dipole moment, and r r is the distance from the source point to the obseration location. For this source, the size of the source body and its magnetization strength merge as one parameter. The mathematical representation for the MWD measurement of the systematic erroneous magnetic field superimposed upon the global magnetic field is Figure 3. Component measurements represent the pure error only. The horizontal axis represents the absolute distance along the borehole from its collar, and the ertical axes show the magnetic field components. The sphere is situated at 2 km east of the well and at a depth of 2.5 km, has a radius of 1km and a susceptibility of 0.1 SI units. Note that errors in all magnetic field components are greater than 50 nt. The resulting steering error is about 46 m. B = R µ 1 global ( ) ( ) 0 ϕ, θ B + R ϕ, θ m 4π r r (4) where the MWD obsered magnetic field components, B, is the superposition of the global field ector and the magnetic produced by the geologic unit. We hae conducted four separate numerical simulations to understand the error associated with a confined magnetic geologic body represented by a sphere. All of the simulations were performed on a synthetic drill path that has 101 planned points. It has a constant dip of 45º and azimuth of 90º east of north. Figure 3 presents one particular scenario with a sphere situated at a distance of 2.3 km away from the borehole. The errors in magnetic field components are well in excess of 50 nt. The maximum error in dip caused by the presence of the sphere is 0.32º. The drill bit ends at a distance of 46 m to the west and 42 m below its intended target point (see Figure 4). It is clear from this simulation that magnetic sources in geologic units that are at a large distance can hae a significant impact on the accuracy of geo-steering based on MWD magnetometers. Figure 4. Corrected and un-corrected paths charted against the planned drill points (black diamonds). The solid blue line represents the un-corrected drill path and its corresponding error from the final planned drilling point. The red line represents the corrected path, which deiates from the final planned point by centimeters. 665

4 Errors in geo-steering using magnetic MWD Correction simulation for accurate magnetic geosteering In order to correct for this type of errors, we inert surface magnetic anomaly data to characterize the source parameters of geologic bodies. Inersion of such data proides us with an estimation of the source properties from which we can then predict the anomalous field produces by the source along the well- bore. Either a parametric inersion for simple source bodies or generalized inersion for a distribution of sources in 3D can be carried out. Much work exists in the exploration geophysics literature on this subject, and many options are aailable. For our simulation, we assume the knowledge of a simple dipole and inert the total-field anomaly on the surface to recoer its position and dipole strength. We then calculate the three-component magnetic field along the well-bore to predict the error component in the MWD data. We carry out a second simulation to steer the directional drilling by correcting for the field. The result is shown in Figure 4. With the inerted dipole source representing the sphere, the errors in each component of magnetic measurements is reduced significantly and, as result, the final error in the steering is much smaller. will proide useful information for drillers to more accurately steer towards laterally remoed targets. Inersion of surface magnetic data to characterize source parameters proides the first step towards the remoal of erroneous magnetic fields. Howeer, data from the initial stages of drilling should be used to supplement the inersion deried source parameters and create an een more accurate characterization of erroneous magnetic fields present in the subsurface. Further research is required to better understand how to use such data for the benefit of a drilling program. Acknowledgment This work was partially funded by Graity and Magnetic Research Consortium (GMRC) under the Center for Graity, Electrical and Magnetic Studies (CGEM). The sponsoring companies are Anadarko Petroleum, Cheron Texaco, and ConocoPhillips. We also thank Rich Krahenbuhl, Kris Dais, and Vinicio Sanchez for helpful discussions and assistance. Figure 4 shows the comparison between corrected and uncorrected steering. The dots represent the planned well path, and blue line shows the actual well resulted from the influence of the sphere adjacent to the well. The red line indicates the actual well when the correction was applied based on the inerted dipole field. Relatie errors associated with the corrected case hae a maximum error in dip that is well below the 0.1º requirement. The final drill bit position is at a negligible distance away from the intended target. Discussion This research shows that geologic magnetic bodies are capable of producing magnetic fields large enough to cause significant errors with respect to the required accuracy of the magnetometer measurements. The scenarios presented proide a glimpse into the full complexities of this problem, but they do proide a basis from which to make educated extrapolations about the impacts of geologic magnetic bodies on directional drilling programs. The results of this research point to promising applications of methods deeloped for magnetic exploration to the correction of errors caused by geology in directional drilling when magnetic MWD techniques are used. The ideas and information presented here represent only the beginning of necessary research into the impacts of magnetic geologic bodies on MWD geosteering methods. Analysis of situations such as magnetic bedrock, magnetized stratigraphic layers, and the impact of a magnetic body s position relatie to different stages of a typical directional drilling sequence 666

5 EDITED REFERENCES Note: This reference list is a copy-edited ersion of the reference list submitted by the author. Reference lists for the 2007 SEG Technical Program Expanded Abstracts hae been copy edited so that references proided with the online metadata for each paper will achiee a high degree of linking to cited sources that appear on the Web. REFERENCES Clark, T. D. G., and E. Clarke, 2001, Space weather serices for the offshore drilling industry: British Geological Surey. Gill, P. E., W. Murray, and M. H. Wright, 2004, Practical optimization: Elseier. Li, Y., and D. W. Oldenburg, 2000, Joint inersion of surface and three-component borehole magnetic data: Geophysics, 65, Ngugi, P. K., 2002, Technical, economic and institutional ealuation of adopting directional drilling by Kengen, Kenya: Geothermal Training, Program paper 9,