SPE 63141 Applications of a New Multiple Sensor Production Logging System for Horizontal and Highly-Deviated Multiphase Producers

David Chace, Jianrong Wang, Roman Mirzwinski, Jorge Maxit, and Darryl Trcka, Baker Atlas

Copyright 2000, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2000 SPE Annual Technical Conference and Exhibition held in Dallas, Texas, 14 October 2000. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract This paper focuses on the first applications of an improved, second-generation multiple sensor production logging system designed for use in horizontal and highly deviated multiphase producers. The system integrates several key measurements to provide a comprehensive analysis of well performance under a variety of conditions and flow rates. An expanded and improved 2-dimensional capacitance array is used to define flow regime and measure 3-phase holdup, velocities and flow rates. A new 3-detector pulsed neutron instrument provides an independent measurement of water velocity, 3-phase holdups, and formation water saturation. Auxiliary sensors include an acoustic transducer and distributed temperature sensors useful for gas entry, liquid entry, and behind-casing channel identification. A quartz-pressure gauge measurement is recorded which is also useful in mechanistic models of multiphase flow. While a brief description of the system components will be provided in order to familiarize readers with the measurement concepts, this paper will concentrate on field examples from the Middle East that demonstrate the first use of the improved logging system in horizontal openhole multiphase producing wells. Determination of multiphase holdups, cross-wellbore velocity profiling, and production inflow profiling is demonstrated. Openhole logs are also shown, including resistivity image data, which clearly differentiate the inflow points as producing bed layers or conductive fractures.

complicated flow profiles. Understanding the reservoir performance and behavior, as well as diagnosing poor well performance, requires the ability to reliably measure and monitor the multiphase inflow profile throughout the life of the well. Measuring horizontal multiphase flow is complicated by the effects of gravity segregation of the phases, changes in wellbore inclination due to undulations, changes in flow regime, and associated asymmetries in the phase and velocity distributions across the wellbore. Conventional, single sensor, center-line measuring production logging tools and interpretation methods have therefore been largely unsuccessful in the horizontal environment. The development of new instrumentation and methods has been necessary in order to better determine key flow parameters and conditions, and to adequately measure horizontal flow. In general, these new methods and instrumentation rely on multiple, distributed sensors to provide across-the-wellbore measurements of holdup and velocity distributions, in conjunction with flow regime identification, which is essential for selecting appropriate interpretation models. Measurement interpretation methods have also been expanded to include mechanistic multiphase flow models to help check the quality of measured data and consistency of the interpreted results.

Logging System Components

The described logging system (POLARISsm) comprises 3 major components. The components are briefly described here, including the primary function and purpose with a corresponding acronym for subsequent reference in the text: The Multi Capacitance Flowmeter (MCFMsm) uses a 2dimensional capacitance array calibrated for dielectric measurements to determine 3-phase holdups (phase fractions), across-the-wellbore velocity profiles, and 3phase flow rates and inflow profile. Additional measurements include distributed temperature, quartz pressure, passive acoustic flow noise, and wellbore orientation. The Reservoir Performance Monitor (RPMsm) is a pulsed neutron instrument that provides measurements of formation capture cross-section () or C/O ratio for formation water saturation, 3-phase holdups, and water flow velocity.

Introduction
Horizontal and highly-deviated multiphase producers often traverse large intervals of a reservoir and typically have

CHACE, WANG, MIRZWINSKI, MAXIT, AND TRCKA

SPE 63141

A small-diameter, center-line spinner flowmeter provides a measurement of velocity along the center axis of the wellbore. Figure 1 illustrates the logging string used in the field examples. The tool string has a total length of 80 feet, a diameter of 1-11/16 inches, and a weight of 375 lbs.

The Multi Capacitance Flowmeter

The Multi Capacitance Flowmeter (MCFM) is the result of a joint development project between Baker Atlas and Shell International Exploration and Production (SIEP). The measurement concepts are derived from original work on surface pipeline flow sponsored by SIEP in the early 1990s, which includes the identification of fluids by measured dielectric constants and velocity measurement by crosscorrelation of dielectric signals from sensors distributed across the wellbore1. Sensor and Measurement Concepts The primary sensor package consists of a 2-dimensional array of parallel-plate capacitors arranged along the inside surfaces of a pair of rigid parallel plates 2.6 feet (0.8 meters) long, referred to as the wing, which pivots around its center where it is hinged to the instrument body. Figure 2 illustrates the wing and sensor configuration. In operation, the wing is motored open to span the wellbore and expose the capacitor array to the wellbore fluids. The array measurements provide the two key components required for measuring multi-phase flow rates in horizontal wellbores: Phase fractions (holdup distribution) - The sensors are calibrated on the surface for water, oil, and gas responses so that the measured dielectric constants of the fluid mixture downhole can be used to identify the wellbore fluid types and fractions (holdups) at 8 separate points across the wellbore. Velocity profile Flow velocity at different points across the wellbore can be determined by correlating responses from adjacent sensors within the same row (variations in fluid dielectric) as disturbances flow along the wellbore and through the wing assembly. The velocity profile is constructed using the correlation-based velocities from the wing and the center-line velocity from the spinner. Additionally, the MCFM provides the following: Determination of the flow regime (e.g. stratified, bubble or intermittent) by direct holdup imaging, or by acoustic methods2. Quartzdyne pressure (can be compared with predicted pressure from mechanistic multiphase flow modeling3,4). Channel and fluid entry detection using distributed temperature measurements (top and bottom of the wing) or by acoustic sensing 4,5. Ultimately, the velocity and holdup profiles are combined at the 8 measurement points across the wellbore to determine the multi-phase flow rates and inflow profile.

High resolution measurements and flow images are possible due to the simultaneous measurement of velocity and holdup profiles in the same sensor package. Figure 3 shows one example from nearly 500 flow conditions studied at Shells flow loop in Rijswijk, The Netherlands during prototype testing 4,6. In this 3-phase measurement, the flow loop was inclined at one degree and the metered production was 8 m3/hr water (1207 BPD), 33 m3/hr oil (4981 BPD), and 16 m3/hr gas (13,560 ft3/day). Approximately 35 seconds of data are shown in the figure. Reviewing the data in Figure 3 starting with the bottom track, we have: Track 1 Holdup image - shows the water, oil, and gas holdups in the flow loop test section, constructed by summing the contributions from individual sensors shown in Track 2. Phase distribution profile shows individual sensor responses at eight levels across the flow loop pipe. Slug flow is clearly visible with intermittent gas pockets moving through the flowing liquid. Velocity profile shows the velocity profile across the flow loop pipe, constructed from the three transit time (correlation) velocities on the outer two rows of the wing (Tracks 5, 6, 8, and 9) and spinner velocity (Track 7). Flow rates - the water, oil, and gas flow rates are calculated from the combined phase distribution and velocity profiles. Correlation and spinner velocities shows correlations from 3 spacings on each row, and centerline spinner velocity

Track 2

Track 3

Track 4

Tracks 5-9

Figure 3 provides a comprehensive picture of the 3-phase flow in the inclined pipe. The velocity measurements and the resulting velocity profile show asymmetry in the profile; the holdups show the variation of the holdups in the slug flow; and the resulting flow rate measurements show the oscillations in flow rates in slug flow. Accumulated time averages yield the net flow rates, which in this case are within 10% of the flow loop metered values. Development History of the MCFM The original prototypes have gone through two revisions in which various improvements in the system have been implemented. A brief summary of the original prototype implementation is given below, followed by the key improvements implemented in subsequent upgrades. For more complete descriptions of MCFM measurement concepts, laboratory testing, operation, and field trial examples, the reader is directed to previous publications on the subject4,6.

SPE 63141

APPLICATIONS OF A NEW MULTIPLE SENSOR PRODUCTION LOGGING SYSTEM FOR HORIZONTAL AND HIGHLY-DEVIATED MULTIPHASE PRODUCERS

1st Generation (Research Prototypes) The MCFM prototypes entered field trials for PDO in Oman in late 1997 after comprehensive multi-phase testing in Shells flowloop in Rijswijk, The Netherlands. Holdups 12-bit resolution signals from 8 rows for dielectric measurement and holdups using a single transmitter scheme (response to both conductive and non-conductive fluids). Velocities - 4 rows (2 on top, 2 on bottom of wing) for cross-correlation velocity of pairs of differential capacitance sensor responses (12-bit signals decimated to 1-bit for transmission then correlated at surface). 2nd Generation (Field Prototype Instruments) The upgraded instruments were deployed in mid-1999 with a number of improvements. Holdups The single transmitter scheme has been upgraded to a dual transmitter scheme (two transmitters in quadrature), allowing better separation of the conductive and non-conductive responses for improved water, oil, and gas holdup measurements7. Velocities 4 correlation rows as before. The differential sensor correlation has been supplemented with single sensor correlations for long-spaced sensors (leading and trailing sensors in each row) for an improved measurement of slug front and tail velocities. A downhole data compression scheme was implemented to allow reconstruction of 12-bit signals at surface for correlation (a significant improvement over 1-bit signal correlation). Mechanical Modifications were made to the wing sensor assembly contact points to improve performance in openhole wellbores where debris is more probable. Specifically, new pads were developed to improve movement over rough borehole surfaces. The pads also provided better protection to the sensors on the wing (capacitance and temperature). (Note: The examples in this paper were recorded with the 2nd generation MCFM instruments.) Next-Generation Enhancements The latest revision currently in progress is the implementation of additional capacitance sensor rows for correlation velocities. An additional row is being added on both the upper and lower halves of the wing near the tool body. This will increase the velocity measurement points from 4 to 6 across the wellbore (from 5 to 7 including the center-line spinner measurement). The additional rows are expected to improve performance in larger wellbores (>> 7 inch diameter) and in wells where, due to deviation and/or particular flow conditions, the fluid interface stays close to the center of the wellbore.

Measurement Concepts The 3-detector array is designed to provide optimum spacing for several pulsed neutron measurements, including C/O, pulsed neutron capture (PNC), oxygen activation for water velocity, and 3-phase holdups. Another benefit of the multiple detector configuration is that it allows for simultaneous acquisition of PNC () and oxygen activation water flow data. Figure 4 illustrates the RPM configuration in water flow mode with water flow along the wellbore. The RPM provides the following measurements in multiphase producers: Formation water saturation from PNC () or C/O to help determine the source of produced fluids. Water velocity from oxygen activation8,9 to provide an independent measurement of water velocity in multiphase flow rate calculations. Also useful for detecting the presence of water channeling behind pipe or for monitoring water flow and velocity in annular regions. 3-phase holdups from C/O and other inelastic gamma measurements (a measurement jointly developed with Shell 10,11). When combined with the MCFM holdup measurement (which is only sensitive to flow inside the casing), this measurement can be useful in detecting the presence of gas, oil, or water behind pipe or in the annulus. For a more complete description of RPM, 3-phase holdup, and water flow measurement concepts, laboratory testing, operation, and field trial examples, the reader is directed to previous publications 8,9,10,11,12,13.

Field Examples
The following three field examples are from wells in the Middle East. In all cases, the wells are openhole producers that were logged with the POLARIS system conveyed on 11/2 inch coiled tubing. Field Example 1 Oil producer with 40 % water cut and water producing fractures The first well is an oil/water producer that has approximately 2300 ft of horizontal 6 1/8 hole through a carbonate reservoir as a openhole completion. The hole was drilled slanted through the reservoir with the deviation more or less constant at 83 degrees and never reaches 90 degrees. The well is producing at about 40 percent water cut and the bulk of water was found to be coming from a fractured zone. Flow Profile Results The flow profile results are shown in Fig.5 and the following table summarizes the measured and surface flow rates:

The Reservoir Performance Monitor

The Reservoir Performance Monitor (RPM) is a new pulsed neutron instrument that utilizes a 3-detector array and a multifrequency neutron generator to allow multiple modes of operation.

CHACE, WANG, MIRZWINSKI, MAXIT, AND TRCKA

SPE 63141

POLARIS Downhole Surface Total (bpd) Oil (bopd) Water (bwpd) Water Cut (%) 7310 5089 2221 30.4 6164 3925 2221 36.1

Well Test While 2 mos. Logging before 5-7000 7830 3-4000 2-3000 40 4749 3081 39.3

Surface flowrates computed from the logged flow profile are in reasonably good agreement with the test trap averages. Surface measured rates during the logging passes were not as stable as in other logging operations, and in part show that flow rates can be affected due the presence of coiled tubing in the wellbore. The following is a summary of the major conclusions drawn from the POLARIS (MCFM / RPM), open hole resistivity image and lithology data: Most oil and water flow is from a fractured dolomitized interval. Additional oil entries are indicated at several points throughout the well below the fractured interval. Water is not produced from below the fractured interval, however, the water in the sump below the fractured interval is circulating due to the movement of oil along the top side of the hole. In Fig.5, the holdup and across-the-wellbore velocity profiles show that the water in the sump is static with oil moving along the top side of the wellbore. At the major fluid entry point, where the holdups show significantly more oil, the velocity profile shows positive velocities across the entire wellbore, indicating that water is being produced to the surface. The temperature curve superimposed on velocity profile responds to oil entries with a slight decrease and an increase at the water entry points. In Fig.6, details of the phase distribution profile (oil and water distribution for at each of the 8 sensor rows across the wellbore) and the individual velocity profile measurements from correlations and spinner are shown. In Fig.7, the resistivity image and lithology tracks have been added, with focus on the interval where water production was detected. The temperature data shows a marked increase in the interval of flow rate increases due to the entry of warmer water with the oil entries. The fractured dolomitized section in this interval is marked with a red bar in the figure. Bedding planes can be seen in the resistivity image. In Fig.8, the scale has been expanded so that the vertical fractures (relatively straight lines on the resistivity map) can be identified amidst the more horizontal bedding planes (sinusoidal bands on the images). The combined flow and

openhole images allow a clear determination that the fractures are the major water entry points in this well. In Fig.9, taken from the interval in the water sump, the holdup image shows oil entry in 2 major intervals and an accumulation of the oil along the top of the wellbore. Note that the oil entry is seen by the MCFM along the entire perimeter of the horizontal bedding plane interface intersected by the slanted wellbore (compare the MCFM holdup and resistivity images). The entry point is spread out along the wellbore due to the angle at which the wellbore intersects the bedding plane. No water is produced in the sump, but the water column is circulating due to the movement of oil along the top of the hole. Water is essentially dragged along by the oil. Note that the spinner in this interval showed slight indications of downflow (in the center of the borehole) under the upward flowing oil. In addition, the RPM oxygen activation log data shown in Fig.10 indicates slight upward water velocity at the stationary measurement points taken in the interval. However, the count rates associated with the water velocity are low and indicate that only a small volume of water is moving upward in this interval. All of the evidence points to a water sump that is circulating due to the oil production and flow along the high side of the hole. Field Example 2 Oil producer with 8 % water cut The second example is a two-phase oil/water producer that has approximately 2200 ft of horizontal 6 1/8 hole through a carbonate reservoir as an openhole completion. The hole was drilled slanted through the reservoir with the deviation more or less constant at 81 degrees with no sumps or traps along its length. Flow Profile Results The flow profile results are shown in Fig.11 and the following table summarizes the measured and surface flow rates: POLARIS Downhole Surface 10246 8034 9672 574 5.6 7460 574 7.1 Well Test Surface 7455 6826 629 8.4

Total (bpd) Oil (bopd) Water (bwpd) Water Cut (%)

% Diff. + 7.8 + 9.3 - 8.7 - 1.3

Surface flowrates computed from the logged flow profile are in good agreement with the test trap averages. Oil, water, and total flow rates are within 10 percent, and water cut is within 1.5 percent. The profile looks in many ways similar to the previous example; however, there are a few notable differences. Here are some of the key observations:

SPE 63141

APPLICATIONS OF A NEW MULTIPLE SENSOR PRODUCTION LOGGING SYSTEM FOR HORIZONTAL AND HIGHLY-DEVIATED MULTIPHASE PRODUCERS

The large inflows correlate with zones of high porosity; however, no image logs were run in this well, so it is not possible to say with any certainty whether these inflows are fracture related. This well has a very long water sump with no fluid flow (bottom 600 feet). This sump interval includes a normally productive zone. This lack of production may either be grounds for stimulation, or it may lead to revised new well trajectories in this part of the field. Water shut-off could involve sealing off the bottom zones near the water sump where oil entry is minimal. Water entry in the upper intervals would be more difficult to isolate since it appears to be co-mingled with the major oil entries.

Whereas few correlations were detected on the bottom of the wing in Example 1, the correlation velocities were stronger in the turbulent flow near the heel in total flow where the water / oil mixture layers dropped sufficiently to give strong events on row 2. Holdups and phase distribution are affected by inclination angle and flow rates, and in this case a significantly lower water cut and high flow turbulence in the heel provided good conditions for velocity measurement at all points across the wellbore. The water flow profile was constructed from the continuous RPM oxygen activation passes at 20 and 30 ft/min with stationary measurements (not shown) providing data over the lower part of the profile where the water velocity was less than the practical logging speed. Modeled results The interpreted results were compared with flow conditions predicted by Shells pressure-based mechanistic two-phase flow modeling software. The Shell software uses modeling originally designed for use in surface pipelines, and is readily applicable to downhole conditions in horizontal wells. Essentially, the model takes input such as the inflow profile, fluid parameters, PVT data, inclination, and other information to predict water holdup, pressure gradient, and other flow data along the wellbore 3,4. The model was applied to the measured flow rates in this well and a predicted water holdup was computed. A comparison of the measured and theoretical water holdups is shown in Fig.11 and shows good agreement. This lends confidence that the interpreted results are reasonable and consistent. Stationary measurements The lower intervals are producing mainly oil in elongated bubbles (slugs) on the high side of the hole over a standing water column. Fifteen stationary readings were made to verify the low oil and water flow rates. The stationary readings illustrate that rows 7 and 8 are in the oil flow and exhibit good velocity correlations, whereas the lower rows 1 and 2 are immersed in water and show none due to a continuous water phase. The flowmeter registers a

slight downflow in the water at the center of the wellbore as each oil slug displaces some water in the downhole direction. The stationary reading shown in Fig.12 shows oil slugs passing by the wing every 10 to 15 seconds with good row 7/8 velocity correlations and the flowmeter again responding to the water downflow with every slug. An oil flowrate was computed from a time average over the 300 seconds of each station. Fig.12 also shows a stationary measurement from a deeper depth where the oil slugs have not fully developed and are smaller in size and higher in frequency. The stationary measurements support give confidence to the profile at low flow rates over this region of elongated oil bubble (oil slug) flow. Water flow stationary measurements indicate no water flow in the sump below the first fluid entry point. As observed in the first example well, the water velocity in the interval above the first fluid entry point corresponds with oil slug flow and suggests that while some water is being produced, water in the wellbore is also being dragged along with the oil slugs. Concurrently, the spinner shows some indications of downflow in the presence of upflow on the oxygen activation measurements. This again supports the scenario that circulation of the water column is occurring. Shut-in pass The well was shut-in for approximately 4.5 hours prior to logging. The data is shown in Fig.13. An oil/water contact can be seen in the wellbore. The spinner (not shown) indicated no water cross-flow, however, slight oil flow was observed traveling on top of the static water column from the lowest entry points indicating that the well had not reached a stable shut-in condition. The oil slugs moving upward were detected on rows 7 and 8 and indicated a flow rate of less than 150 b/d. Typical shut-in times of 24 hours are required to obtain a stable shut-in condition. Field Example 3 - Continuous water flow log in an undulating horizontal producer (70 % water cut) The example well is an openhole producer with approximately 2250 feet of horizontal 6 1/8 hole. This well is quite different than the first two examples in that rather than being drilled as a slant hole through all of the major producing zones in the reservoir, it was drilled as a true horizontal well to target a single producing zone within the reservoir. In addition, the production rate is relatively low in comparison to the first two examples. The well undulates above and below 90 degree deviation which makes it an interesting example with regard to the flow characteristics associated with these types of wells. The undulations create traps and sumps where lighter and heavier phases, respectively, can accumulate, and which can create additional pressure drops along the horizontal section. In addition, the change in inclination from positive (uphill) to negative (downhill), relative to the flow, can cause dramatic changes in phase holdups and velocities.

CHACE, WANG, MIRZWINSKI, MAXIT, AND TRCKA

SPE 63141

For the purpose of this discussion, the focus will be on the water flow profile determined from the MCFM water holdup measurement and the RPM real-time continuous oxygen activation water velocity measurement. Specifically, the data in Fig.14 illustrates the effects on holdup and velocities caused by the undulating wellbore. For reference, the measured flow rates from the POLARIS logging data are shown in the following table. The flow profile (not shown) indicates that nearly all of the oil and water production is from the toe, so the flow rate is relatively constant along the entire interval: POLARIS Downhole Surface 2060 1904 650 1410 68.4 494 1410 74.1 Well Test Surface 1732 436 1296 74.8

Total (bpd) Oil (bopd) Water (bwpd) Water Cut (%)

% Diff. + 10 +13.3 + 8.8

would have been helpful as the relative flow is still too fast (with reduced holdup and volume) and the velocity is somewhat underestimated in the downhill sections. In the current instrument version, an ultra-long spaced utilizing the gamma ray detector at the top of the tool is used for faster water velocities. The spinner (not shown) was also affected by wellbore inclination changes. The spinner and oxygen activation measurements respond similarly in the uphill sections where water nearly fills the wellbore, but show quite different responses in the downhill sections. In these intervals, the oxygen activation responds to faster water velocity and the spinner decreases significantly (or stops) in the slowly moving oil which occupies most of the wellbore space. In some cases, the water holdup can be high enough to touch the spinner in the center of the wellbore, and the response can become a complicated combination of the two phase velocities.

Summary
- 0.7 This paper has summarized the key components and functionality of a new multiple sensor production logging system designed for use in horizontal and highly-deviated multiphase producers. The field examples highlight the first applications of the 2nd generation instrumentation that incorporates a number of significant improvements over the original prototype design. The system integrates several key measurements to provide a comprehensive analysis of well performance under a variety of conditions and flow rates: An improved 2-dimensional capacitance array is used to define flow regime and measure 3-phase holdup, velocities and flow rates. A new 3-detector pulsed neutron instrument provides an independent measurement of water velocity, 3-phase holdups, and formation water saturation. Auxiliary sensors include an acoustic transducer and distributed temperature sensors useful for gas entry, liquid entry, and behind-casing channel identification. A quartz-pressure gauge measurement allows comparison with pressure-based mechanistic multiphase flow models. The system improvements include: A new dual transmitter scheme for the capacitance sensor measurements which allows better separation of the capacitive and conductive responses, thereby improving measurement of water, oil, and gas holdups. A differential measurement system for cross-correlation velocities that now incorporates data compression to maximize sensitivity and resolution, thereby improving the correlation velocity measurements. Implementation of long-spaced cross-correlations of single sensor capacitance responses for improved slug front and tail velocity measurements. Implementation of a real-time water velocity calculation based on oxygen activation measurements with a

A brief summary of the data shown in Fig.14 will be helpful for our discussion: Track 1 Track 2 Track 3 Sigma, GR, and caliper OA2 and OA3, the oxygen activation counts at detectors 2 and 3 Real-time water velocity measurement compared with the coiled tubing logging speed (continuous and stationary meas.) MCFM water holdup and wellbore trajectory showing oil and water holdup in the wellbore Water rate from the MCFM holdup and RPM water velocity (continuous and stationary measurements)

Track 4 Track 5

The water flow profile was determined from the MCFM water holdup measurement and the RPM oxygen activation passes at 12 and 50 ft/min and supplemented with stationary measurements over the lowest part of the interval where the water velocity was less than the minimum practical coiled tubing logging speed. Water flow from the toe of the well is detected by the stationary measurements nearest the toe, and a second water entry is indicated by the increase in water velocity about 100 feet above in the interval where the continuous and stationary measurements overlap. Water holdup and velocity were affected by the wellbore inclination with increases in velocity and decreases in holdup observed in the downhill sections near the three sumps in the well. (Note that the oil holdup and velocity respond in the opposite way in these sections.) Faster (50ft/min) logging passes were used over these intervals to reduce the relative water velocity (to the instrument) and improve the measurement statistics. Indeed, an even faster logging speed

SPE 63141

APPLICATIONS OF A NEW MULTIPLE SENSOR PRODUCTION LOGGING SYSTEM FOR HORIZONTAL AND HIGHLY-DEVIATED MULTIPHASE PRODUCERS

simultaneous PNC () measurement for formation water saturation. Upgrades in progress include: Addition of capacitance sensor rows for correlation-based velocity measurements across the wellbore. Better isolation of the wing-tip temperature probes for improved thermal response. Field examples from three wells were presented that demonstrate: Use of the new logging system in highly-deviated (slant) and undulating horizontal openhole multiphase producing wells. Detailed determination of multiphase holdups, across-thewellbore velocity profiling, fluid entry points, and production inflow profiling. Good agreement between measured and well test flow rate data. The effects of changing inclination in undulating wellbores on the holdups and velocities of the phases present in the multiphase flow, as shown in the continuous oxygen activation log example. The presence of static, yet circulating, water sumps caused by the flow of oil along the top side of the wellbore. Corresponding openhole logs (lilthology, porosity, acoustic and resistivity images) which clearly differentiate the inflow points as producing bed layers or conductive fractures. The use of multiphase flow modeling for comparison of theoretical and measured water holdups. Additionally, experience from logging campaigns thus far indicates that the new integrated multiple-sensor production logging system provides measurements with the degree of completeness, accuracy, and repeatability required for effective long-term monitoring of horizontal wellbore and reservoir performance.

ETCE/OMAE 2000 Joint Conference, New Orleans, LA, Paper PROD-10020, Feb.14-17, 2000. 6. Chace, D.., Trcka, D.., Georgi, D., Wang, J., Bousche, O., van der Spek, A., al Nasser, H., New instrumentation and methods for production logging in multiphase horizontal wells, 1999 SPE MEOS, Bahrain, Paper 53220. 7. Maxit, J., Reittinger, P., Wang, J., Kostelnicek, D., Downhole instrumentation for the measurement of three-phase volume fractions and phase velocities in horizontal wells, ETCE/OMAE 2000 Joint Conference, New Orleans, LA, Paper PROD-10030, Feb.14-17, 2000. 8. Trcka, D.E., and Chace, D.M., 1993, Improved method for measuring annular water flow in injection wells using continuous oxygen activation logging, SPE 68th Annual Technical Conference and Exhibition, Houston, Texas, Paper 26450. 9. Chace, D.M., Trcka, D.E., and Dawe, B.A., 1994, Application and interpretation of continuous oxygen activation logs for measuring complex water flow profiles in injection wells, SPE 69th Annual Technical Conference and Exhibition, New Orleans, Louisiana, Paper 28412. 10. Peeters, M., Oliver, D.W., Wright, G.A., 1994, Pulsed neutron tool applied to three-phase production logging in horizontal wells, SPWLA 35th Annual Logging Symposium, Paper L. 11. Trcka, D., van den Berg, F., Manan, W.A., Peeters, M., and Mickael, M., 1996, Measuring three-phase holdups in horizontal wellbores using pulsed neutron instruments, SPE Annual Technical Conference and Exhibition, Denver, Colorado, Paper 36561 12. Gilchrist, W.A. Jr., Prati, E., Pemeper, R., Mickael, M.W., and Trcka, D., Introduction of a New Through-Tubing Multifunction Pulsed Neutron Instrument, SPE paper 56803, Annual SPE Technical Conference and Exhibition, Houston, Texas, Oct. 1999. 13. Trcka, D., Oliver, D., Mickael M., Chace, D., Georgi, D., and Wang, J., Reservoir performance monitoring in the new millennium, Proceedings of the Tunisian Oil Conference, April, 2000

References
1.Den Boer, Johannis Josephus, Inventor, 1992, Method and apparatus to measure multiphase flow properties, European Patent Application No. 0 510 774 A2, Shell Internationale Research Maatschappij B.V. 2. Van der Spek, A. and Thomas, A., 1998, "Neural Net Identification of Flow Regime Using Band Spectra of Flow Generated Sound," SPE 50640, 1998 SPE European Petroleum Conference, Hague, Netherlands, Oct. 20-22., 1998. 3. Van der Spek, A., and Veran, Y., Horizontal well production log evaluation in multiphase flow, ETCE/OMAE 2000 Joint Conference, New Orleans, LA, Paper PR-032, Feb.14-17, 2000. 4. Chace, D.M., Trcka, D.E., Wang, J., Bousche, O., van der Spek, A., al Nasser, H., A new production logging service for horizontal wells, SPWLA 39th Annual Symposium, Keystone, CO, 1998, Paper H. 5. Wang, J., Georgi, D., Chace, D., Maxit, J., New method for detection of channeling behind casing in producing wells,

CHACE, WANG, MIRZWINSKI, MAXIT, AND TRCKA

SPE 63141

CTU head - Coil Tubing Pressure Deployment Bar 3913 Swivel 8250 Power Supply 8248 Telemetry / CCL

8281 Reservoir Performance Monitor Tool XLS Detector LS Detector SS Detector Pulsed Neutron Source

8262 Gamma Ray 8221 Temp 8260 Centralizer 8291 POS Powered Swivel 8260 Centralizer

8290 MCFM Multi-Capacitance Flowmeter

8260 Centralizer 8258 Spinner Flowmeter

Figure 1. The POLARIS logging string.

SPE 63141

APPLICATIONS OF A NEW MULTIPLE SENSOR PRODUCTION LOGGING SYSTEM FOR HORIZONTAL AND HIGHLY-DEVIATED MULTIPHASE PRODUCERS

Figure 2. The MCFM wing spans the borehole exposing the capacitance sensors (red rectangles) to the flow. Holdup and liquid level are determined from the capacitance measurements at 8 levels. Bi-directional velocities are determined from transit times of disturbances flowing through the sensor rows at levels above and below the tool body. Water bubbles in oil/water flow are shown on the left, and 3-phase slug flow with water fall-back is shown on the right.

Figure 3 This example of 3-phase flow data from the SIEP flow loop at an inclination of + 1 degree shows the oscillation of holdups and flow rates characteristic of slug flow. Water (blue), oil (green), gas (red). Rotating this image 90 degrees facilitates visualization of the flow in the horizontal test section (flow from right to left).

M CFM Flow Des criptio n

T im e ( se c) P ha se H oldu ps P ha se D istr ibution V elo city P r ofile F lo w R ates

C orrela tion and Sp inn er V elo cities

R ow 1 R ow 2 S pin ner R ow 7 R ow 8

Figure 4. Oxygen activation measurement from RPM As water molecules move past the neutron source, the oxygen becomes activated (blue O16 changes to orange N16 in area shown as orange ellipse). The N16 decays back to O16 with a half-life of 7.13 seconds and emits gamma radiation that is detected at the various detectors in the tool. The water velocity is a function the ratio of count rates at the detectors, the spacing between the detectors, and the decay half-life.

W a ter

O il G as

10

CHACE, WANG, MIRZWINSKI, MAXIT, AND TRCKA

VELOCITY

SPE 63141

CORREL

Figure 5 Example 1 Well path, holdup and velocity profiles, flow profile, and temperature for an 83 degree horizontal oil/water producer

Figure 6 Example 1 Details of holdup and velocity profiles

SPE 63141

APPLICATIONS OF A NEW MULTIPLE SENSOR PRODUCTION LOGGING SYSTEM FOR HORIZONTAL AND HIGHLY-DEVIATED MULTIPHASE PRODUCERS

11

Figure 7 Example 1 - Fractured dolomitized limestone interval producing majority of water

Figure 8 - Example 1 Zoom in on fractured interval showing near-vertical fractures as source of water production

12

CHACE, WANG, MIRZWINSKI, MAXIT, AND TRCKA

SPE 63141

Figure 9 Example 1 Oil entry in lower section with static water sump in well, showing oil entry from bedding planes and segregation to top of borehole

Figure 10 Example 1 RPM oxygen activation log data confirming circulating water sump

SPE 63141

APPLICATIONS OF A NEW MULTIPLE SENSOR PRODUCTION LOGGING SYSTEM FOR HORIZONTAL AND HIGHLY-DEVIATED MULTIPHASE PRODUCERS

13

VEL

Figure 11 Example 2 Well path, holdup and velocity profile, flow profile, and temperature for an 81 degree horizontal oil/water producer, showing that deepest 600 feet is non-productive. Also shown in the far right track is a comparison of the measured and modeled water holdups. The good agreement confirms that the interpreted results are consistent with Shells pressure-based model.

Figure 12 Example 2 Stationary MCFM measurements at points A and B in Fig. 11 showing details in flow characteristics as oil slug development increases along the wellbore

14

CHACE, WANG, MIRZWINSKI, MAXIT, AND TRCKA

SPE 63141

Figure 13 Example 2 Shut-in pass indicates slight oil flow along top of wellbore. Shut-in time was approximately 4.5 hours.

Figure 14 Example 3 MCFM water holdup and RPM oxygen activation log data showing effects on water holdup and velocity due to changing deviation in an undulating horizontal wellbore. Most water in this well was determined to be coming from the toe and a second point within 100 feet of the toe.