OA17024A - Multiphase meter to provide data for production management. - Google Patents

Abstract

A method and apparatus for determining a volume of a phase of a multiphase fluid flowing in a production tubular is provided. A magnetic field is imparted on the fluid to align nuclei of the multiphase fluid along a direction of the magnetic field. A radio frequency signal is transmitted into the multiphase fluid to excite the nuclei, and a signal is detected from the nuclei responsive to the transmitted radio frequency signal. An amplitude of the detected signal is determined and the volume of the phase flowing in the production tubular is determined using the determined amplitude and an amplitude of a calibration signal.

Description

MULTIPHASE METER TO PROVIDE DATA FOR PRODUCTION MANAGEMENT

CROSS REFERENCE TO RELATED APPLICATIONS

This application daims the benefît of U.S. Application No. 13/028553, fîlcd on February 16, 2011, which is incorporated herein by référencé in its entirety.

BACKGROUND OF THE DISCLOSURE

1. Fîeld ofthe Disclosure [0001] The présent disclosure relates gcncrally to a mcasurcmcnt apparatus and methods for estimating downhole fluid characteristics.

2. Description ofthe Related Art [0002] In the oil and gas industry it has become increasingly important in recent years to obtain measurcmcnts of the flow rate and phase ratio of multiphase fluids such as those produced by drilling operations and the compositions ofthe downhole fluid.

[0003] In order to measure the flow rate and ratio properties ofsuch multiphase fluids accurately enough to satisfy the operatoris requirements it îs currently known to use techniques such as Nuclear Magnetic Measuremcnt (NMR) and Electronic Spin Résonance (ESR) analysis. Howcver, currently available Systems for measuring such properties using these techniques rcquirc a number of separate components which cmploy a varicty of operational and analytical techniques and often involve a number of discrète devices each adapted to measure a particular property of the fluid flow. For example a device for detecting the fraction of one phase may be supplied along with a device for detecting the fraction of another phase and another device to measure the overall flow rate. Also, such techniques are generally not utilized for composition^ analysis of hydrocarbons in downhole fluids.

SUMMARY OF THE DISCLOSURE [0004] In one aspect, the présent disclosure provides a method of determining a volume of a phase ofa multiphase fluid flowîng in a tubular, including: imparting a magnetic fîeld on the fluid to align nuclci of the multiphase fluid along a direction ofthe magnetic fîeld; transmitting a radio frequency signal into the multiphase fluid to excite the nuclei; detecting a signal from the nuclci responsive to the transmîtted radio frequency signal; determining an amplitude ofthe dctected signal; and determining the volume ofthe phase flowîng in the tubular using the determined amplitude and an amplitude of a calibration signal.

» ’ [0005] In another aspect, the présent disciosure provides an apparatus for determining a volume of a phase of a multiphase fluid flowing in a tubular, the apparatus including a source confîgured to impart a primary magnetic field on the fluid to align nuclei of the multiphase fluid along a direction of the primary magnetic field; a source confîgured to transmit a radio frequency signal into the multiphase fluid to excite the nuclei; a detector for detecting a signal from the nuclei responsive to the transmitted radio frequency signal; and a processor confîgured to déterminé an amplitude of the detected signal and the volume of the phase flowing in the tubular using the determined amplitude of the detected signais and an amplitude of a calibration signal.

[0006] In yet another aspect, the présent disciosure provides a method of determining stability of an émulsion flowing in a production string, the method including imparting a primary magnetic field on the émulsion to align nuclei of the émulsion along a direction of the primary magnetic field; transmitting a radio frequency signal into the émulsion flowing in the production string; detecting a signal from the nuclei of the émulsion responsive to the transmitted radio frequency signal; determining an amplitude of the detected signal; determining a water eut of the émulsion using the obtained amplitude; determining a relaxation rate of a signal obtain from nuclei of the émulsion excited in response to the transmitted radio frequency signal; obtaining a viscosity of the émulsion from the determined relaxation time; and determining the stability of the émulsion from the determined émulsion viscosity and the water eut of the émulsion.

[0007] Examples of the more important features ofthe methods and apparatus for analyzing the composition of a hydrocarbon hâve been summarized rather broadly in order that the detailed description thereof that follows may be better understood and in order that the contributions they represent to the art may be appreciated. Thcre are, of course, additional features of the methods and apparatus that are described hereinafter and which will fbrm the subject of any daims that may be made pursuant to this disciosure.

BRIEF DESCRIPTION OF THE DRAWINGS [0008] For detailed understanding ofthe apparatus and methods for compositions! analysis of hydrocarbons in a downhole fluid, référencé should be made to the following detailed description, taken in conjunction with the accompanyïng drawing, in which like éléments arc generally designated by like numcrals, and in which:

FIG. I is a transverse side view of one embodiment of the apparatus according to the présent disciosure;

FIG. 1 a is a transverse side view of the apparatus of FIG. 1 showing magnetic gradient coils which act in the direction of the z-axis with respect to the référencé axes indicated on FIG. 1;

FIG. 2 is a transverse side view of the apparatus of FIG. 1 showing magnetic gradient coils which act in the direction of the y-axis with respect to the référence axes indicated on FIG. 1;

FIG. 3 is a cross sectional viewof the apparatus of FIG. 1 showing the components of the magnetic gradient coils which act in the x, y and z directions with respect to the rcfcrcncc axes indicated on FIG. 1;

FIG- 4 is a schematic view of the component of the gradient coils which act in the zaxis direction with respect to the référencé axes of FIG. I, arranged around combined transmission and réception coils in accordance with a particular embodiment of the disclosure;

FIG. 5 is a schematic view of the gradient coils which act in the y-axis direction with respect to the référencé axes of FIG. 1, arranged around combined transmission and réception coils in accordance with one embodiment of the présent disclosure;

FIG. 6 is an illustration of a magnetic ficld orientation in order to produce the homogcncous magnet used in accordance with the présent disclosure;

FIG. 7a is a schematic circuit diagram showing the interaction between the various components of the recciving circuit of the combined rcceîving and transmission coils;

FIG. 7b is a schematic circuit diagram showing the interaction between the various components of the transmitting circuit of the combined rcceîving and transmission coils;

FIG. 8 is a schematic cross sectional diagram of the primary magnet composition used in accordance with the présent disclosure;

FIG. 9 is a transverse side view of another embodiment of the apparatus according to the présent invention without the gradient and transmission coils shown;

FIG. 10 is a schematic perspective view of the magnet configuration used in the apparatus of FIG. 9;

FIG. 11 is a transverse side view of the apparatus of FIG. 9 showing the gradient and 30 transmission coils;

FIG. 12 is a schematic view of the component ofthe gradient coils of FIG. 10 which act in the z-axis direction;

FIG. 13 is a schematic view of the component ofthe gradient coils of FIG. 10 which act in the x and y-axis directions;

t

FIG. 14 is a functional diagram of a circuit for performing compositional analysis of hydrocaibons according to one embodiment of the disclosure;

FIG. 15 is a visual représentation of the relationshlp between certain specîes of a hydrocarbon and the frequency shift (also refened to as chemical shift) between an imparted 5 radio frequency signal on a fluid and a detected radiofrequency signal from the fluid;

FIG. 16 is a frequency shift spectrum of certain specîes of a hydrocarbon;

FIG. 17 shows an exemplary Nuctear Magnetic Résonance (NMR) flow meter device for estimating a volume of a fluid phase using the exemplary methods of the présent disclosure; and

FIG. 18 shows an exemplary relation between émulsion viscosity and watercut or volume fraction of water in the émulsion.

DETAILED DESCRIPTION OF THE DISCLOSURE [0009] Referring to FIG. 1 the apparatus 10 in accordance with the first embodiment of the présent invention comprises an outer housing 12 which surrounds a section of a fluid 15 flow pipe 14, such as production tubing, by locking thereto via a suitable lockîng mechanism. Inside the housing 12 is locatcd a primary permanent magnet 16 in an outermost recess 18 and a sccondary clcctromagnet housing 20 locatcd in an innermost recess 22. The electromagnet housing 20 has located within it an electromagnet 21 which comprises electromagnet coils Gx, Gy and (as shown in FIG. la) Gz. Combined transmission and réception coils 24 are also provided within the inner diameter of the electromagnet housing 20.

[0010] Outer housing 12 provides magnetic shielding which substantially minimizes leakage of magnetic field outsîde the apparatus 10, and provides safe handling ofthetool This also improves the signal transmission and réception performance of the coils 24 by minimizing interférence from surrounding radio signais such as FM radio signais. Housing 25 12, in the présent embodiment, comprises tow permeability iron, (typically μτ<1.00) which provides the main outer body of the apparatus. The material is typically around 10 mm thick around the mid portion of the apparatus 10 and thickertoward the ends ofthe apparatus 10, typically up to a thickness of around 60 mm. The skillcd reader will realize that different thickncss and material may be used in the housing 12 in order to suit the particular application.

[0011] Referring particularly to FIGS. 6 and 8, the primary permanent magnet 16 comprises a number of conccntrically arranged magnetic cells 26 which are stacked together.

c

Each magnctic cell 26 comprises a number of outer segments 28 (FIG. 8) arranged adjacent a number of inner segments 30 such that a circumfercntial band of inner segments 30 are arranged within a circumferential band of outer segments 28. Fiat plates 32 are positioned between the circumferential band of outer segments 28 and the circumferential band of inner segments 30 such that a circumferential band of plates 32 is located between the outer segments 28 and the inner segments 30. The plates 32 are typically formed of an iron based material having a permeability ofgreater than 1000.

[0012] Apcrturc 34 is providcd in the centre of each cell 26 to allow the flow of fluid thercthrough as will be discussed subsequently. When the cells 28 are stacked together they 10 form a throughbore 36 (as shown in FIG. 6) afong the length of the magnet 16. The iron plates 32 ensure that the résultant magnetic field produeed by inner segments 30 and outer segments 28 îs focused toward the center of the aperture 34 of each cell and hencc along the throughbore 36 of the apparatus 10.

[0013] The skilled reader will understand that the term permanent magnet in this

COntext is taken to mcan a magnet which provïdes a constant magnctic field wiihout requiring, for example, an clectric current in order to create the magnetic field. In an alternative embodiment, the permanent magnet may be an electromagnet which provides a continuons and substantially homogcncous magnctic field.

[0014] The direction ofthe magnetic field vectors (indicated by MF in FIG. 6) of each outer 28 and inner 30 segment is carefiilly arranged during manufacture in order to create a résultant magnetic field for the magnet 16 which is as close to being homogeneous as possible throughout the throughbore ofthe magnet 16. This ensures that the magnetic field présent within the throughbore 36 ofthe magnet 16 remains consistent within the throughbore 36 irrespective of the location within the throughbore 36 that the magnetic field is expericnccd. Typically, the required homogeneity is in the région of around 1.0 ppm. This ensures accurate measurements are possible using the apparatus 10 in conjunction with the NMR techniques as will be discussed subsequently.

[0015] The secondary clcctromagnct housing 20 is providcd with a combincd transmission and réception coil 24 which is capable of both transmitting a radio frequency 3Q puise and dctecting the radio frequency emitted by nuclei excitcd by such a radio frequency puise. In the embodiment shown in the Figures, the coil 24 comprises a pair of circular loops 24a at the top and bottom of the coil 24 connectcd by circumferentially spaced connecting coils 24b to form a '’birdcage configuration. This provides the apparatus 10 with the ability to both transmit a radio frequency puise cvenly throughout the throughbore 36 and t

β competent ly dctect radio frequency signais emitted by nuclei at any location within the throughbore 36 of the apparatus 10. Rather than a birdcage*' configuration the coils may altematively be arranged to provide a saddle coir configuration depending upon the application.

₅ [0016] Referring to FIG. 7a, the receiver circuit 40 of the combined transmission and réception coils 24 comprises a refcrence signal input generator 42 and a 90° phase shifter 44 connected to a standard amplification and filtering system 46 in order to provide a real and imaginary output signal as a rcsult of the signal rcccivçd from the coit 24. Referring to FIG. 7b, the transmitter circuit 48 of the combined transmission and réception coils 24 comprises a signal generator input module 50 and an oscillator 52 which are linked to an amp 54 and a puise programmer 56 in order to transmit the required radio frequency through coil 24. Though illustrated separatcly in FIGS. 7a and 7b, it will be understood that these circuits may be combined or integrated in order to provide the required transmission and réception capability of combined transmission and réception coils 24.

[0017].The secondary electromagnet housing 20 provides the magnetic gradient using coils Gx, Gy, and Gz which selectively (depending upon whether the electromagnet is on or off) provide a graduated. magnetic fietd within the throughbore 36 of the apparatus in the x, y, and z directions rcspcctivcly indicatcd by the référence axes R in FIG. I. This arrangement provides the graduated magnetic fietd required by the flow rate calculation process described subsequcntly.

[0018] The profile of both the primary permanent magnet 16 and the secondary electromagnet 20 are arranged in the présent embodiment, such that they can be housed within the outermost recess 18 and innermost recess 22 respectively in order to maintain a consistent diameter of throughbore 36 through the apparatus 10 such that disturbance of the 25 fluid flowing from the pipe 14 through the apparatus 10 is minimized.

[0019] A second embodiment ofthe présent invention having a number of modifications will now be described. Many componcnts ofthe second embodiment are the same as thosc described in relation to the first embodiment. Such components will not be described any fiirther. In addition, a number of components in the second embodiment correspond to similar components previously described in relation to the first embodiment, and where this apptîes, similar refcrence numérale will be used.

[0020] Referring to FIGS. 9 to 13, the apparatus 100 in accordance with the second embodiment of the présent invention comprises an outer housing 120 surrounding a primary magnet 160. Primary magnet 160 has an inner ring 160A and an outer ring 160B. A

I sccondary clcctromagnct is providcd in housing 215 as discussed subscquently. Transmîssion/reception coil housing 205 is provided on the internai bore of the apparatus 100. The housing 205 may be made of a material such as Poly-Ethcr-Ether-Ketone (PEEK) or a nickel alloy such as Inconel®. The required pressure rating using (PEEK) is generally achievcd using a housing 205 having a very thick wall (in the région of 20 mm). Such a wall generally dégradés the magnetic ficld strength at the ccnter of the flow path since magnet strcngth dccrcascs with radial distance from the magnet. The thickness required using InconcLRTM. is much lcss (in the région of 7 mm). In addition, the use of Inconel® (which has permeability comparable with free space (μτ=Ι)), concentrâtes the magnetic ficld into the 10 flow path, thereby increasing the magnetic strength homogeneity.

[0021] The housing 205 in the présent embodiment is providcd with recessed tracks (not shown) which are machined onto the outer surface of the housing 205 during manufacture. Additional shapes may also be machined onto the outer surface in order to _x accommodate components such as the transmission and réception coil capacitors used in the 15 transmission and réception circuit. Electrical insulation (not shown) such as adhesive insulant is also providcd between the transmission/reception coil and the housing 205.

[0022] In further contrast, with the first embodiment, the apparatus 100 has gradient coils Gx, Gy, Gz mounted in tubing 215 between the primary magnet portions 16QA and 160B. This séparâtes the magnets 160A and 160B from one another which increases the 20 comblned efiiciency of the magnets in producing a high strength homogeneous magnetic field in the flow path. The tubing 215 also provides mechanical support to retain the primary magnet and to provide support against the pressure exerted from the flow. In the présent embodiment, the tubing 215 is made from high permeability iron and is dodecagonal in shape (as shown in FIG. 10). A pair of axial end members 215A are also provided in order to 25 provide a magnet ically permeable path for the magnetic ficld.

[0023] As seen in FIG. 12, tubing 215 houses the axial gradient coil along the flow path (Gz) on the inner surface and the orthogonal gradients (Gx and Gy) on the outer surface (sce FIG. 13). Again, thèse coils arc providcd in rcccsscd tracks on the tubing 215 and are insulated from the tubing itself using adhesive insulant. The gradient coils are capable of 30 imparting a variable magnetic field as discussed subsequently and in this regard can be considercd as an electromagnet.

[0024] The tubing 215 is provided with a tubular inner diameter in order to provide minimal frictional losses to the fluid passing thercthrough, and a dodecagonal outer surface which allows the tubing to fit with in the rings of magnets.

t β

[0025] In use, each embodiment ofthe apparatus 10 opérâtes in an idcntical fashion by utilizing Nuclear Magnetic Résonance (NMR) techniques in order to détermine the volume fraction of multiphase flow produced from a wellbore. In addition to determining the fraction of each phase présent in the flow, the invention may also be used to détermine the rate of fluid flowing from the wellbore. The embodiments described détermine the phase fraction of fluid containing oïl, gas and water phases; however, it will be understood by the ski lied reader that fùrther and/or different phases may be determined using the apparatus and method described.

[0026] For clarity the phase fraction analysis process will firstly be described followed by a description of the flow measurement process; however, both of these processes may bc eflectively carried out simultaneously by confîguring the control System of the apparatus 10 to rapidly altemate between fraction analysis mode and flow measurement mode. This alternation between modes is typically performed at a rate of approximately one second for each mode i.e. the control System will allow the fraction analysis mode to operate for one second and then allow the flow measurement mode to operate for one second before swîtching back to the fraction analysis mode and so on as required. The skilled reader will note that this time may be altered to suit the spécifie situation.

[0027] The method of using the first embodiment of the apparatus will bc described in the following description; however, the skilled reader will realize that either embodiment 20 may bc used.

[0028] In the embodiment shown, the apparatus 10 is installed in-line with a fluid flow pipe 14. As produced fluids flow into the apparatus 10, they enter the substantially homogeneous primary magnetic field generated by primary magnet 16. Atomic nuclei having a non-zero magnetic moment that are présent in the fluids flowing through the apparatus 10 25 align themselves with the axis of the primary magnetic field. Fluids having a non-zero magnetic moment include *H, ¹³C, ^3IP and ^1JN. In this embodiment (and in many NMR applications in general) ’H is the most commonly measured of these since it is naturally présent in hydrocarbons such as thosc produced from wcllborcs. The nuclci of flow within the throughbore 36 of the apparatus 10 including water, oil and gas are now aligned with the 30 direction of the primary magnetic field.

[0029] A radio frequency (RF) puise signal is now transmitted into the throughbore 36 using the transmission circuit 48 ofthe combined transmission and réception coils 24. The frequency ofthe RF puise will be transmitted at a frequency which is known to excite the atomic nucléus of’H (typically in the région of between 40-45 MHz for a 1 Tesla statîc

I >

magnetic field such that it resonates at its natural résonant frequency (this is known as the Lannour frequency). This ensures that any ’H nuclei présent in fluid flowing through the throughbore 26 will resonate in response to the RF puise signal. The frequency (v) required to resonate the nuclei may be determined using the following équation:

where .gamma, is the gyromagnetic ratio ofthe nucléus and B is the magnetic field.

[0030] While resonating, the nuclei emits a radio signal at a frequency corresponding to its resonating frequency.

[0031] The frequency at which the nuclei présent in the fluid flow rcsonatc after having being excited by the RF puise signal is detected by the receiver circuit 40 of the combined transmitter and réception coils 24. In a mixture of phases such as in the présent embodiment, the résonance described provides molecular information such as the bond type and the environment surrounding the nuclei. From this, the ratio of the signal being received from the resonating nuclei to the background frequency of the RF puise may be calculated. The skilled reader will understand that this value is known as Chemical Shîft and is measured in parts per million (ppm).

[0032] The chemical shift (δ) recorded by the apparatus may now be used to détermine the ratio ofoil and gas (combined) to water using the following équation:

<5 = ^Vnfmc’xlQ⁶ppm (Eq. (2) [0033] In this regard, the séparation between the phases is ïncreased by ensuring that good magnetic field homogeneity is provided by the primary permanent magnet 16 in order to produce a relaxation time graph peak with a small bandwidth.

[0034] However, as stated prcviously it is désirable to measure the ratio ofoil to gas also in order to détermine the ratios of oil, gas and water in the multiphasc fluid, without assumîng présence of other phases. In general, the chemical shift between oil and gas nuclei is too small to measure accurately by using the chemical shift method. There fore, the présent invention déterminés the ratio of oil to gas by comparing the Tj relaxation times (described subsequently) of each hydrocarbon. This is possible since the Ti relaxation times of gaseous hydrocarbons are longer compared to the T] relaxation times of liqutd hydrocarbons.

[0035] In addition to causing the nuclei of each phase to resonate· the energy supplied by the RF puise signal from the combined transmission and réception coil 24 causes the nuclei of each phase to bc knocked ofî their previous alignment with the primary magnetic

I )

field. After the RF signal has been pulsed, the spins (nuclei which hâve becn subjected to a magnetic field) will tend to relax back to their state of equilibrium in which they are realigned along the primary magnetic field. The time takcn for the spins to relax back to their state of equilibrium after the RF signal has been pulsed off is known as the Ti relaxation time of the nuclei.

[0036] It is possible to measure the Ti relaxation times of the oil and gas using the apparatus 10 by monitoring the angle through which the nuclei of each phase of the flow is tîltcd with respect to the primary magnetic field at any given time (which must be less than the relaxation time) after the RF signal has been pulsed. This is donc by measuring the time taken for the magnitude of the radio ftequcncy receîved from the nuclei to reach a maximum value in the direction of the primary magnetic field and the time taken for a minimum value in the direction orthogonal to the primary magnetic field direction, which may be performed using the combincd transmission and réception coils 24. This résulte in two distinct Tj relaxation times being détectable; one for the oil phase and one for the gas phase. The proton density (PD) of each hydrocarbon phase is now calculated by integrating the area under each peak of the accumulated Tj relaxation time density. The graph is derived by applying an inverse atgorithm to the T| relaxation time measurement extracted using an inversion rccovcry scquencc. Using the proton density measurement the volume fraction is now calculated using the foliowing équation:

Eq.(3) where MW, is the molecular weight, p» is the density of the sample, Av is the Avogadro number, PD is the proton density, a is the natural abundance of’H and Rm is the number of ¹H for 1 molécule ofthe phase.

[0037] The sequcnce applied herc is such that the required measurement time is less than the transit time (τ) of the flow. The method of determining the proton density is performed using a t-dimensional hydrogennuclei (1D-1 H) sequcnce in combination with an inversion recovery sequence for Tj measurement and Carr-Purccll-Meiboom-Gill (CPMG) sequcnce for T₂ measurement.

[0038] However, the above merely retums values for the volume of the relevant phases and, as prcviously mentioned, not the phase fraction. In order to calculate the phase fraction, the foliowing équation may be used:

I 1

Λ/W PD, ----^Lx---1

Eq.(4) where n is the number of phases présent in the sample.

[0039] It should be noted that in a sample containing just two phases (a and b), the équation can be simplified to:

[0040] Each of the fractions ofοΐζ gas, and water hâve therefore been calculated using the apparatus 10 wïthout (as in somc previous Systems) requiring to assume that once the ratios of two phases in the flow hâve been calculated the third makes up the rest of the fluid.

[0041] The method and apparatus for determinîng the flow rate of the fluid flow will now be described.

[0042] Now that the ratio of each phase has been calculated, the Tj relaxation time of each phase is known. The embodiment shown is capable of employing two alternative methods of calculating the flow rate of each phase through the apparatus 10. The first method is based upon the Time of Flight (TOF) of the spins abng the apparatus 10. In this method a puise signal is applied in a 'slice' at a first location along the throughbore 36 of the apparatus 10 in order to tilt the nuclei at that location. A détection area is then monîtored downstrcam from the location at which the puise signal was applied. The résultant NMR. signal received by the réception circuit 40 of the combincd transmission and détection coils 24 will now be increased by every fuily tilted spin entering the détection area and will be decreased with every fully tilted spin leaving the détection area. The overall net signal can therefore be rclatcd back to the flow of phase through the apparatus. This alfows the velocity of the flow (v) to be calculated using the transit time (τ) and the distance of the détection area (d) using the folio wing équation:

r=— Eq.(7) v

[0043] The second alternative method of mcasuring the flow through the apparatus 10 uses the gradiated magnetic field provided by the sccondary clectromagnet 20. A gradient ccho scquence is imparted on the flow such that the nuclei of the flow rotate about the ir axes.

In a stationary flow this results in no net accumulation of phase signais since the nuclei expérience the same balanced gradient with respect to time. However, in a dynamic flow the magnetic fîeld experienccd by the nuclei wili change as the nuclei flow along the throughbore 36 of the apparatus 10 due to the magnetic fîeld gradient provided by electromagnet 20. This 5 variation of magnetic fîeld, dépendent upon the movement of the flow along the throughbore ofthe apparatus 10, results in an accumulation of phase signal. This is dépendent upon the velocity of the flow through the apparatus 10 and the strength and duration of the magnetic fîeld gradient supplied by the electromagnet 20. The accumulation in phase (¢) which may be directly correlated to the velocity of the flow is given by:

φ = dt +y{(t)A Eq. (8) where Bo is the magnetic fîeld provided by the primary magnet, n represents the position of the spins within the throughbore in either the x, y, or z axes (as shown in FIG. 1) and G_a is the magnitude ofthe magnetic fîeld gradient being applicd by the electromagnetic 20 in the n-axis direction.

[0044] The method described previously allows both the flow rate and proportion of 15 each phase to be calculated using a single apparatus 10. Furthermore, the System and apparatus described does not rcquire users of the apparatus to be safeguarded from levels of opcrational danger other than that normal ly expected in such oil and gas exploration operations. Specifically, the apparatus and method described does not require the user to be protected against e.g. radiation and biological hazards.

2₀ [0045] In other aspects, compositional analysis of hydrocarbons produccd from the wellbores may be provided using the imparted and detected radio frequency signais. FIG. 14 shows onc embodiment of an apparatus that may be utilized for estimating spccics of hydrocarbons of downhole fluids. The apparatus of FIG. 14, in one aspect, may include a processor, such as a microprocessor or a computer 1410 and a data storage device 1420, which may bc any suitable device, inciuding, but not limited to, a solid state memory, compact dise, hard dise, and tape. One or more programs, models and other data (collectively referred to as programs and designated by numéral 1430) may be stored in the data storage device 1420 or another suitable device accessible to the processor 1410 for executing instructions containcd in such programs. A display device 1440 may be provided for the processor 1410 to display information relating to the compositional analysis, as described in more detail below.

t [0046] In one aspect, the processor 1410 may compute the frequency différence or frequency shift 1460 between the original imparted or perturbing signal 1402 and the detected signal 1401. This phenomenon is also known as the chemical shift. In one aspect, the processor 1410 may estimate or détermine the composition (species of the produced hydrocarbons) using the frequency or chemical shift.

[0047] FIG. 15 shows a relationship 1500 of the frequency or chemical shift and various species of a hydrocarbon. For example, a chemical shift between -2 to -1.0 parts per million (ppm) indicatcs the présence of alkanc, a chemical shift between -7 to -5.5 ppm indicates the présence of Alkene, and a chemical shift between -8.0 and -7.0 ppm indicates 10 the presence of aromatic compounds. Aromatic compounds are the compounds that hâve a benzene ring structure, such as toluene, benzene and zylene. The data shown in FIG. 15 may be stored in the data storage device 1420 for use by the processor 1410.

[0048] FIG. 16 shows a chemical shift curve or spectrum 1600 relating to the various species of a hydrocarbon produced from the wellbores. In another aspect, the processor 1410 15 may estimate an amount of the species from the spectrum 1600. In one aspect, an area under the chemical shift curve 1600 may be integrated to estimate the amount of the species in the hydrocarbon. Therefore, the area under the curve section 1610 wüt provide the amount or fraction of alkanc in the fluid sample, while the area under section 1620 will providc the amount or fraction of oromatics in the fluid sample. Composition of the species may therefore 20 be estimated as a relative value or as an absolute value.

[0049] ln another aspect, the programs 1430 may include instructions for the processor to détermine the relaxation times and provide therefrom a dctailed analysis of the species, such as a breakdown of the types of alkanes. Relaxation time, as described prcviously, is the time the signal cmitted by the nuclei takes to decay. There is a direct 25 relationship between the density of the alkane (which is linked to the length of the carbon chain) and the relaxation time. The higher the density of alkane, the longer the relaxation time.

[0050] ln another aspect, the compositionai analysisofhydrocarbon described herein may bc utilized to provide information for a PVT analysis of the hydrocarbon, From the estimated composition of the hydrocaibon, as described above, the overall PVT properties of the hydro carbon may bc determined.

[0051] In prior art techniques, the PVT analysis of the hydrocarbon is typically done by taking a sample of the downhole fluid at a known pressure and température and various tests are conducted to détermine the properties of the fluid, such as the bubble point and density and viscosity at various températures and pressures. The brcakdown of the hydrocarbon to its core components is then made. The PVT properties of the core components are well known and may be reconstitutcd to provide an overall hydrocarbon PVT property. A known reconstitution technique is used. For example, for gas hydrocarbons, gas 5 chromatography may be pcrformcd to break down the gas into its individual components and based on this composition, an overall gas property may be recalculated.

[0052] The technique of compositional analysis described herein bypasscs taking a fluid sample and rigorous tests typically pcrformcd to détermine the PVT properties of the hydrocarbon. Furthcrmore, in the présent method, the measurements are donc in real time (in10 situ) as opposcd to ‘samplïng* of the hydrocarbon.

[0053] The dimensions ofthe apparatus may be altered during the manufacturing stage dépendent upon the particular downhole or subsea conditions in which it is to be used. In this regard, the space requirements ofthe components may be balanced based on the accuracy of desired measurement, which may be relevant for the primary magne116 and the 15 clectromagnct 20. Additionally, the apparatus described above may be used in a wellbore or in-linc with any portion of the production tubing. Altematively the apparatus may be used off site as an off-site measurement and analysis tool.

[0054] ln another aspect, the présent disclosure provides a method of determining volumes of liquid and gas phases of a multiphase fluid. Phase volume is determined for a 20 multiphase fluid in a constant measurement volume, such as in a volume of the NMR apparatus of FIG. 17. FIG. 17 shows an exemplary Nuclear Magnetic Résonance (NMR) flow meter device 1700 for estimating a volume of a fluid phase using the exemplary methods ofthe présent disclosure. In one embodiment, the fluid is a multiphase fluid. In another embodiment, the fluid is a fluid flowing in a production system or a pipe for transportation of hydrocarbons. The exemplary NMR flow meter 1700 inciudes a section

1710 for providing NMR excitation puises to the fluid and obtaining NMR signais in response to the NMR excitation puises from the fluid, and a testing unit 1726 for receiving the NMR response signais from the détection section 1710 and performing calculations on the rcccivcd NMR response signais to obtain a phase volume.

3Q [0055] The device 1700 inciudes a magnet 1714 which may be exterior to the détection pipe section 1712 for providing a static magnetic field in a volume of the détection pipe section 1712, and a radio frequcncy (RF) coil 1716. As fluid passes through the static magnetic field of magnet 1714, nuclear spins of atoms and molécules within the fluid align along the direction ofthe static magnetic field. The RF coil 1716 encloses a volume within « b the détection pipe section and is arranged to provide one or more NMR excitation puises to the fluid in the device and to detect one or more NMR response signais from the fluid. Testing unit 1726 inciudes various circuitry for obtaining one or more NMR response signais from the fluid and estimating a parameter of interest of the fluid from the obtained NMR response signais. In one embodiment, the exemplary testing unit 1726 is coupled to the RF coil 1716 via preamplifîer 1720. The exemplary testing unit 1726 inciudes a transmitter 1724 for providing an NMR excitation puise to the RF coil 1716 via preamplifier 1720. In one embodiment, the transmitter 1724 provides multiple NMR excitation puise séquences, cach NMR excitation puise sequence tuned to a selected nuclear résonance frequency. The exemplary testing unit 1726 also inciudes a recciver 1722 for receiving NMR response signais detected at the RF coi! 1716 via the preamplifier 1720. Testing unit 1726 also inciudes an NMR a control unit 1728 for estimating one or more parameters of the fluid from the received NMR response signais using exemplary methods ofthe présent disclosure. In one embodiment, the control unit 1728 may include a processor 1730, one or more computer programs 1732 that are accessible to the processor 1730 for executing instructions contained in such programs to obtain one or more fluîd-related parameters such as a fluid phase volume, viscosity, water eut, and an émulsion stability parameter, for example, and a storage device 1734, such as a solid-statc memory, tape or hard dise for storing the one or more parameters obtained at the processor 1730.

[0056] NMR signal amplitude measurcments in a constant volume measurement device can be used to détermine a tiquid phase fluid volume fraction vs. gas phase volume fraction. As shown with référencé to FIG. 17, fluids flow into apparatus 1700 where they enter the substantially homogeneous primary magnetic field generated by primary magnet 1714. The nuctei ofthe produced fluids thereby polarize along the direction ofthe primary magnetic field. A radio frequency (RF) puise signal ïs transmitted into the throughbore 1712 using the transmitter 1724 ofthe combined transmission and réception coils 1716. A signal amplitude for the nuctei présent in the fluid flow after having being excited by the RF puise signal is dctcctcd bythe recciver 1722 ofthe combined transmitter and réception coils 1716.

[0057] NMR signal strength for a phase is a fonction of phase volume and hydrogen index for the phase. Hydrogen index is proportïonal to the proton density in the sensitive volume ofthe measurement device. The hydrogen index ofméthane CHi, for cxample, is proportïonal to the gas density:

Eq.(9)

» t with/1=2.25. Gas density, in tum, is température and pressure dépendent and can be determined by équation of state (EOS) corrélations such as those described by Londono et al (SPE paper #75721 (2002)) and by Drumdruk et al “Calculation of z-factors for naturel gases using équations of states” in J. of Canadien Petro., v.14, pp.24-26 (1975).

[0058] In order to détermine liquid and gas phase volumes, a calibration signal is first obtained. The calibration signal is used to scalc NMR signal strength or signal amplitude. In one aspect, calibration signal strength M is determined for a calibration signal using 100% water, such as distillcd water (Hl=l) flowing in the production tubular. A calibration constant can be obtained using c-M/V, where V is the sensitive volume and M is the strength of the calibration signal at a given température. The value of the calibration constant may be temperature-dependent and can therefore be determined either by using a theoretical prédiction or by obtaining values for calibration constants at a plurality of températures and interpolating the values when necessary.

[0059] ln a multiphase fluid having liquid and gas phases, NMR signal amplitude m is described by

Eq.(10) where c(T) is the temperature-dependent calibration constant, is the volume of a gas phase, is the volume of a liquid phase, Hï_gtti is the hydrogen index of the gas phase and Hhiquid is the hydrogen index ofthe liquid phase. Most formation water has a hydrogen index that is close to unit y. Salinity of water can affect the hydrogen index, and the hydrogen index can be determined when salinity is known or estimated. The hydrogen index of liquid hydro carbon is also substantially close to one.

[0060] When Hlu^id is I and HI_gat is known for example from Eq. (9), then Eq. (10) can be solved for gas volume to obtain:

Eq.(ll)

An équation similar to Eq. (II) can be derived for when ΗΙα^Μ is a known value not equal to

1. Such an équation may be suitable for obtaining V_gal in subsea or downhole locations.

[0061] In various aspects, phase volume may be obtained at a surface location, wherein pressure is substantially close to 1 atm. At this (low) pressure, HI_gai is substantially close to zéro and thus \-HIpu is substantially close to 1. For an équation more suitable to volume phase estimation, Eq. (10) can be obtained for low pressure environments by noting that Y · Hl_taa « ν_αφύΛ · . If Υι^υ is substantially non-zero, then

I m * c(r) and the résultant NMR signal représenta primarily the liquid phase(s):

Eq.(12) îfHuq_Vid= l.then

Eq. (13)

M and

Eq. (14)

Altematively, Vii^m and Ÿ_tiu can be estimated using a first order Taylor expansion of Eq. (10):

Eq.(15) and

Eq.(16) [0062] Thus, liquid phase volume and gas phase volume can be determined for a multiphase fluid flowing in a production tubing. Additionally, volumétrie flow rates of each 10 phase may be obtained from the obtained phase volumes and flow velocities, [0063] In another aspect, the présent disclosure provides a method ofdctcrmining a stability of an émulsion (te., multiphase fluid) flowing in a production tubing. Emulsion stability can vary from very tight émulsions characterized by small and closely distributed droplcts to very loose émulsions characterized by large and widely distributed droplets.

Emulsion stability can be characterized by an émulsion stability parameter, the value of which indicates the type of émulsion. For instance, an émulsion stability parameter of about 7.0 indicates a very tight émulsion and an émulsion stability parameter of about 3.0 indicates a very loose émulsion.

[0064] As discussed below, the émulsion stability parameter is related to viscosity

2o and water eut of the émulsion. Viscosity measurcments can be determined from relaxation times ofNMR signais obtained from the cmulsion. Relaxation times ofthe émulsion i.c, Ti (spin-lattice relaxation constant) and Tï (spin-spin relaxation constant), for example, can be measured using the exemplary apparatus of FIG. 17. An exemplary relation between Τι, T2 and the viscosity of a fluid is given below:

_ 1 _ 40nA/ja’p Ί\~Ί\~ 9kT whcre

Eq. (17)

Eq. (18) wherein €□ is the magnctic pcrmcability of free spacc, h is Planck’s constant dividcd by 2π, γ is the gyromagnetic ratio for an H proton, r is a distance between nearcst protons in a molécule, μ is the viscosity of the fluid, k is the Boltzmann constant, and T is the absolute température.

[0065] FIG. 18 shows an exemplary relation between émulsion viscosity and watercut (WC) or volume fraction of water in the émulsion. Water eut can be determined using the exemplary methods discio sed herein or any other methods known in the art such as infrared 10 measurements. In various émulsions, the viscosity of the émulsion is typically greater than the viscosity of either oil or water at a given température. For instance, FIG. 18 shows various curves of viscosity vs. water eut at different températures (75°F (1801), 100°F (1803), I25°F(1805), 150°F (1807)) for an émulsion of water and oil At water eut of0%, the émulsion is 100% oil and the émulsion has the viscosity of oil (μο). At water eut of

100%, the émulsion is 100% water and the émulsion has the viscosity ofwater (μ«). As the water eut of the émulsion increascs from 0%, the viscosity of the émulsion increases up to a water eut inversion point. In this région between 0% water eut and the inversion point, the émulsion is a water-in-oil émulsion. In the région between the inversion point and 100% water eut, the émulsion is an oil-in-water émulsion. The inversion point in exemplary FIG.

17 occurs at approximately a water eut of 80% water.

[0066] The viscosity of an émulsion dépends on several factors: the viscosîties of oil and water, the volume fraction ofwater (watercut), dropfet size distribution, température, shear rate, amount of solids présent, for cxample. The viscosity of the émulsion can be substantially higher than the viscosity ofoil or water at a given température. The viscosity of 25 an émulsion is relatcd to the viscosity of the virgin crudc oil at the same température by the following équation for WC<WCi_n/.

= p_oe^(l-3WC+a(WC)²) Eq. (19) whcre a is the émulsification stability, μο is oil viscosity, μν is water viscosity, WC is the water eut and WCi_nv is the water eut inversion point. In addition, the following relation holds 30 forWC>WC_lnv:

k

[0067] Eq.(19) can be used to détermine émulsion stability parameter a for an émulsion flowing in a production tubing. The method includes obtaining a relaxation time of an NMR signal of the émulsion, determining a viscosity of the émulsion from the obtained relaxation time, and determining émulsion stability using the determined viscosity and a value of water eut for the émulsion.

[0068] In various aspects, information of fluid émulsification assista production cnginccrs in pipeline modeling by providing information on the drag exerted by the fluid. The présent invention further provides an ability to image a flow pattern in the pipeline. This provides data for diagnosing plugs formations within the pipeline, thereby enhancing production pipeline and wellbore modela. Knowing fluid composition also enables fluid analysis to obtaining fluid P VT properties.

[0069] The NMR measurement apparatus provides real-time measurement data which can be continuously updated. Updated compositional information can be fed back to a production model to improve flow analysis. Multiple technologies such as but not limited to infrared measurement (IR) can be used with the magnetic résonance mcasuremcnts, [0070] Therefore, in one aspect, the présent disclosure provides a method of determining a volume of a phase of a multiphase fluid flowing in a tubular, including: imparting a magnetic ficld on the fluid to align nuclei of the multiphase fluid along a direction of the magnetic field; transmitting a radio frequency signal into the multiphase fluid to excite the nuclei; detecting a signal from the nuclei responsive to the transmitted radio frequency signal; determining an amplitude of the detected signal; and determining the volume of the phase flowing in the tubular using the determined amplitude and an amplitude of a calibration signal. When the phase is a liquid phase, the method further includes determining a relaxation time ofthe detected signal; and determining a viscosity ofthe multiphase fluid using the determined relaxation time. In one embodiment, the method includes determining a water eut of the liquid phase; and determining an émulsification stability parameter of the liquid phase from the determined viscosity and the determined water eut. Determining water eut ofthe liquid phase may inciude determining a proton density of the liquid phase from an accumulated relaxation time density. Also, the calibration signal may be obtained from a flow ofwater. Determining the volume of the phase may inciude determining a hydrogen index of a phase ofthe multiphase fluid. The method may further inciude determining a flow velocity ofthe phase; and determining a flow rate ofthe

phase using the determined volume of the phase and the determined flow velocity of the phase. In various embodiments, the amplitude of the calibration signal is a temperaturecorrected amplitude.

[0071] In another aspect, the présent disciosure provides an apparatus for determining 5 a volume of a phase of a multiphase fluid flowing in a tubular, the apparatus including a source confîgured to impart a primary magnetic field on the fluid to align nuclei ofthe multiphase fluid along a direction o f the primary magnetic field; a source confîgured to transmit a radio frequency signal into the multiphase fluid to excite the nuclei; a dctcctor for detecting a signal from the nuclei responsive to the transmitted radio frequency signal; and a 10 processor confîgured to détermine an amplitude ofthe detected signal and the volume ofthe phase flowing in the tubular using the determined amplitude ofthe detected signais and an amplitude of a calibration signal. The processor may be confîgured to détermine a relaxation time of the detected signal; and détermine a viscosity of the multiphase fluid using the determined relaxation time. In one embodiment, the processor is confîgured to détermine a water eut ofthe liquid phase; and détermine an émulsification stability parameter of the lïquid phase from the determined viscosity and the determined water eut. The processor may be further confîgured to déterminé water eut ofthe liquid phase by determining a proton density ofthe liquid phase from an accumulatcd relaxation time density. The calibration signal may be obtained from a flow ofwater. The processor may be confîgured to détermine the volume

2o ofthe phase by determining a hydrogen index ofa phase ofthe multiphase fluid. In various embodiment, the processor is confîgured to déterminé a flow velocity o f the phase and détermine a flow rate of the phase using the determined volume ofthe phase and the determined flow velocity ofthe phase. The amplitude of the calibration signal is typically a temperaturc-corrected amplitude.

[0072] In another aspect, the présent disciosure provides a method of determining stability of an émulsion flowing in a production string, the method including imparting a primary magnetic field on the émulsion to align nuclei of the émulsion along a direction of the primary magnetic field; transmittiog a radio frequency signal into the cmulsion flowing in the production string; detecting a signal from the nuclei of the émulsion responsive to the transmitted radio frequency signal; determining an amplitude of the detected signal; determining a water eut ofthe émulsion using the obtained amplitude; determining a relaxation rate ofa signal obtain from nuclei ofthe émulsion excited in response to the transmitted radio frequency signal; obtaining a viscosity ofthe émulsion from the determined relaxation time; and determining the stability of the émulsion from the determïned émulsion viscosity and the water eut of the émulsion.

[0073] While the foregoing disclosure is directed to certain embodiments, varions modifications will be apparent to those skillcd in the art. It is intended that ail such modifications fall within the scope and spirit of this disclosure and any daims that arc or may be presented.

Claims (17)

1. What îs claimed is:

   1. A method o f determining a vo lume o f a phase o f a multiphase fluid flowing in a tubular, comprising:

   impartïng a magne tic field on the fluid to align nuclei of the rnultîphasc fluid along a direction of the magnetic field;

   transmitting a radio frequency signa! into the multiphase fluid to excite the nuclei; dctccting a signa! from the nuclei responsive to the transmitted radio ftcqucncy signal; determining an amplitude of the detected signal; and determining the volume of the phase flowing in the tubular using the determined amplitude and an amplitude of a calibration signal.
2. 2. The method of claim 1, wherein the phase îs a liquid phase, the method further comprising:

   determining a relaxation time of the detected signal; and determining a vîscosity of the multiphase fluid using the determined relaxation tîme.
3. 3. The method of claim 2 further comprising:

   determining a water eut of the liquid phase; and determining an émulsification stability parameter of the liquid phase from the determined vîscosity and the determined water eut
4. 4. The method of claim 3, wherein determining water eut of the liquid phase further comprises determining a proton densityofthe liquid phase from an accumulated relaxation time density.
5. 5. The method of claim 1, wherein the calibration signal is obtained from a flow of water.
6. 6. The method of claim 1, wherein determining the volume of the phase further comprises determining a hydrogen index of a phase of the multiphase fluid.
7. 7. The method o f claim 1, further comprising:

   determining a flow vclocity of the phase; and determining a flow rate of the phase using the determined volume of the phase and the determined flow velocity of the phase.
8. 8. The method of claim 1 wherein the amplitude of the calibration signal is a temperaturc-corrcctcd amplitude.
9. 9. An apparatus for determining a volume of a phase of a multiphase fluid flowing in a tubular, comprising:

   » I a source configurée! to impart a magnetic fïcld on the fluid to align nuclei of the multiphase fluid along a direction of the magnetic field;

   a source configured to transmit a radio frequency signal into the multiphase fluid to excite the nuclei;

   a detector for detecting a signal from the nuclei responsive to the transmitted radio frequency signal; and a processor configured to détermine an amplitude of the detected signal and the volume of the phase flowing in the tubular us in g the determined amplitude of the detected signais and an amplitude of a calibration signal.
10. 10. The apparatus of claim 9, wherein the processor is further configured to: détermine a relaxation time of the detected signal; and détermine a viscosity of the multiphase fluid using the determined relaxation time.
11. 11. The apparatus of claim 10, wherein the pro cessor is further configured to : détermine a water eut of the liquid phase; and détermine an émulsification stability parameter of the liquid phase from the determined viscosity and the determined water eut.
12. 12. The apparatus of claim 11, wherein the processor is further configured to détermine water eut of the liquid phase by determining a proton density of the liquid phase from an accumulated relaxation time density.
13. 13. The apparatus of claim 9, wherein the calibration signal is obtained from a flow of water.
14. 14. The apparatus of claim 9, wherein the processor is further configured to détermine the volume of the phase by determining a hydrogen index of a phase of the multiphase fluid.
15. 15. The apparatus of claim 9, wherein the processor is further configured to: détermine a flow velocity of the phase; and détermine a flow rate of the phase using the determined volume of the phase and the determined flow velocity of the phase.
16. 16. The apparatus of claim 9 wherein the amplitude of the calibration signal is a temperature-corrected amplitude.
17. 17. A method of determining stability of an émulsion flowing in a production string, comprising:

    imparting a magnetic field on the émulsion to align nuclei of the émulsion along a direction of the magnetic field;

    transmîtting a radio frequency signal into the émulsion flowin g in the production string;

    detecting a signal from the nuclei of the émulsion rcsponsive to the transmitted radio frequency signal;

    5 determining an amplitude of the detected signal;

    detemrining a water eut ofthe émulsion using the obtained amplitude;

    determining a relaxation rate of a signal obtain from nuclei ofthe émulsion excited in response to the transmitted radio frequency signal;

    obtaining a viscosity of the émulsion from the determined relaxation time; and

    10 determining the stability of the émulsion from the determined émulsion viscosity and the water eut of the émulsion.