Method for operating a nuclear magnetic flowmeter

Abstract

A method for operating a nuclear magnetic flowmeter for determining the flow of a multi-phase medium flowing through a measuring tube that is suitable for media exhibiting phase slip with which the characterization of the gaseous phase is simplified, is achieved by a pulse spoiling the magnetization at least in the direction of the magnetic field or a pulse sequence spoiling the magnetization in the direction of the magnetic field being emitted by a coil-shaped antenna, particularly in combination with dephasing gradients, and then, after a waiting time t.sub.W, a nuclear magnetic measurement is carried out in that the medium is excited with excitation pulses by the coil-shaped antenna and the measuring signals generated by the excitation in the medium are detected.

Claims

1. Method for operating a nuclear magnetic flowmeter for determining the flow of a multi-phase medium flowing through a measuring tube, having a pre-magnetization unit for pre-magnetization of the medium, a magnetic field generator for generating a magnetic field interfusing the medium and a measuring device that includes at least one coil-shaped antenna for generating excitation signals exciting the medium and/or for detecting measuring signals emitted by the medium, comprising the steps of: emitting a pulse spoiling the magnetization at least in the direction of the magnetic field or a pulse sequence spoiling the magnetization in the direction of the magnetic field by the coil-shaped antenna, waiting a waiting time tw, and then, carrying out a nuclear magnetic measurement by exciting the medium with excitation pulses by the coil-shaped antenna and by detecting measuring signals generated by the excitation of the medium, wherein the waiting time tw is determined in accordance with the relationship $t_W\frac{L_1}{v_{\mathrm{gas}}}$ wherein L1 is the length of the coil-shaped antenna and v.sub.gas is the flow velocity of the gaseous phase; wherein the waiting time t.sub.w is iteratively determined using the signal amplitude and the signal amplitude S.sub.gas of the gaseous phase, wherein the signal amplitude ratio and the signal amplitude of the gaseous phase are at a maximum for $t_W=\frac{L_1}{v_{\mathrm{gas}}},$ and wherein flow velocities of individual phases are determined using the measured values obtained by the nuclear magnetic measurements.

2. Method according to claim 1, wherein emitting the pulse spoiling the magnetization at least in the direction of the magnetic field or the pulse sequence spoiling the magnetization in the direction of the magnetic field by the coil-shaped antenna is combined with dephasing gradients.

3. Method according to claim 1, wherein the pulse spoiling the magnetization is a P90 pulse.

4. Method according to claim 3, wherein the P90 pulse is followed by a dephasing gradient pulse.

5. Method according to claim 1, wherein portions of the individual phases in the medium are determined using the signal amplitude ratio of an amplitude of a gaseous phase signal S.sub.gas to an amplitude of a liquid phase signal S.sub.liquid.

6. Method according to claim 1, wherein the nuclear magnetic measurement is implemented using a CPMG sequence.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The sole FIGURE is a diagrammatic sectional view of a known type of nuclear magnetic flow meter that is suitable for use in practicing the method of the invention.

DETAILED DESCRIPTION OF THE INVENTION

(2) FIG. 1 depicts a nuclear magnetic flowmeter 1 for determining the flow of a multi-phase medium flowing through a measuring tube 2, having a pre-magnetization unit 3 for pre-magnetization of the medium, a magnetic field generator 4 for generating magnetic field pulses that interfuse the medium and a measuring device 5 that includes at least one coil-shaped antenna 6, 7 for generating excitation pulses for exciting the medium and/or for detecting measuring signals emitted by the medium.

(3) A particular implementation of the method according to the invention is characterized in that the pulse spoiling the magnetization is a P90 pulse, particularly followed by a dephasing gradient pulse. However, any other pulse or any other pulse sequence that can spoil the magnetization in the direction of the magnetic field is possible. Thus, the method according to the invention is not limited to the use of a P90 pulse as spoil pulse.

(4) A further preferred implementation of the method according to the invention is characterized in that the waiting time t.sub.W is given by

(5) $t_W\frac{L_1}{v_{\mathrm{gas}}}$
wherein L.sub.1 is the length of the coil-shaped antenna and v.sub.gas is the flow velocity of the gaseous phase.

(6) As already described above, the medium continues to flow in the measuring tube after the magnetization has been spoiled in the direction of the magnetic field by the spoil pulse. The time that the gaseous phase requires for covering the entire length L.sub.1 of the path containing the coil-shaped antenna results from the length L.sub.1 divided by the flow velocity of the gas v.sub.gas. If the waiting time t.sub.W has a greater value according to this, this means that the entire gas portion exhibiting spoiled magnetization has flowed out of the coil-shaped antenna and thus the area of the coil-shaped antenna has been completely refilled with fresh gaseous medium and emits a maximum possible nuclear magnetic measuring signal. The longer the waiting time is, the more medium of the liquid phase that leaves the coil. Accordingly, the portion of fresh liquid medium becomes greater with increasing waiting time, and thus also again provides a greater contribution to the measuring signal.

(7) It should be taken into consideration that a longer waiting timemeaning a waiting time longer than t.sub.W=L.sub.1/v.sub.gasleads to a decrease in the effect caused by the method according to the invention, namely a decrease in the amplification of the portion of the measuring signal generated by the gaseous medium. If more fresh liquid medium flows into the area of the coil, the contribution of the liquid medium to the measuring signal increases until, finally, the original state is reinstated, i.e., the entire coil is refilled with fresh gaseous and fresh liquid medium and the measuring signal is dominated by the liquid phase. Thus, a suitable choice for the waiting time t.sub.W is indispensable.

(8) A particularly preferred implementation of the method according to the invention is characterized in that the waiting time t.sub.W is iteratively determined using the signal amplitude ratio S.sub.gas/S.sub.liquid and the signal amplitude S.sub.gas of the gaseous phase, wherein the signal amplitude ratio and the signal amplitude of the gaseous phase are at a maximum for t.sub.W=L.sub.1/v.sub.gas.

(9) As already described above, it is absolutely of relevance to maintain a suitable value for the waiting time t.sub.W between spoiling of the magnetization in the direction of the magnetic field and beginning nuclear magnetic measurement. If it is said here to maintain a suitable value, then this means to use a value for the waiting time t.sub.W, in which as much fresh gaseous medium as possible and as little fresh liquid medium as possible are found in the area of the coil-shaped antenna. An indication for this state is the value of the signal amplitude ratio S.sub.gas/S.sub.liquid in combination with the value of the signal amplitude S.sub.gas of the gaseous phase. The signal amplitude of the gaseous phase has a maximum value when the gaseous portion found in the coil-shaped antenna is formed completely of fresh medium. This is the case for a waiting time greater or equal to L.sub.1/v.sub.gas. The signal amplitude of the gaseous phase has a value increasing with time for smaller waiting times. The signal amplitude ratio S.sub.gas/S.sub.liquid has a maximum value for a waiting time t.sub.W=L.sub.1/v.sub.gas, the value decreases with increasing waiting time. The point in time at which both the signal amplitude ratio and the signal amplitude of the gaseous phase are at maximum corresponds to the optimum value for the waiting time t.sub.W and is t.sub.W=L.sub.1/v.sub.gas.

(10) In an iterative method, for example, it can be assumed that the gaseous phase flows twice as fast as the liquid phase, i.e., that v.sub.gas=2 v.sub.liquid=2 v.sub.bulk. This initially leads to a waiting time of t.sub.W=L.sub.1/2v.sub.bulk. After this waiting time, a nuclear magnetic measurement can be carried out and the signal amplitudes for the fast gaseous and the slow liquid phase can be determined using this measurement data. A more exact value for the flow velocity of the gaseous phase v.sub.gas can be determined from the signal amplitudes of the gaseous phase determined in this manner. This can then be used to determine a new waiting time. Using the new waiting time, a next nuclear magnetic measurement can be carried out and, again, the signal amplitudes of the respective phases as well as the flow velocity of the gaseous phase can be determined more accurately using the measurement data. This iterative method is preferably used as long as a constant value results for t.sub.W, which results in determining an optimum waiting time.

(11) In turn, a preferred implementation of the method according to the invention is characterized in that the flow velocities of the individual phases are determined using the measured values obtained by the nuclear magnetic measurement.

(12) If it is said above that the flow velocities of the individual phases, in particular the gaseous phase, are determined in an iterative method in order to define a suitable value for the waiting time t.sub.W, then it should be taken into consideration that the previously determined flow velocities do not necessarily correspond to the real flow velocity, but rather have iteratively approached the real flow velocity. If it is now said that the individual phases have been determined using the measurement values, this means that the real flow velocities of both the gaseous phase and the liquid phase are determined, since an optimum value for the waiting time t.sub.W is used. In particular, it can be provided that the flow velocities of the individual phases are determined by suitably fitting the recorded measurement data.

(13) A particularly preferred implementation of the method according to the invention is characterized in that the portions of the individual phases of the medium are determined using the signal amplitude ratio S.sub.liquid/S.sub.gas.

(14) If it is presently said that the portions of the individual phases are determined, then the relative portions a of the individual phases are meant. The signal ratio is given by

(15) $\frac{S_{\mathrm{liquid}}}{S_{\mathrm{gas}}}=\frac{{\left(HI\right)}_{\mathrm{liquid}}}{{\left(HI\right)}_{\mathrm{gas}}}\frac{\min \left(1,\frac{v_{\mathrm{liquid}}{t}_w}{L_1}\right)}{\min \left(1,\frac{v_{\mathrm{gas}}{t}_w}{L_1}\right)}=S_{0}F\left(v_{\mathrm{liquid}},v_{\mathrm{gas}},t_W\right)$
wherein HI is the hydrogen index of each phase, is the relative portion of the respective phase in the medium, v is the flow velocity, t.sub.W is the time that is waited between the spoil pulse and the beginning of measurement, and L.sub.1 is the length of the coil-shaped antenna.

(16) The term v, t.sub.W/L.sub.1 gives the relative portion of fresh medium in the coil-shaped antenna, wherein x can represent liquid or gas, i.e., thus describes the relative portion of fresh liquid medium or the relative portion of fresh gaseous medium. The term

(17) $\min \left(1,\frac{v_x{t}_W}{L_1}\right)$
means that for waiting times t.sub.W after which the entire area of the coil-shaped antenna has not yet been filled with fresh medium, the relative portion is determined using v.sub.x t.sub.W/L.sub.1. For longer waiting times, the entire area of the coil-shaped antenna is filled with fresh medium, thus the relative portion is equal to 1.

(18) A further, preferred implementation of the method according to the invention is characterized in that the nuclear magnetic measurement is implemented using a CPMG sequence. The method according to the invention, however, is not to be limited to a certain measuring sequence in the measuring device. Any measuring sequence applicable in the field of flow measurement is possible here.