AT505032B1 - METHOD AND DEVICE FOR DETERMINING THE CONVEYOR PARAMETERS OF A PARTICULAR LEADING FLUID BY FLOWING A LINE - Google Patents

Description

2 AT 505 032 B1

The invention relates to a method for determining the delivery parameters of a particle-carrying fluid flowing through a conduit, having an electrode arrangement having at least two electrodes, wherein the coupling capacitance is measured between at least two electrodes located at the peripheral region of the conduit and from the measured values at least the flow velocity of detected particles is determined.

Likewise, the invention relates to a device for determining the delivery parameters of a particle leading, flowing through a conduit fluid with an electrode located on the peripheral region of the line electrode assembly having at least two electrodes, a measuring device for real-time measurement of the coupling capacitance between the at least two electrodes as well an evaluation unit, to which the measured values measured by the measuring device are supplied and which is set up to determine at least the flow velocity of detected particles.

Sensors and measuring devices for the determination of delivery parameters, such as the conveying speed and the material concentration of conveyed material in delivery lines, are known from the prior art in a variety of embodiments, the applied principles, based on sensors are very different For example, DE 19949985 C2 describes a suitable sensor for capacitive level measurement. A structure comprising a plurality of sensor fields arranged in the form of a matrix can be found in DE 10008093 A1. From DE 19916979 A1 a method for level measurement with a multiplicity of capacitive sensors arranged side by side along a filling path is known. Also EP 0760467 A1 describes a level measurement in a pipe by means of a capacitive method.

The determination of the complex dielectric constant can take place, for example, with a device of intermeshing edge field electrodes, as disclosed in DE 4131999 C2.

Methods and devices for determining density profiles in closed conveyors are also state of the art. These methods include the class of "Electric Capacitance Tomography" sensors. An example is shown in EP 0326266, in which corresponding reconstruction methods are also disclosed. A device for capacitive measurement with the reduction of stray field effects is described in DE 4442711 A1. Here, controlled auxiliary electrodes ("active guarding") are used, which, however, brings with it a correspondingly high circuit complexity.

For measuring the flow velocity of a conveyed material flow, many methods and corresponding devices have become known. For example, DE 4025952 A1 describes the measurement of the flow velocity of fine-grained bulk solids in a pneumatic or hydraulic suspension by a non-contact measurement with capacitive sensors. In this case, two sensor electrodes of a sensor electrode are spatially opposite each other on the outside of a measuring tube, wherein an alternating voltage is applied in opposite phase to the sensor electrodes. Downstream or upstream therefrom are once again two sensor electrodes and one sensor electrode, in which case the supply takes place at a different frequency. Using phase-sensitive rectifiers and signal processing by cross-correlation, statistical fluctuations are detected and from these, the flow rate is deduced.

A similar measuring arrangement with two pairs of electrodes is known from DE 3909177 A1 as known. Methods and apparatus for determining delivery parameters by correlation flowmeters are described, for example, in "Cross-Correlation Flowmeters - Their Design and Application", M.S. Beck, A. Plaskowski, IOP Publishing Ltd, 1987. 3 AT 505 032 B1

The document DE 40 25 841 A1 describes the measurement of the mass concentration of conveyed material in pipelines by means of a determination of the relative permittivity. In this case, special, namely slotted measuring electrodes divided into partial electrodes are used. The special electrode formation should lead to a more homogeneous field distribution. Only a capacity, a frequency evaluation, speed measurement or the application of a Fourier transformation are not disclosed by the document.

In flow measurements by methods of the type specified above, a dielectric property of the conveyed material is determined at least two points in the flow direction. It is necessary that this dielectric property at each observation site have temporal fluctuations, which may be either of natural origin, e.g. B. Concentration variations in turbulent flow, or intended to be introduced, for. B. by injecting another medium in the Fördergutstrom.

Other devices or methods require the additional use of magnetic fields or free charge carriers. Thus, the document JP 2004117011 describes an electromagnetic flow measuring device in which an alternating magnetic field is generated by means of a coil. Due to the Lorentz force, a force is applied to moving charge carriers which is proportional to the speed. This in turn leads between an electrode pair to an alternating voltage signal which is amplified and evaluated by means of a Fourier transformation in order to be able to display an average conveying speed. A disadvantage is not only the high cost, due to the need for the use of coils on the one hand and electrodes on the other hand, but also the fact that this known method requires charged particles. Fluctuations, a velocity profile or the material density are also not detectable with the known method.

Magnetic fields for generating signals which are representative of the flow in a tube are also used in US 6 615 149 B1 (corresponding to WO 01/90702 A2) in addition to electrodes. Here, too, a signal processing by Fourier transformation takes place.

When material is transported in conveyors, a velocity profile and a particle concentration profile usually form over the pipe cross-section. The deviation from a uniform or constant profile can be seen particularly in pneumatically transported material under thin flow conveying conditions in horizontal and vertical pipes. In order to reliably determine the mass flow through the conveyor in multi-phase flows, it is necessary to determine the two flow parameters speed and material density quickly and accurately.

A disadvantage of the above-described known methods and devices, the fact that electrode assemblies must be provided at remote locations of a pipe, the cost of the measurement and evaluation circuits is high and installation in existing conveyor systems is complex and costly.

It is an object of the invention to provide a method and / or a corresponding device which, with little effort, can deliver reliable and practical measurements with regard to the essential delivery parameters, particularly in the case of thin-flow delivery conditions. In addition, the measurement of the delivery parameters regardless of the flow direction, d. H. be possible even with reversal of the flow direction.

This object is achieved by a method of the type mentioned in the introduction, in which according to the invention the measured values are subjected to a short-time Fourier transformation and a selected number n of the frequency components obtained is summed up in order to obtain a particle velocity and / or density value. 4 AT 505 032 B1

Likewise, the object is achieved with a device of the type mentioned at the outset, in which according to the invention the evaluation unit is set up to subject the measured values to a short-time Fourier transformation and to subject a selected number n of the frequency components obtained to a summation to obtain a particle velocity. and / or density value.

Thanks to the invention, it is in principle possible to record dielectric properties of the flow in only a single measurement plane lying orthogonally to the propagation direction of the flow and to determine a particle velocity profile of the tube cross section by evaluating the frequency spectrum. Another advantage of the invention is that it requires no additional coils for generating magnetic fields and charged particles are not a prerequisite for the measurement.

In an expedient development of the invention, it is provided that individual frequency components are weighted before summing. By this weighting an adaptation to specific conditions can be achieved, whereby individual frequencies are preferred, mitigated or even discarded.

It can also be advantageous if the particle density in the flow is determined from the number of measured values occurring per unit of time, wherein the evaluation unit is designed accordingly.

Furthermore, it can be provided that a particle size is determined from the amplitude of the measured values that occur. Statements about the particle size are important in many cases, and here too, the evaluation unit is designed accordingly.

Another expedient embodiment is characterized in that, in the case of a plurality of electrodes distributed at least along one line circumference, a velocity profile of the flow is determined by evaluating the measurement at different electrode pairs, the electrode arrangement correspondingly having a plurality of electrodes distributed along a line circumference. In this way, a spatial distribution of the measuring area can be achieved.

The sensitivity of the measurement can be increased if the electrode arrangement has a plurality of electrodes distributed along two adjoining line peripheries.

A simplified, but in many cases expedient variant provides that the electrode assembly comprises two, adjacent to each other, located on line circumferences annular electrodes.

In another variant, it is provided that the individual electrodes are connected to a controlled selection circuit which connects two electrodes each with an input of the measuring device.

The measurement on individual pairs of electrodes can be performed sequentially in time or in parallel in time.

If a number of pairs of electrodes is present, by evaluating the measurement at different pairs of electrodes, a pair of electrodes can be determined which delivers the strongest signal and this pair of electrodes can be used for further measurement.

Finally, the selected number n of the frequency components may be equal to 1, i. H. only a single frequency component is used which has previously been found to be most useful empirically. Naturally, the circuit complexity can be kept lower. 5 AT 505 032 B1

The invention together with further advantages is explained in more detail below with reference to exemplary embodiments, which are illustrated in the drawing. In this show:

1 and 1a an exemplary embodiment of the invention with eight distributed around the circumference of a tube electrodes E-i to E8 in a view normal to the axis of the tube,

2 shows an embodiment of the invention with 16 electrodes, which together form eight pairs of electrodes,

3 shows an embodiment of the invention with two continuous electrode rings,

4 and 4a an embodiment of the invention with a divided into two parts electrode ring,

5 and 5a in an illustration corresponding to FIG. 1 and 1a an invention with exemplary particle flow in the tube interior,

6a and 6b are graphs showing the time course of a measuring signal on a pair of electrodes and the associated occurrence of individual particles generating this signal;

7 shows exemplary frequency spectra for different particle velocities.

8 shows the relationship between summation values for frequency components and the mean particle velocity on the basis of a recorded characteristic curve,

Fig. 9 is a simplified block diagram of a device according to the invention and

10 is an example of a flowchart for determining the flow velocity after

Invention.

In principle, the determination of coupling capacities, which belong to a common measuring plane normal to the flow direction, can take place in various ways. Four exemplary embodiments are described in FIGS. 1 to 4 which make use of the principle on which the invention is based: in all four variants, the interior of a line LEI, here a conveying pipe, is detected in a measuring plane e. This measurement plane is formed by capacitive sensors, i. H. Electrodes which form one or more capacitors, generally pairs of electrodes, and which use as dielectric a fluid flowing in the tube and particles in this fluid.

For use, the invention may come in particular in the Dünnstrom conveyor technology, z. B. gravel, plastic granules, wood chips, etc. So, in particles that are usually transported with a high proportion of conveying air and high conveying speeds. Usual pipe diameters are in the order of about 10 cm. However, the invention can be easily adapted for any small / large diameter, although the fluid has corresponding particle sizes. The conveying speeds can range from a few cm / s to many m / s. Here, too, an extension of the measuring range up and down by adjusting the measuring rate is possible.

By dividing the electrodes along the circumference of a line, a spatial resolution of the flow cross section can be achieved. According to Fig. 1 and 1 a eight electrodes E ^ to E8 are uniformly distributed along a pipe circumference. The coupling capacities Ci2 (between E and E2), Ci8 (between Ei and E8), C27 (between E2 and E7), Ci7 (between Ei and E7) and C78 (between E7 and E8) are plotted. The sensitive area for the capacitance Ci7 is shown in FIG. 1a by the indication of field lines - changes in the flow in this area change the value of the capacitance Ci7 in particular. It is understandable that the coupling capacity for different electrode combination is determined by different regions of the cross section. This circumstance can be exploited to determine spatial distributions.

The embodiment according to FIG. 2 shows a measuring plane e, which is associated with eight electrodes E1a... E8a and E1b... E8b located at mutually adjacent circumferences. As in the embodiment according to FIG. 1, the use of different switching patterns and combinations of electrodes, in particular of neighboring electrodes, can also be used to determine a profile. The formation of the electric field takes place in this embodiment mostly 6 AT 505 032 B1 parallel to the flow direction, provided that adjacent electrodes form a pair as a capacitor, z. B. the electrodes E1a and E1b.

The simple embodiment of the invention according to Fig. 3 shows two continuous, parallel and adjacent to each other, annular electrodes Ea and Eb, which form a single capacitor whose electric field can form parallel to the flow direction.

In the likewise simple embodiment of the invention illustrated in FIGS. 4 and 4a, only two, almost semicircular electrodes Eo and Eu are provided along a circumference of the line LEI whose plane of symmetry normal to the pipe axis is the measuring plane e.

Although it is not possible to reconstruct a distribution of the velocity and / or the concentration over the pipe cross-section in the embodiments according to FIGS. 3 and 4, a determination of the highest and lowest occurring particle velocity as well as an average particle velocity and average particle concentration is also possible with these simple arrangements. However, these simple embodiments are particularly recommended because of their low switching and billing expense and sufficient in many cases.

The attachment of the electrodes in the region of the measuring plane is by no means limited to the outer circumference of the conveying tube. If the conveyor tube consists of a metallic conductor, the electrodes can also be mounted on the inside, electrically insulated from the tube, and the measuring leads leading from the electrodes to a measuring device can be led out. For electrically non-conductive production pipes, in practice, due to the possibly abrasive behavior of the material to be conveyed, an embodiment of the pipe with electrodes incorporated in the pipe wall or mounted on the outside of the pipe will be most sensible.

To determine the coupling capacity, several methods are available, which are well known to the skilled person and belong to the prior art, for. B. Measurement by means of measuring bridges, potential measurement, "capacitance-to-digital" converter, detuning a resonant circuit, etc.

As already mentioned, in embodiments of the invention selectable electrodes, i. H. be provided a capacitor-forming electrode pairs. The capacity between the selected transmitting and receiving electrode is z. B. determined by applying one high-frequency signal to one electrode and receiving a signal at the other electrode of the pair whose amplitude depends on the known distance between the electrodes and the nature of the dielectric (i.e., εΓ of the tube contents). In Fig. 1, the propagation of the electric field through the interior of the tube has already been shown when the electrodes E, and E7 are used as receiving and transmitting electrodes, respectively. Particles that in this configuration in the sensitive area of the tube, d. H. are in the propagating field, provide according to their speed a spectral contribution to the measurement result.

At this point, FIGS. 9 and 10 anticipate in order to better explain the measuring method and a device used therefor. According to FIG. 9, z. B. eight electrodes Εί ... E8 attached to an outer circumference of a pipe LEI, in which a particle laden fluid can flow, such as a wood chips leading air flow. The electrodes are connected to a selection circuit which follows a capacitance measuring device CME and whose output signals are fed to an evaluation unit AWE. This evaluation unit AWE can in turn control the selection circuit. The result determined by the evaluation unit can be fed to a display / recording unit AZE.

In the invention, changes in the nature of the material flow are determined by capacitive means. This means that every particle with a dielectric constant εΓ greater than 1 provides a contribution to the capacitive result of the measurement. The invention is applicable to both dense and for thin flow promotion. With pneumatic dense phase flow promotion, only small differences in the particle velocities of individual material plugs occur, and the particle concentration is also very homogeneous. In contrast, a promotion under thin-stream conditions on a variety of different particle velocities and particle distributions in the delivery pipe. The particular advantages of the present invention are thus usable, especially in the Dünnstromförderung.

A particle concentration or material density distribution in the tube is achieved by the proportionality between the number of particles in the sensitive area between two electrodes and the measurable electrode capacity: with denser occurrence of particles, the dielectric constant of the material lying between the measuring electrodes is increased, and thus the measured one also increases Capacity. For thin-flow conveying processes granular material, the number of particles in the delivery volume is very low, the particle concentration can also be determined here by the signal fluctuations caused by individual particles are detected and counted.

The measurement plane e has a small but finite dimension in the flow direction, d. H. In any case, the word level is not to be understood in a geometric sense. Fast-moving particles have a correspondingly shorter residence time in this sensitive measuring range. Slowly transported particles are in the sensitive area of the measuring electrodes for a long time.

It is now well known that with mathematical methods (eg the Fourier transformation) the time domain of a signal can be transformed into the frequency domain.

Slow processes or long-lasting signal characteristics can be found in the frequency spectrum in the low frequency range, fast processes or short-duration signal characteristics in the higher frequency range. Static operations, d. H. Operations that cause a constant signal waveform are represented as a DC component in the frequency spectrum. The transformation from the time domain of the signals in their frequency domain is carried out, for example, by a short-time Fourier transformation, d. H. the changes of the spectrum over time are evaluated.

The different speeds of the particles, or the different residence time of the particles within the measurement plane, are evaluated in the invention by the fact that the frequency spectrum of the capacitive measurement data, d. H. the output signal of the capacitance measuring device CME is analyzed.

The possibility of determining the velocity and density distribution in the conveying device is to be clarified with reference to FIGS. 5 and 5a by the particle concentration varying over the pipe cross section and the different particle velocities. For the determination of the particle concentration in Fig. 5, for example, the capacitance between electrodes E2 and E3 is less than the capacitance between electrodes E6 and E7 because the particle concentration increases towards the bottom of the tube and more particles than dielectric (e.g. , Partikei = 6) are available. Likewise, the determined particle velocity will have different values in different sensitive measuring ranges.

FIGS. 6a and 6b show by way of example that, for the slow particle P2, the residence time t2 in the measurement plane is longer than the residence time ti for the fast particle Pi. As shown in FIG. 6, from the measured capacitance values over time both a statement about the frequency of occurring particles, i. H. a material concentration or, if several electrodes distributed around the circumference are used in the measuring plane, a material concentration profile, a statement about the size of the particles in the measuring plane, d. H. on the amplitude of the signals of the measurement plane e, with large particles correspondingly more volume with increased εΓ and cause a higher signal amplitude, as well as - a statement about the residence time of the particles and thus their frequency components or speed, or velocity profile when using several order the circumference distributed electrodes are taken in the measurement plane.

FIGS. 7 a to d show exemplary signals in the frequency domain, which were determined by using the Fast Fourier Transformation (FFT) or "Short-time Fourier Transform" in FIG. 10. It can be seen that with increasing speed of the conveyed medium (higher particle velocities), the higher frequency components increase. By summing several frequency components, the noise exhibited by these signals can be reduced, giving the summation value. The result is a relationship between speed and summation value, as shown in Fig. 8. Such a characteristic curve can be recorded in the course of a calibration of the device. Furthermore, it is then possible with the aid of this characteristic curve to determine from the summation value the conveying speed or also the volume flow (depending on the selected calibration). Due to the approximately linear course of the characteristic curve, it is usually sufficient to multiply the summation value by a factor in order to determine the particular measured variable sought.

The choice of operation is determined by a previously defined switching pattern (FIG. 10) and the selection circuit AWS. If one or more electrodes are operated as transmitters, i. H. the corresponding electrode is acted upon by a high-frequency signal, forms, in accordance with the dielectric of the tube contents, an electric field in the tube interior. Partial volumes of the tube contents containing higher dielectric constant material prefer the propagation of the electric field compared to areas where only air prevails. For example, if there is material with a higher εΓ between the electrodes E5 and E6 and air between the electrodes E6 and E7, the propagation of the electric field between E6 and E5 is preferred when operating the electrode E6 as transmitter and E5 and E7 as receiver and thus also the measurable Capacity between E5 and E6 increased. In an arrangement according to FIG. 1 or FIG. 4, the electric field therefore develops mainly in the cross-sectional plane of the tube, i. H. normal to the flow direction.

In contrast, in Fig. 2, the capacitive plane e is realized by two rings of electrodes. All electrodes of a ring work in the same mode, either as a receiver or as a transmitter and form a capacitive sensor or a pair of electrodes with the electrode located in the adjacent ring. The transmission electrodes of these sensors are either parallel in time, d. H. Frequency division multiple access (FDMA) or code division multiple access (CDMA) or sequential, d. H. Time division multiple access (TDMA) with a high-frequency alternating signal driven. In an arrangement as in FIG. 2, the electric field develops mainly parallel to the flow direction, wherein the distance between the two electrode rings determines the penetration depth and the sensitivity of the arrangement for central regions or outer regions of the tube contents.

Before the actual measurement, it is necessary to select those frequencies in Fourier space where the relationship between velocity and relative spectral amplitude is most accurately determined. In the simplest case, a linear relationship between these quantities is assumed, the measurement uncertainties and the proportionality constants can then be determined experimentally. A theoretical determination is conceivable with knowledge of the movement properties of the conveyed material. Similarly, the relationship between signal energy (the temporal squared average of the signal) and the particle concentration can be determined. With low particle concentration, it is also possible to determine the number of particles by counting the signal peaks.

In the actual measurement, a switching pattern is first selected. Depending on the switching pattern, different segments are actuated and thus the spatial distribution of the sensitivity of the sensor changes. The capacity curves are recorded and from this the particle concentration within the sensitive area is determined by the previously determined relationship. Furthermore, the signal profiles are each subjected to a short-time Fourier transformation. The result is normalized with the particle concentration. From the normalized signals, the speed of the conveyed material within the sensitive area can be determined via the previously determined relationship.

It should be noted that in cross-correlative methods, as they belong to the above-mentioned prior art, signals are tapped at two different locations in order to determine a transit time between these levels, the invention, however, only with a "measurement level "E works.

The advantage of adjacent electrodes, as shown in FIGS. 2 and 3, lies in the fact that the field lines form substantially parallel to the flow direction and thereby obtain a more favorable spatial distribution of the sensitivity. In order to be able to exploit this for cross-correlation, one would have to use a second such arrangement with a total of 4 or 3 electrode rings.

The Fourier transformation is in theory an integral "over all time" which, of course, can not be realized in practice and is also constant for a signal. Since the signals are taken only over a short period of time and a Fourier transform is calculated therefrom, the short-term Fourier transform can change over time and thus a change in velocity over time can be detected. Mathematical: with x [i] as the i-th measured value, N the number of recorded (evaluated) measured values and the discrete frequency k.

The determination of the speed is carried out as described by the evaluation of the short-time Fourier transform KFT. You can z. B. use a weighted sum over the amplitudes of all frequencies such that: where c represents a proportionality constant and g (i) represent weights, in the simplest case all weights are g (i) = 1. These weights will, in principle, be a selection of Frequencies hit, whereby also all weights can be up to a zero, which means that only a single spectral line is evaluated.

Of course, in order to determine the speed from the measurement data obtained by the transformation, the prior art also offers numerous general estimation methods which differ with regard to the achievable accuracies and the required computational effort. In the simplest case, as stated above, a weighted or unweighted sum over the amplitudes of the frequency components can be used as an estimator for the speed; this estimator is linear and advantageous due to the low computational complexity. Another possibility is, for example, so-called maximum-likelihood (ML) estimators, in which, for a given course of the frequency components, by solving a maximum value problem, that speed is calculated which has the highest probability of supplying exactly that course of the frequency components. The determination of the ML estimator generally requires comparatively complex, iterative and non-linear arithmetic operations.

Claims (20)

Thanks to the use of a single measuring plane and the associated symmetrical construction in the flow direction (see Fig. 1 to Fig. 5), the invention without modifications for the measurement in both directions, z. B. in pressure and suction promotion, are used. It should also be noted that it may be useful in certain cases to increase the sensitivity, by evaluating the measurement at different pairs of electrodes to determine a pair of electrodes, which provides the strongest signal and to drive the selection unit so that this electrode pair is used for further measurement. 1. A method for determining the delivery parameters of a particle leading, flowing through a conduit fluid, with an electrode assembly having at least two electrodes, wherein the coupling capacitance between at least two, located at the peripheral region of the line electrodes is measured and from the measured values at least the flow velocity of detected particles is determined by subjecting the measured values to a short-term Fourier transformation and summing a selected number n of the frequency components obtained in order to obtain a particle velocity and / or density value. 2. The method according to claim 1, characterized in that individual frequency components are weighted before summing. 3. The method according to claim 1 or 2, characterized in that the particle density in the flow is determined from the number of measured values occurring per unit time. 4. The method according to any one of claims 1 to 3, characterized in that from the amplitude of the measured values occurring a particle size is determined. 5. The method according to any one of claims 1 to 4, characterized in that in a plurality, at least along a line circumference distributed electrodes by evaluating the measurement at different pairs of electrodes, a speed and / or density profile of the flow is determined. 6. The method according to any one of claims 1 to 5, characterized in that by evaluating the measurement at different pairs of electrodes, a pair of electrodes will determine which delivers the strongest signal and this pair of electrodes is used for further measurement. 7. The method according to claim 5 or 6, characterized in that the measurement is carried out at individual electrode pairs in time sequentially. 8. The method according to claim 5 or 6, characterized in that the measurement is performed on individual pairs of electrodes in time parallel. 9. The method according to any one of claims 1 to 8, characterized in that the selected number n of the frequency components is equal to 1. 10. Device for determining the delivery parameters of a particle-carrying fluid flowing through a conduit (LEI), having an electrode arrangement located at the peripheral region of the conduit, comprising at least two electrodes (Ε ^, Εβ; E1a, E1b ... E8a, E8b : Ea, Eb: E0, Eu) has 1 1 AT 505 032 B1 of a measuring device (CME) for real-time measurement of the coupling capacitance between the at least two electrodes as well as with an evaluation unit (AWE), to which the measured values measured by the measuring device are fed and which for Determination of at least the flow velocity of detected particles is set up characterized in that the evaluation unit (AWE) is adapted to subject the measured values to a short-time Fourier transformation and to subject a selected number n of the obtained frequency components to a summation to obtain a particle velocity and / or or to get density value. 11. The device according to claim 10, characterized in that the evaluation unit (AWE) is adapted to weight the individual frequency components obtained before the summation. 12. Device according to claim 10 or 11, characterized in that the evaluation unit (AWE) is set up to determine the particle density in the flow from the number of measured values occurring per unit of time. 13. Device according to one of claims 10 to 12, characterized in that the evaluation unit (AWE) is adapted to determine from the amplitude of the measured values occurring a particle size. 14. The device according to one of claims 10 to 13, characterized in that the electrode arrangement comprises a plurality of distributed along a line circumference electrodes ((E1 ... E8). 15. Device according to one of claims 10 to 13, characterized in that the electrode arrangement has a plurality, along two adjacent mutually adjacent line circumferences distributed electrodes (Eia, E1b ... E8a, E8b). 16. The device according to one of claims 10 to 13, characterized in that the electrode arrangement has two, adjacent to each other, located on line circumferences annular electrodes (Ea, Eb). 17. The apparatus of claim 14 or 15, characterized in that the individual electrodes are connected to a controlled selection circuit (AWS), which connects two electrodes each with an input of the measuring device (CME). 18. The device according to claim 17, characterized in that the evaluation unit (AWE) is adapted to determine by evaluating the measurement at different pairs of electrodes, a velocity and density profile of the flow. 19. The device according to claim 17, characterized in that the evaluation unit (AWE) is adapted to determine by evaluating the measurement at different pairs of electrodes, a pair of electrodes, which provides the strongest signal and the selection unit to control so that this pair of electrodes used for further measurement becomes. 20. Device according to one of claims 10 to 19, characterized in that the selected number n of the frequency components is equal to 1. For this purpose 3 sheets of drawings