WO2010085980A1

WO2010085980A1 - Coriolis flowmeter and method for calculating the gas fraction in a liquid

Info

Publication number: WO2010085980A1
Application number: PCT/EP2009/000624
Authority: WO
Inventors: Thomas Bierweiler; Wolfgang Ens; Henning Lenz
Original assignee: Siemens Aktiengesellschaft
Priority date: 2009-01-30
Filing date: 2009-01-30
Publication date: 2010-08-05

Abstract

The invention relates to a Coriolis flowmeter (1) and to a method for calculating a gas fraction in a liquid (9) flowing through a line (3). The flowmeter comprises a measuring unit (2) that determines a measured value and an evaluation unit (17) that is connected to the measuring unit (2) and that comprises a data memory (21) and a processor (19). The invention is characterised in that a maximum curve (MAX) and a minimum curve (MIN) calculated from measured values of the actuator current and the vibration amplitude for known gas fractions are stored in the data memory (21), the determined measured value being inserted into the maximum curve (MAX) by means of the processor, thus determining a maximum limit and being inserted into the minimum curve (MIN) by means of said processor, thus determining a minimum limit of the actuator current and of the vibration amplitude versus the gas fraction. The gas fraction in the liquid is calculated from the maximum limit and the minimum limit and is displayed.

Description

description

CORIOLIS MASS FLOWMETER AND METHOD FOR CALCULATING GAS IN A LIQUID

The invention relates to a Coriolis mass flowmeter for measuring the mass flow of a liquid, wherein the liquid may contain a gas portion. The invention also relates to a corresponding method for operating a Coriolis mass flow meter.

A Coriolis mass flow meter is known from DE 10 2004 014 029 A1. In the process, measuring and automation technology, Coriolis mass flowmeters are often used for the measurement of physical parameters of a medium flowing in a pipeline, by means of a measuring transducer inserted in the course of the medium-carrying pipeline and through which the medium flows during operation Measuring operation circuit with the density of corresponding inertial forces cause and derived from these generate a respective mass flow measurement signal. For guiding the medium, the measuring sensors each comprise at least one measuring tube held in a supporting frame with a bent or straight tube segment which is vibrated during operation in order to produce the above-mentioned reaction forces. For detecting the vibrations of the pipe segment, the sensors also each have a responsive to movements of the pipe segment, physical-electrical sensor arrangement. The measurement is based on the medium flowing through the measuring tube inserted into the pipeline and oscillating during operation, whereby the medium experiences Coriolis forces. These in turn cause the inlet-side and outlet-side regions of the measuring tube to oscillate out of phase with one another. The size of these phase shifts serves as a measure of mass flow. The oscillations of the measuring tube are therefore detected by means of two vibration sensors of the aforementioned sensor arrangement which are spaced apart from one another along the measuring tube, and in Vibration measurement signals converted from the mutual phase shift of the mass flow is derived.

When using mass flow meters of the vibration type, however, it has been shown that in the case of non-homogeneous media, especially two- or multiphase media, the variables derived from the vibrations of the measuring tube, in particular also the mentioned phase shift, are subject to considerable fluctuations and thus, if necessary can be completely unusable for the measurement of the respective physical parameter without remedial measures. Such inhomogeneous media may be, for example, liquids in which a gas present in the pipeline, in particular air, is introduced. DE 10 2004 014 029 A1 proposes, in order to solve this problem, to generate an intermediate value derived from the exciter current, which represents an apparent viscosity, on the basis of which a correction factor for the mass flow meter is derived.

In the article "New parameters for the analysis of the disturbances of a Coriolis mass meter", Technical Measurement, 74, 2007, pp. 577-588, various measures are presented for the detection of gas fractions in a liquid whose flow is to be measured The amplitude characteristic is then defined as the difference between the accumulated useful frequencies and the accumulated noise frequencies, as further characteristic numbers are: the ratio of the excitation current to the sensor voltage, the amplitude characteristic is defined as the discrete spectrum of the sensor signal of one of the two sensors. the ratio of the excitation current without gas bubbles to the excitation current with gas bubbles, the ratio of the sensor voltage with gas bubbles to the sensor voltage without gas bubbles and the shift of the resonance frequency.

WO 2006/130415 A and WO 2006/127527 A describe methods for the detection and counting of gas bubbles by means of additional excitation of other frequencies and comparison of the test results. word with a reference spectrum. The determination of both the mass flow rate and the density of the liquids and the gas is thereby possible. A problem with this approach is that it requires the use of a suitable excitation source.

DE 10 2006 017 676 B3 discloses a method for operating a Coriolis mass flow meter. A two-phase flow is detected by using different characteristics from at least two different groups. The groups are subdivided as follows: magnitudes based on friction losses within the multiphase flow, magnitudes based on inhomogeneity of the multiphase flow, magnitudes based on a modeling of the nominal behavior of the Coriolis mass flow meter, and magnitudes based on transit time effects of the multiphase flow between two longitudinally of the measuring tube spaced detection points for the resulting vibrations. Although this method allows the recognition of a two-phase flow but not its quantification.

In US 2005/0284237 Al a method for correcting the error of the mass flow measurement is described. A neural network, which has temperature, attenuation and density drop as input parameter, corrects the error of the mass flow meter. This achieves only a partial compensation, that is, the error remains limited, but the specification is abandoned. The method also requires considerable computing and memory costs.

The object of the invention is to provide a Coriolis mass flow meter, by means of which a gas component can be determined when flowing through with a liquid. Another object of the invention is to specify a method for operating a Coriolis mass flow meter, in which a gas component can be determined when flowing through a liquid. According to the invention, the object directed to a Coriolis mass flowmeter is achieved by a Coriolis mass flowmeter for measuring the mass flow of a liquid through a line, with a measuring unit which determines a measured value of an actuator current and / or a vibration amplitude, and with a measuring unit associated evaluation unit, which comprises a data memory and a processor, wherein in the data memory, a maximum curve and a minimum curve are stored, wherein by means of the processor, the determined measured value used on the one hand in the maximum curve and calculates a maximum limit and on the other hand inserted into the minimum curve and from this a minimum barrier is calculated and wherein from the maximum barrier and the minimum barrier, a gas content in the liquid is calculated and displayed.

An evaluation unit of a Coriolis mass flowmeter calculates or measures u. a. the natural frequency and the phase difference between the incoming and outgoing pipe part of the vibrating tube.

The invention is based on the recognition that certain features of the measurement of a Coriolis mass flow meter over a wide range of flow rates and different types of liquid / gas mixtures produce a characteristic signature that allows a quantitative inference to the gas content. For certain features, such. As for the Aktorstrom and the amplitude estimate, resulting in dependence on the gas content characteristic changes that are valid for different flow directions (horizontal, vertical upward, vertically downward) and different back pressures. This characteristic signature can be mapped by including the range of feature values as a function of the gas fraction between a maximum curve and a minimum curve. By now storing this maximum curve and minimum curve as limitations of the feature values in the memory, by inserting the determined measured value into these curves, an interval of a possible gas flow can be established. partly determined and from this the gas content can be determined with a defined error.

As soon as gas components (gas void fraction) are conveyed during the mass flow measurement of liquids, the Coriolis mass flow meter i. A. erroneous readings. The proposed method can be used to estimate the proportion of entrained gas and the uncertainty of this information. This enables condition monitoring of both the field device and the process. In addition to detecting a faulty measurement, this provides the operator with valuable information about his process status.

The maximum curve and the minimum curve are preferably calculated by forming the inverses of an envelope from a series of measured values previously determined for different gas fractions. More preferably, the envelope is formed by one of the following types of functions: a linear or piecewise linear curve, an n-th order polynomial, a spline, a neural network, an exponential or a logarithmic curve. An envelope is thus modeled in a known form from a set of measured values. For this purpose, measured values are preferably determined once depending on known gas fractions. The maximum curve and the minimum curve are then calculated by inverting the envelopes.

Preferably, respective maximum and minimum limits are calculated from measured values of the actuator current and the oscillation amplitude and their respectively associated maximum and minimum curves, and the smallest maximum barrier and the largest minimum barrier are used for the calculation of the gas fraction. More preferably, an error indication is calculated and displayed from the difference between the largest maximum barrier and the smallest minimum barrier. Preferably, the gas fraction is calculated as the arithmetic mean of the smallest maximum and the largest minimum barrier.

Preferably, the Coriolis mass flow meter is designed to measure digitally. As a rule, the described calculation of the gas fraction can already be implemented without hardware changes via the processor and memory by depositing a software suitable for carrying out the method in the program memory of the device.

The determined measured values are preferably filtered before the further processing.

According to the invention, the object directed to a method is achieved by a method for operating a Coriolis mass flowmeter for measuring the mass flow of a liquid through a line with a measuring unit which determines a measured value of an actuator current and / or a vibration amplitude, and one with the Measuring unit connected evaluation unit, which comprises a data memory and a processor, wherein in the data memory, a maximum curve and a minimum curve are stored, by means of the processor, the measured value used on the one hand in the maxima ximum curve and calculates a maximum limit and on the other hand used in the minimum curve υ (From this a minimum barrier is calculated and from the maximum barrier and the minimum barrier, a proportion of gas in the liquid is calculated and displayed.

The invention will be explained in more detail with reference to the drawing in an embodiment. In part, they show schematically:

1 shows a Coriolis mass flowmeter, FIG. 2 shows a flowchart for measuring a gas fraction,

3 shows measured values of the actuator flow as a function of the gas fraction for different flow rates, 4 shows envelopes for measured values of the actuator flow at different gas fractions,

FIG. 5 shows envelopes for the measured values of the oscillation amplitude for various gas components in logarithmic representation and FIG

6 shows a comparison of calculated gas fractions with the actual gas fractions.

1 shows a Coriolis mass flow meter 1. The Coriolis mass flow meter 1 has a measuring unit 2 in a line 3. The line 3 splits at a flow measuring point 4 into two sub-lines 4A, 4B. The two sub-lines 4A, 4B extend parallel to each other in an approximately radial direction away from the flow direction in the line 3, then run parallel to this flow direction and then extend again approximately radially to the flow direction.

An excitation unit 13 is connected to the flow measuring point 4 with an excitation coupling 14 so that they can put them into vibration. For this purpose, a value of an actuator current is set in a control, which is required to excite the partial lines 4A, 4B to oscillations with a certain amplitude. An evaluation unit 17 is connected to the flow measuring point 4 by an evaluation coupling 15 and an evaluation coupling 16 so that they can measure the deflections on the one hand of the sub-line 4A and on the other hand, the sub-line 4B or their relative position to each other. The evaluation unit 17 has a processor 19 and a data memory 21. The evaluation unit 17 is connected to a display unit 23.

During operation of the Coriolis mass flow meter, a liquid 9 flows through the line 3 and divides into the partial lines 4A, 4B into two partial streams 9A, 9B. As described above, the measuring principle for the mass flow rate and the density is based on the fact that the inertial forces acting on the partial flows change the relative phase position of the Vibrations of the sub-lines 4A, 4B to each other, which are detected by the Auswertekopplungen 15 and 16, or cause changes in the natural frequency of the vibration of the flow measuring point 4, which are detected by the evaluation unit 17. A proportion of gas in the liquid 9 can significantly falsify this measurement. Therefore, a determination of this proportion of gas is essential. The following describes how this gas content can be determined easily.

2 shows a flow chart for a method for determining the gas content. On the basis of the measurement signals which are generated by the evaluation couplings, the actuator current IA in step SO1 and the oscillation amplitude AE in step S02 are measured. The measured values are filtered in steps S03 or S04 and then used in steps S05 and S06, respectively, in assigned, inverse hull curves stored in the data memory 21 (FIG. 1). The assigned hull curves were previously modeled from measured values present in known gas fractions in step S07, inverted in step S08 and stored in step S09 as inverse hull curves HK ^-1 UA) and HK ^"1 (AE) in data memory 21 (FIG Maximum in-line curves (MAX) and minimum curves (MIN) in the data memory 21 (FIG. 1), into which the measured values of the actuator current IA and the oscillation amplitude AE In this case, different types of hull curves can be used, for example, hull curves with different types of functions, such as polynomials of the nth order or splines, on the basis of the minimum curve stored for the actuator current

(MIN) and maximum curve (MAX) of the gas component, a lower limit GVF _m1n (IA) and an upper limit GVF _ma χ (IA) are determined for the measured actuator current. Analogously, the measured oscillation amplitude AE is evaluated. For the gas content GVF, intermediate values result between the limits GVF _min (IA) and GVF _max (IA) and between the limits GVF _min (AE) and GVF _max (AE). From this, a maximum barrier GVF _{max is determined} as the smallest of the values of GVF _max (IA) and GVF _max (AE). Continue a minimum bound GVF _{mIn is determined to} be the largest of the values of GVF _1nIn (IA) and GVF _mIn (AE). Finally, in step S, the final value of the gas fraction GVF is calculated as half the sum of the maximum and minimum limits, that is, (GVF _m i _n + GVF _max ) / 2. Advantageously, an error indication is also possible in a simple manner with this method, calculated from the difference of the largest value of GVF _ma x (IA) and GVF _max (AE) and the smallest value of GVF _m in (IA) and GVF _rain (AE ). The gas component GVF is displayed with the error indication in step S12 on the display unit 23 (FIG. 1) and output in step S13, for example via a field bus, to a higher-level control station in an automation system for further processing.

3 shows a series of measured values 30 of the actuator current IA as a function of the gas content. On the abscissa, the gas components are plotted as gas-void fraction GVF, on the ordinate actuator currents IA. From these measurements, the maximum values are taken to determine a maximum envelope and the minimum values to interpolate a minimum envelope. This is illustrated in FIG. 4, wherein the envelopes HK _ma χ (IA) and HK _m i _n (IA) were modeled as section-wise linear curves. By forming the inverse function of HK _max (IA), a maximum curve MAX, not shown, and a minimum curve MIN by inverting the function HK _min (IA) are determined. FIG. 5 shows a representation of the oscillation amplitude AE with envelopes HK _max (AE) and HK _m i _n (AE) with logarithmic scaling of the ordinate, the envelopes being determined substantially analogously to the procedure described with reference to FIGS. 3 and 4.

FIG. 6 shows how the calculated gas fraction GVF represented by crosses 61 coincides with an actual gas fraction GVF, which is indicated by points 60. A good match is achieved. On the abscissa the number of the measurement N, on the ordinate the respective gas content GVF in% is plotted. The embodiment thus describes a combination of two mutually constructive aspects: first, the selection and determination of features that are particularly suitable. The second aspect arises from the way in which these features are used with envelopes to estimate a minimum or maximum gas fraction as a model for gas fraction determination. This results in the following advantages:

- Unlike other methods, no hardware changes are necessary. - No additional sensors or actuators are required.

- The control loop of the excitation unit does not need to be changed.

- Only existing sensor signals are used or evaluated.

The algorithm requires little CPU and memory resources.

The algorithm can be easily implemented in a DSP. - The algorithm can be used for a wide range of flow rates, flow directions (horizontal, vertical upward and vertical downward flow) and back pressures.

- The algorithm can be used for different media (eg water, oil).

- A diagnostic statement about the process status and the field device is provided:

- From a certain gas content, the field device is out of specification. This means that the flow measured value is no longer reliable and should no longer be considered as valid process information. The knowledge of the gas content is also the basis for a compensation of the measurement error. - The indication of the gas content also diagnoses the core problem: it is gas (eg air) in the process, thus providing the starting point for process optimization.

Claims

claims

1. Coriolis mass flowmeter (1) for measuring the mass flow of a liquid (9) through a line (3), with a measuring unit (2) for determining at least one

Measured value of an actuator current and / or a vibration amplitude and with an evaluation unit (17) connected to the measuring unit (2), which comprises a data memory (21) and a processor (19), characterized in that in the data memory (21) a maximum curve ( MAX) and a minimum curve (MIN) are stored, and that the evaluation unit (17) is designed such that the measured value determined on the one hand in the maximum curve (MAX) used and determines therefrom a maximum barrier and on the other hand in the minimum curve (MIN) used and from this a minimum barrier is determined and that from the maximum and the minimum barrier a proportion of gas in the liquid is calculated and displayed.

2. Coriolis mass flowmeter (1) according to claim 1, characterized in that the maximum curve (MAX) and the minimum curve (MIN) is calculated by forming the inverse of an envelope (HK) from a series of measured values.

3. Coriolis mass flowmeter (1) according to claim 2, characterized in that the envelope (HK) is formed by one of the following types of functions: a linear or piecewise linear curve, an n-th order polynomial, a spline, a neural network, an exponential or a logarithmic curve.

4. Coriolis mass flowmeter (1) according to one of the preceding claims, characterized in that from measured values of the Aktorstroms and the oscillation amplitude of these respective assigned envelopes (HK) are calculated that with these envelopes (HK) respectively associated maximum and minimum barriers are determined and that for the calculation of the gas fraction, the smallest maximum barrier and the largest Minimum barrier can be used.

5. Coriolis mass flowmeter (1) according to claim 4, characterized in that from the difference of the largest maximum limit and the smallest minimum limit an error is calculated and displayed.

6. Coriolis mass flowmeter (1) according to any one of the preceding claims, characterized in that the proportion of gas is calculated as the arithmetic mean of the maximum and the minimum barrier.

7. Coriolis mass flowmeter (1) according to one of the preceding claims, which is formed digitally measured.

8. Coriolis mass flowmeter (1) according to one of the preceding claims, characterized in that the measured values are filtered prior to further processing.

9. Method for operating a Coriolis mass flowmeter (1) for measuring the mass flow rate of a liquid (9) through a line (3), comprising a measuring unit (2) which is used to determine at least one measured value, an actuator current and / or an oscillation amplitude, and with an evaluation unit (17) connected to the measuring unit (2), which comprises a data memory (21) and a processor (19), characterized in that in the data memory (21) a maximum curve (MAX ) and a minimum curve (MIN) are deposited, wherein by means of the processor, the determined measured value on the one hand in the maximum curve (MAX) used and determines a maximum barrier and on the other hand inserted in the minimum curve (MIN) and from this a minimum barrier is determined and where the maximum barrier and the minimum barrier, a proportion of gas in the liquid is calculated and displayed.