METHOD FOR COMPENSATING THE INFLUENCE OF THE REYNOLDS NUMBER ON THE MEASUREMENT OF A CORIOLIS MASS FLOW METER, AND CORRESPONDING DEVICE

Abstract

A method for compensating the influence of at least one of the parameters from the group consisting of flow rate, viscosity, density and Reynolds number of a fluid to be measured on the measured flow rate and/or density of this fluid in a Coriolis mass flow meter with the aid of an equation using the parameters of the current Reynolds number of the fluid to be measured in the Coriolis mass flow meter, the maximum compensation value for Reynolds numbers approaching zero, the Reynolds number at which the curve of the compensation value has the largest slope, and the slope of the curve of the compensation value at the point Re.sub.c. Moreover, the invention relates to a Coriolis mass flow meter with a control device for carrying out the method.

Claims

1-10. (canceled)

111. A method for compensating the influence of at least one of the parameters from the group consisting of flow rate, viscosity, density and Reynolds number (Re) of a fluid to be measured, on the measured flow rate and/or density of this fluid in a Coriolis mass flow meter, comprising: determining a compensation value (M.sub.f(Re)) for the flow rate and/or density of the fluid to be measured using the following equation ${M_{f} (R e) = \frac{f_{\max}}{2} .Math. {1 - \tanh (- \frac{2}{f_{\max}} .Math. \frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}} .Math.  [\log_{1 0} (R e) - \log_{1 0} (R e_{c})])},$ where f.sub.max is the maximum compensation value (M.sub.f(Re)) for Reynolds numbers (Re) approaching zero, Re.sub.c is the Reynolds number (Re) at which the curve of the compensation value (M.sub.f(Re)) has the largest slope, and ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}}$ is the slope of the curve of the compensation value (M.sub.f(Re)) at the point Re.sub.c, and correcting the value measured by the Coriolis mass flow meter for the flow rate and/or the density of the fluid to be measured using the compensation value (M.sub.f(Re)).

212. The method according to claim 11, wherein determining the compensation value (M.sub.f(Re)) and correcting the measured value is performed during operation of the Coriolis mass flow meter by a control device of the Coriolis mass flow meter.

313. The method according to claim 11, wherein for correcting the measured value, the compensation value (M.sub.f(Re)) is interpreted as a relative deviation of the measured value, in particular as a negative relative deviation, and that the absolute deviation determined from it is added to the measured value.

414. The method according to claim 11, comprising acquiring the value of a current Reynolds number (Re) of the fluid to be measured in the Coriolis mass flow meter during operation.

515. The method according to claim 11, wherein determining the compensation value (M.sub.f(Re)) is preceded by experimentally determining the parameters f.sub.max, Re.sub.c and ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}} .$

616. The method according to claim 15, wherein the experimental determining is performed exclusively in a range in which the fluid to be measured in the Coriolis mass flow meter has a Reynolds number (Re) of the order of magnitude of Re.sub.c up to at least 10.sup.5, in particular in a range in which the lower limit for the Reynolds number (Re) corresponds exactly to Re.sub.c.

717. The method according to claim 11, wherein determining the compensation value (M.sub.f(Re)) is preceded by determining the parameters f.sub.max, Re.sub.c and ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}}$ using the finite element method (FEM).

818. The method according to claim 11, wherein only the current Reynolds number (Re), f.sub.max, Re.sub.c and ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}}$ are used as input parameters for compensating the influence of the at least one parameter of the group.

919. The method according to claim 11, wherein the determining of the compensation value (M.sub.f(Re)) is performed for a plurality of Reynolds numbers (Re) before the Coriolis mass flow meter is put into operation and the determined compensation values (M.sub.f(Re)) are recorded in a memory, and that a control device of the Coriolis mass flow meter fetches the compensation values (M.sub.f(Re)) from the memory during operation of the Coriolis mass flow meter and correcting the measured value is performed based on these compensation values (M.sub.f(Re)).

1020. A Coriolis mass flow meter, comprising: a housing with an inlet and with an outlet for a fluid medium, at least one measuring tube configured to allow the fluid medium to flow through it and arranged between the inlet and the outlet, a vibration exciter configured to set the at least one measuring tube into vibration, two vibration sensors for detection of the movements of the measuring tubes, and a control device, wherein: the control device is configured for carrying out the method according to claim 11.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The invention will be explained in more detail below by reference to the embodiment examples shown in the figures. In the schematic figures:

[0023] FIG. 1: is a side view of a Coriolis mass flow meter;

[0024] FIG. 2: shows the measuring tube inside the housing of the Coriolis mass flow meter according to FIG. 1;

[0025] FIG. 3: is a diagram with exemplary measured value deviations and the graph of the compensation function according to equation (1);

[0026] FIG. 4: is a graph of the compensation function according to equation (1) to explain various parameters;

[0027] FIG. 5: is a diagram illustrating the influence of different values of Re.sub.c on the graph of the compensation function according to equation (1);

[0028] FIG. 6: is a diagram illustrating the influence of different values of the slope of the graph of the compensation function according to equation (1) at the point Re.sub.c on the graph;

[0029] FIG. 7: is a diagram illustrating the influence of different values of f.sub.max on the graph of the compensation function according to equation (1);

[0030] FIG. 8: is a diagram showing the corrected measured values of the example according to FIG. 3; and

[0031] FIG. 9: is a flow chart of the method.

DETAILED DESCRIPTION

[0032] Like parts, or parts acting in a like manner, are designated by like reference numerals. Recurring parts are not designated separately in each figure.

[0033] FIG. 1 shows a Coriolis mass flow meter 1 with a transmitter 2 and a housing 3. The transmitter 2 of the Coriolis mass flow meter 1 accommodates the electronics inter alia for the vibration exciter and the vibration sensors. It is connected to the housing 3 via a neck 34. In the present case, a control device 5 is configured to carry out the method according to the invention. For this purpose, it comprises in particular a data memory in which, for example, equation (1) and various values, for example for the parameters f.sub.max, Re.sub.c and

[00012] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}}$

determined during calibration, are stored. During operation, the Coriolis mass flow meter 1 with its housing 3 is fitted into a pipeline transporting the fluid to be measured. More particularly, the Coriolis mass flow meter 1 includes connectors 30, which in turn include an inlet 31 for connection to a supply line 40 and an outlet 32 for connection to a discharge line 41 of the pipeline.

[0034] Moreover, as shown in FIG. 2, the Coriolis mass flow meter 1 has a tube housing 33 in which two measuring tubes 4 are housed, one of which is not visible in the figure because it is located behind and covered by the other one. FIG. 2 also shows the progression of the measuring tubes 4 through the housing 3 from the inlet 31 via the tube housing 33 to the outlet 32. The progression of the measuring tubes 4, which is U-shaped in the example shown, also defines the flow direction of the fluid inside the measuring tube 4 and thus inside the Coriolis mass flow meter 1. The measuring tubes 4 are fixed by a respective fixing element 35 in both the region of the inlet 31 and the region of the outlet 32, said fixing element being configured as a gusset plate in the present example. As can also be seen from FIG. 2, a vibration exciter D is arranged on the measuring tube 4, which in operation of the Coriolis mass flow meter 1 is employed to set the measuring tubes 4 into vibration, in particular resonant vibration. In FIG. 2, the vibration excited by the vibration exciter D is directed into and out of the paper plane, respectively. A first vibration sensor S1 and a second vibration sensor S2 are arranged on the measuring tubes 4 in the flow direction upstream and downstream of the vibration exciter D, respectively. The vibration sensors S1, S2 detect the movements of the measuring tubes 4 and in particular the vibration induced by the vibration exciter D. Moreover, a temperature sensor RTD is arranged on the measuring tubes 4, which is, for example, configured as a resistance thermometer.

[0035] FIG. 3 shows a diagram illustrating the relationship between the deviation of the value measured by the Coriolis mass flow meter 1 and equation (1). The abscissa of the diagram shows the dimensionless Reynolds number Re, the ordinate shows the deviation of a measured value from the actual value in percent. The plotted points F show the deviation of values measured on a specific example of a Coriolis mass flow meter 1 for the flow rate or mass flow or the density of the fluid to be measured from the respective actual value. Since the measured values were obtained in a calibration or test setup, the actual values are known, so that the deviation of the individual values measured at the respective Reynolds number Re can be given as points F. In the present experiment, the Reynolds numbers Re for which measured values were obtained ranged approximately between 10.sup.2 and 10.sup.6. If the values measured by the Coriolis mass flow meter 1 were independent of the Reynolds number Re, the points F would all have to lie on the horizontal line at 0.0% deviation or scatter symmetrically around this line due to other influencing factors. However, as can be seen from the distribution of points F of the respective measured value deviations, this is not the case. More specifically, the measured values deviate more and more from the actual value with decreasing Reynolds numbers Re. As can also be seen from FIG. 3, the Coriolis mass flow meter 1 increasingly underestimates the actual values for small Reynolds numbers Re, i.e. the relative measured value deviation in percent takes on increasingly negative values as the Reynolds number Re approaches zero. It can also be seen that the measured value deviations for large Reynolds numbers Re become smaller and smaller, i.e. the measured values come closer and closer to the actual value and therefore scatter around the zero line as expected. As mentioned earlier, this is due to the fact that the Coriolis mass flow meter 1 is typically calibrated in operation with high Reynolds numbers Re, since these are easier to realize experimentally.

[0036] The function according to equation (1) can now be fitted to the distribution of the points F of the measured value deviations. How such a fit must be performed is known to the skilled person and therefore does not need to be explained in more detail here. From the fit, specific numerical values can be obtained for the parameters used in equation (1). For example, in the specific example shown in FIG. 3, the fit yields the values f.sub.max=−0.00951, Re.sub.c=842, and

[00013] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}} = 0.00334 .$

The graph G obtained with these values from equation (1) is also shown in FIG. 3. The graph G follows the distribution pattern of the points F of the measured value deviations and satisfies the requirements found in experimental observations: For a Reynolds number Re approaching zero, the graph G approaches a maximum (negative) measured value deviation f.sub.max (see also FIG. 4). For very high Reynolds numbers Re, on the other hand, the graph G approaches zero. Using equation (1), a function has thus been found that covers the measured value deviations over the entire range of Reynolds numbers Re from zero to infinity. It is therefore irrelevant in which range of Reynolds numbers Re the parameters of the function are determined. As long as the corresponding determination is sufficiently accurate, equation (1) may also be used to accurately determine the measured value deviation of the Coriolis mass flow meter 1 in other ranges of Reynolds numbers Re. For this purpose, it is not necessary to also perform experiments in the ranges of Reynolds numbers Re which are difficult to realize experimentally. This leads to a considerable limitation of the calibration effort.

[0037] FIG. 4 graphically illustrates various parameters of equation (1). The Reynolds number Re is plotted on the abscissa while the compensation values Mf(Re) calculated from equation (1) are plotted on the ordinate. As can be seen from graph G of the function according to equation (1), f.sub.max denotes the maximum measured value deviation and thus the maximum compensation value M.sub.f(Re) for Reynolds numbers Re approaching zero. The Reynolds number Re at which the compensation value M.sub.f(Re) assumes half of its maximum negative value f.sub.max is called Re.sub.c. At point Re.sub.c, graph G of the function according to equation (1) has its highest slope. In other words,

[00014] $\frac{d M_{f}}{d (\log_{1 0} R e)}$

reaches its maximum for Re=Re.sub.c.

[0038] For a better understanding of equation (1), FIGS. 5, 6 and 7 discuss the influence of parameters f.sub.max, Re.sub.c and

[00015] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}}$

on the graph of the function. FIG. 5, for example, shows the influence of Re.sub.c. The other two parameters remain constant, i.e. in the graphs 5a to 5e shown at f.sub.max=−0.03 and

[00016] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}} = 0. 1 .$

The value of Re.sub.c, on the other hand, was increased in powers of ten for graphs 5a to 5e. More specifically, for the graph according to 5a, Re.sub.c has the value 1, for 5b it has the value 10, for 5c it has the value 10.sup.2, for 5d it has the value 10.sup.3, and for 5e it has the value 10.sup.4. With the resulting shift of the graphs to the right on the abscissa, the regions of highest slope shift accordingly, while the limits f.sub.max for Reynolds numbers Re approaching zero and zero for Reynolds numbers Re approaching infinity remain the same. The value of the maximum slope also remains the same.

[0039] FIG. 6 shows the influence of different values for

[00017] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}},$

i.e. the slope of the graph at Re.sub.c. In contrast, the other two parameters are constant at f.sub.max=−0.03 and Re.sub.c=10.sup.3. Parameter

[00018] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}},$

on the other hand, has the value 0.002 for the graph according to 6a, 0.005 for 6b, 0.01 for 6c, 0.02 for 6d, and 0.05 for 6e. Higher slopes at Re.sub.c result in a faster approach of the compensation values Mf(Re) towards the respective limits f.sub.max for Reynolds numbers Re approaching zero and zero for high and infinite Reynolds numbers Re, respectively.

[0040] FIG. 7 shows the influence of different values for f.sub.max, i.e. the maximum compensation value M.sub.f(Re), which in this case is maximally negative. Again, the other two parameters are kept constant, more specifically at Re.sub.c=10.sup.3 and

[00019] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}} = 0. 1 .$

The parameter f.sub.max considered in FIG. 7, on the other hand, has the value −0.005 for the graph according to 7a, −0.01 for 7b, −0.02 for 7c, −0.03 for 7d, and −0.05 for 7e. Again, maximum values of the measured value deviation or the compensation values M.sub.f(Re) closer to zero result not only in the vertical compression of the graphs but also in a correspondingly steeper approximation towards the limits f.sub.max for Reynolds numbers Re approaching zero and zero for Reynolds numbers Re high or approaching infinity.

[0041] FIG. 8 illustrates the compensation effect of the function according to equation (1) on the particular measured example of FIG. 3. In particular, the measured values of the Coriolis mass flow meter 1 were corrected by the compensation values Mf(Re) obtained from equation (1) as described above. Figuratively, graph G of FIG. 3 is used as a new zero line for the points F and this has been plotted in a new diagram. Again, the abscissa of the diagram according to FIG. 8 shows the Reynolds number Re. The deviation of the measured values from the actual value is again plotted on the ordinate. As can be seen from the plot, after correcting the measured values according to equation (1), the measured points F of the measured value deviation now all scatter around the zero line. A dependence of the measured value deviation on the Reynolds number Re can no longer be detected and was therefore compensated. Moreover, it can also be seen that all experimentally determined measured value deviations, with the exception of three outliers, are within the limits of the maximum permissible error (MPE) usual for Coriolis mass flow meters 1, in the present case ±0.2%. In fact, all measured value deviations except for the outliers are even within a limit of ±0.1%. The compensation according to the invention therefore contributes significantly to the quality of the measurement results.

[0042] FIG. 9 illustrates a flow chart of the method 6 according to the invention. In the embodiment example shown, the method 6 starts with an experimental determination 60 of parameters f.sub.max Re.sub.c and

[00020] ${\frac{d M_{f}}{d (\log_{1 0} R e)} .Math.}_{R e = R e_{c}} .$

As explained above, only measured values in an experimentally easily accessible range of Reynolds numbers Re are required for this. The determined parameters are stored, for example, in a memory of the control device 5 of the Coriolis mass flow meter 1, so that the control device 5 can access them. A determination 61 of a compensation value Mf(Re) from equation (1) is then made, preferably during operation of the Coriolis mass flow meter 1, via the Reynolds number Re currently applicable for the fluid in the Coriolis mass flow meter 1. If necessary, the value of this current Reynolds number Re is also acquired 63 for this purpose, for example by receiving it from a central processing unit or an operator, or by calculating the current Reynolds number Re from parameters present in the Coriolis mass flow meter 1, made available to it or determined by it. Finally, correcting 62 the measured value of the Coriolis mass flow meter 1 is performed using the compensation value Mf(Re) determined from equation (1). Determining 61 of the compensation value Mf(Re) is preferably performed whenever the Reynolds number Re changes during operation of the Coriolis mass flow meter 1. Correcting 62, in turn, is performed basically continuously during operation of the Coriolis mass flow meter 1, so that each measured value recorded by the Coriolis mass flow meter 1 is corrected according to the invention. In this way, the influence of the flow rate and/or viscosity and/or density and/or Reynolds number Re on the measured values of the Coriolis mass flow meter 1 is compensated, thereby increasing the measurement accuracy, especially in the case of Reynolds numbers Re fluctuating during operation or deviating from the circumstances of the calibration. At the same time, according to the invention, it is sufficient to perform the calibration only in a range of Reynolds numbers Re that is easily accessible experimentally, thus reducing the cost of calibration.

METHOD FOR COMPENSATING THE INFLUENCE OF THE REYNOLDS NUMBER ON THE MEASUREMENT OF A CORIOLIS MASS FLOW METER, AND CORRESPONDING DEVICE

Inventors

Cpc classification

Classification Explorer

G01F1/84

PHYSICS

Classification Explorer

G01F15/02

PHYSICS

Classification Explorer

G01F1/8422

PHYSICS

Classification Explorer

G01F25/10

PHYSICS

Classification Explorer

G01F1/8436

PHYSICS

International classification

Classification Explorer

G01F1/84

PHYSICS

Abstract

Claims

Description