Method for operating a vortex flowmeter device

Abstract

Method for operating a vortex flowmeter device for measuring the flow of a fluid that flows through a measuring tube in which a baffle is arranged for producing eddies in the fluid. A signal-processing device processes signals of first and sensors produced by pressure fluctuations. A first signal is obtained by multiplication of the signal of the first sensor with a correction factor, and the second signal is obtained by multiplication of the signal of the second sensor with another correction factor such that a wanted signal is obtained from the deviation between the first signal and second signals, and a sum signal is formed from the sum of the first and second signals. A correlation between the wanted signal and the sum signal is determined and the correlation is minimized by variation of the correction factors, whereby same-phase interfering signals superimposed on anti-phase sensor signals are at least minimized.

Claims

1. Method for operating a vortex flowmeter device for measuring the flow of a fluid that flows through a measuring tube in which at least one baffle is arranged for producing eddies in the fluid using at least one first sensor and at least one second sensor for measuring pressure fluctuations in the fluid that accompany eddies formed by the at least one baffle, and using a signal-processing device for processing signals x.sub.1 of the at least one first sensor which include a first interfering signal and the signals x.sub.2 of the at least one second sensor which include an independent second interfering signal, whereby the signals x.sub.1 of the first sensor produced by the pressure fluctuations are in anti-phase opposition to the signals x.sub.2 of the second sensor produced by the pressure fluctuations, comprising the steps of: configuring and arranging the first and second sensors relative to the baffle in such a way that a polarity of a charge that results from a force along the y-axis on the first sensor is opposite to a polarity of a charge that results from a force on the second sensor independent of said first and second interfering signals, directing a flow of a fluid through the measuring tube of the vortex flowmeter device being operated and around the at least one baffle, obtaining a first signal y.sub.1 by multiplication of the signal x.sub.1 of the at least one first sensor with a correction factor v, and obtaining a second signal y.sub.2 by multiplication of the signal x.sub.2 of the at least one second sensor with a correction factor w, forming a difference between the first signal y.sub.1 and the second signal y.sub.2 as a wanted signal y.sub.d, representing the flow and forming a sum signal y.sub.s from a sum of the first signal y.sub.1 and the second signal y.sub.2, determining a correlation between the wanted signal y.sub.d and the sum signal y.sub.s, and minimizing the correlation by variation of the correction factors v and w, the minimum correlation indicating a minimum content of same-phase interfering signals in the wanted signal y.sub.d, whereby same-phase interfering signals superimposed on the anti-phase sensor signals are at least minimized and a measurement representing the flow of a fluid that flows through a measuring tube obtained, outputting a measurement of said flow through the measuring tube that reflects adjustment of the signals from said sensor resulting from said obtaining, forming, determining and minimizing steps.

2. Method according to claim 1, wherein one of the correction factors is 1 and the other of the correction factors is varied.

3. Method according to claim 1, wherein k is a correction factor from a closed interval [0; 1], and v=k and w=1−k.

4. Method according to claim 1, wherein a correlation is determined in time-discrete signal processing for an nth measurement by the correlation factor $\rho \left[n\right]=\frac{Y_{\mathrm{ds}}\left[n\right]}{\sqrt{Y_d\left[n\right]Y_s\left[n\right]}},\mathrm{with}$ $Y_{\mathrm{ds}}\left[n\right]=\underset{i=1}{\overset{n}{.Math.}}{y}_d\left[i\right]y_s\left[i\right],Y_d\left[n\right]=\underset{i=1}{\overset{n}{.Math.}}{y}_d^2\left[i\right]\mathrm{and}{Y}_s\left[n\right]=\underset{i=1}{.Math.}{y}_s^2\left[i\right].$

5. Method according to claim 1, wherein the minimum correlation is determined by a closed loop controller, and the closed loop controller comprises a difference calculator, a controller a signal calculator, and a correlation calculator, wherein the difference calculator forms an error from a non-correlation specified as a target correlation and an actual correlation wherein the error is a reference value of the controller, wherein at least one correction factor that is varied by the is a control value, wherein the signal calculator forms the wanted signal y.sub.d and the sum signal y.sub.s, and wherein the correlation calculator forms the actual correlation between the wanted signal y.sub.d and the sum signal y.sub.s.

6. Method according to claim 5, wherein the controller comprises a proportional-integral regulator.

7. Method according to claim 1, wherein the first and second sensors have different sensitivities which are compensated for by said multiplication of the signal x.sub.1 of the first sensor with the correction factor v and by multiplication of the signal x.sub.2 of the second sensor with the correction factor w.

8. Vortex flowmeter device for measuring the flow of a fluid that flows through a measuring tube having at least one baffle arranged in the measuring tube for producing eddies in the fluid, comprising: at least one first sensor for measuring the pressure fluctuations in the fluid that accompany the eddies and producing measurement signals x.sub.1, at least one second sensor for measuring the pressure fluctuations in the fluid that accompany the eddies and producing measurement signals x.sub.2 in anti-phase to the signals x.sub.1, and a signal-processing device for processing signals x.sub.1 of the at least one first sensor which include a first interfering signal and signals x.sub.2 of the at least one second sensor which include a second interfering signal, wherein the first and second sensors are configured and arranged relative to the baffle in such a way that a polarity of a charge that results from a force along the y-axis on the first sensor is opposite to a polarity of a charge that results from a force on the second sensor independent of said first and second interfering signals, wherein the signal-processing device is adapted for producing a wanted signal y.sub.d representative of the flow from a difference between a first signal y.sub.1 derived from the signal x.sub.1 and a second signal y.sub.2 derived from the signal x.sub.2, and forming a sum signal y.sub.s from a sum of the first signal y.sub.1 and the second signal y.sub.2, whereby same-phase interfering signals superimposed on anti-phase sensor signals are eliminated, wherein the signal processing system is set up in such a way that: a first signal y.sub.1 is obtained by multiplication of the signal x.sub.1 of the at least one first sensor with a correction factor v, and obtaining a second signal y.sub.2 by multiplication of the signal x.sub.2 of the at least one second sensor with a correction factor w, forming a difference between the first signal y.sub.1 and the second signal y.sub.2 as a wanted signal y.sub.d, representing the flow and forming a sum signal y.sub.s from a sum of the first signal y.sub.1 and the second signal y.sub.2, a correlation between the wanted signal y.sub.d and the sum signal y.sub.s, is determined and the correlation is minimized by variation of the correction factors v and w, the minimum correlation indicating a minimum content of same-phase interfering signals in the wanted signal y.sub.d, and means for at least minimizing same-phase interfering signals superimposed on the anti-phase sensor signals so as to obtain a corrected measurement representing the flow of a fluid that flows through a measuring tube.

9. Vortex flowmeter device according to claim 8, further comprising an A/D convertor for converting the signal of each sensor individually from analog to digital.

10. Vortex flowmeter device according to claim 8, wherein the first and second sensors have different sensitivities.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is diagrammatic sectional view of a baffle that is known from the state of the art with first and second piezoelectric sensors,

(2) FIG. 2 is a symbolic depiction of the determination of a wanted signal from voltage signals caused by a mechanical excitation of first and second sensors of the same sensitivity, by the signal processing of FIG. 7,

(3) FIG. 3 is a symbolic depiction of the determination of a wanted signal from the voltage signals caused by the mechanical excitation of first and second sensors of different sensitivities, by the signal processing of FIG. 7,

(4) FIG. 4 is a symbolic depiction of the basic idea according to the invention for implementing compensation of the effect of different sensitivities of the first and second sensors,

(5) FIG. 5 shows an embodiment, depicted as a block diagram, of the control circuit according to the invention having the functionality shown in FIG. 6,

(6) FIG. 6 shows an embodiment, depicted as a block diagram, of the determination according to the invention of a wanted signal and a sum signal from digitized voltage signals,

(7) FIG. 7 shows an embodiment, depicted as a block diagram, of a signal processing that is known from the state of the art, and

(8) FIG. 8 shows an embodiment, depicted as a block diagram, of the signal processing according to the invention, which comprises the functionality shown in FIG. 5.

DETAILED DESCRIPTION OF THE INVENTION

(9) FIG. 1 shows an overview of a baffle 1 that is known from the state of the art arranged in a measuring tube so that a fluid flows around the baffle 1 in a positive x-direction. In the tapered area of the baffle 1, eddies in the fluid that are caused by the baffle 1 and are generated by it produce pressure fluctuations, which exert forces on the baffle 1 along the y-axis in the tapered area of the baffle 1. These forces result in deviations or deformations of the baffle 1. In the area of the deformations, a first piezoelectric sensor 2a and a second piezoelectric sensor 2b are connected to the baffle 1; consequently, the piezoelectric sensors 2 are exposed to mechanical voltages. The mechanical voltages change the polarization of the sensors 2 by which, with reference to FIG. 7, an electrical charge q.sub.1 is produced on the first sensor 2a as a signal and an electrical charge q.sub.2 is produced on the second sensor 2b as a signal. The charge amount is a measure of the acting force. The sensors 2a, 2b are configured and arranged on the baffle 1 in such a way that the polarity of the charge q.sub.1 that results from a force along the y-axis on the first sensor 2a is opposite to the polarity of the charge q.sub.2 that results from the force on the second sensor 2b, whereby the polarities of the charges q.sub.1 and q.sub.2 are the same at forces that act along the x- or y-axis. In an alternative embodiment, the sensors are not connected to the baffle, but rather to a paddle that is arranged behind the baffle in the direction of flow.

(10) FIG. 7 shows a signal processing device 3 that is known from the state of the art. The charges q.sub.1 and q.sub.2 that are produced on the piezoelectric sensors 2 are converted by charge amplifiers 4 into signal voltages to and u.sub.2 that are proportional to the charges q.sub.1, q.sub.2. The signal voltages u.sub.1 and u.sub.2 are subtracted by a subtractor 5 (FIGS. 2 & 3), and the resulting differential voltage u.sub.d=u.sub.1−u.sub.2 is a wanted signal voltage, which is a measure of the flow. Before the digitization of the wanted signal voltage u.sub.d in an analog-digital converter 6, the wanted signal voltage u.sub.d is conditioned. On the one hand, the wanted signal voltage u.sub.d is filtered with a low-pass filter 7 to avoid alias effects, and, on the other hand, the zero-point voltage of the wanted signal voltage u.sub.d is set in a preloading device 8, so that the modulation range of the analog-digital converter 6 is exploited as much as possible. The analog-digital converter 6 is a component of a microcontroller 9, in which the further processing of the digitized voltage signal u.sub.d, which is x.sub.d, is carried out.

(11) If the first sensor 2a and the second sensor 2b have equally high sensitivities, the mechanical excitation of the baffle 1 that is caused by the eddy produces charges q.sub.1=q and q.sub.2=−q that are equally high in terms of value on the piezoelectric sensors with opposite polarities. The charges q.sub.1, q.sub.2 are converted from the charger amplifiers 4 into the voltages u.sub.1, u.sub.2, which are the same both in terms of antiphase and value. An additional mechanical excitation in the z-direction, produced, for example, by vibrations, produces a superposition of the signal voltages u.sub.1, u.sub.2 with same-phase interfering signals, whereby the values of the interfering signals in the two sensors 2 are equally large. FIG. 2 shows a corresponding example. By forming the wanted signal voltage u.sub.d by subtraction of the signal voltages u.sub.1 and u.sub.2 from one another, the same-phase interfering signals of the same value are completely eliminated. If the values of the same-phase interfering signals are different, the same-phase interfering signals are at least reduced.

(12) Actually, the first sensor 2a and the second sensor 2b, however, have different sensitivities. Possible causes lie in the piezoelectric materials of the sensors 2 themselves or are produced by unavoidable low tolerances in the arrangement of the sensors 2 on the baffle 1. FIG. 3 shows the signal voltages u.sub.1, u.sub.2 in the same mechanical excitation of the baffle 1 as in the case described based on FIG. 2; here, only the sensitivity of the second sensor 2b is lower than the sensitivity of the first sensor 2a. By the different sensitivities of the sensors 2, the same-phase interfering signals are not completely eliminated and reduce the quality of the wanted signal voltage u.sub.d that indicates the flow. Both trimming of the two sensors 2 themselves, so that the sensitivity of the sensors 2 is equal, and a calibration of the vortex flowmeter device are associated with high effort and accompanying high costs and are therefore impractical.

(13) FIG. 4 shows the basic idea according to the invention for implementing the compensation or at least the reduction of the detrimental effect of different sensitivities of the sensors 2 on the wanted signal voltage u.sub.d and thus on the measured flow, in particular, for same-phase interfering signals. The idea is to multiply one of the two signal voltages u.sub.1, u.sub.2 (u.sub.2 in FIG. 4), with a correction factor w and to select the correction factor w in such a way that the detrimental effect of the different sensitivities on the wanted signal voltage u.sub.d is minimum. In the depicted embodiment, the sensitivity of the second sensor 2b is less than the sensitivity of the first sensor 2a, and thus, the signal voltage u.sub.2 of the second sensor 2b is amplified with a factor w that is greater than 1. If the sensitivity of the second sensor 2b was to be higher than the sensitivity of the first sensor 2a, the factor w would be less than 1.

(14) Of course, it is also possible, in addition, to amplify (v>1) or to damp (v<1) the signal voltage u.sub.1 of the first sensor 2a with a correction factor v.

(15) The method according to the invention for finding the optimum correction factor w is based on the surprising property that the detrimental effect of different sensitivities on the wanted signal voltage u.sub.d=u.sub.1−wu.sub.2 is then minimum, even if the correlation between the wanted signal voltage u.sub.d and a sum signal voltage u.sub.s=u.sub.1+wu.sub.2 is minimum. In FIG. 6, a block diagram illustrates the method according to the invention for determining a wanted signal y.sub.d and a sum signal y.sub.s. The signals x.sub.1, x.sub.2 are the digitized signal voltages u.sub.1, u.sub.2. By multiplication of x.sub.1 with a correction factor v=k, a first signal y.sub.1 is produced and by multiplication of x.sub.2 with the correction factor w=(1−k), a second signal y.sub.2 is produced. The wanted signal is y.sub.d=y.sub.2−y.sub.1=(1−k)x.sub.2−kx.sub.1 and the sum signal is y.sub.s=y.sub.2+y.sub.1=(1−k)x.sub.2+kx.sub.1. The advantage of using correction factors k and (1−k) is that, as a range for k, the closed interval [0, 1] is adequate.

(16) FIG. 5 shows a closed loop controller 10 as an embodiment of the implementation of the method for finding the optimum correction factor k. The closed loop controller 10 comprises a difference calculator 11, a proportional-integral (PI) controller 12, a signal calculator 13, and a correlation calculator 14. The signal calculator 13 calculates the wanted signal y.sub.d and the sum signal y.sub.s, corresponding to the method that is shown in FIG. 6, from the signals x.sub.1, x.sub.2. The correlation calculator 14 calculates the actual correlation factor ρ.sub.actual between the wanted signal y.sub.d and the sum signal y.sub.s. The target correlation factor ρ.sub.target is zero, i.e., there is no correlation, and the deviation of the actual correlation factor ρ.sub.actual from the target correlation factor ρ.sub.target is the system deviation Δρ=ρ.sub.target−ρ.sub.actual. The actual correlation factor is calculated according to

(17) ${\rho }_{\mathrm{ist}}\left[n\right]=\frac{Y_{\mathrm{ds}}\left[n\right]}{\sqrt{Y_d\left[n\right]Y_s\left[n\right]}}$$\mathrm{with}$$Y_{\mathrm{ds}}\left[n\right]=\left(1-c\right)Y_{\mathrm{ds}}\left[n-1\right]+{\mathrm{cy}}_d\left[n\right]y_s\left[n\right],Y_d\left[n\right]=\left(1-c\right)Y_d\left[n-1\right]+{\mathrm{cy}}_d^2\left[n\right],\mathrm{and}$$Y_s\left[n\right]=\left(1-c\right)Y_s\left[n-1\right]+{\mathrm{cy}}_s^2\left[n\right],$
where c is a time constant. The system deviation Δρ is the initial value of the PI controller, which varies the correction factor k. The regulating process is terminated when the correlation between wanted signal y.sub.d and sum signal y.sub.s is reduced to a minimum.

(18) FIG. 8 shows a block diagram of a signal processing device 3 according to the invention. The charges q.sub.1, q.sub.2 that are generated by the piezoelectric sensors 2 are converted by the charge amplifier 4 into voltages u.sub.1, u.sub.2 that are proportional to the charges q.sub.1, q.sub.2. Before the digitization of the signal voltages u.sub.1, u.sub.2 in the analog-digital converter 6 with two signal voltage inputs, the signal voltages u.sub.1, u.sub.2 are first conditioned. This includes, on the one hand, the filtering of the signal voltages u.sub.1, u.sub.2 in the low-pass filters 7 to avoid alias effects, and on the other hand, the setting of zero-point voltages of sensors for the best possible use of the modulation range of the analog-digital converter 6 in the preloading devices 8. In contrast to the signal processing device 3 that is known from the state of the art and is shown in FIG. 7, a linkage of the signals of the two sensors is carried out only after the analog-digital conversion in the signal processing 3 that is shown in FIG. 8. In this way, the signals can be linked with one another as desired and further processed. In the microcontroller 9, both the calculation of the wanted signal y.sub.d and the sum signal y.sub.s, as explained in FIG. 6, as well as the control circuit 10 are implemented.