Mass flow meter having an oscillation sensor and method for eliminating noise signals from the measurement signal

Abstract

Mass flow meter having at least one measurement tube, through which mass flows, as an oscillation body which can be set in mechanical oscillation by means of an excitation unit, the oscillation behavior of which varying as a function of the mass flow can be recorded via at least one oscillation sensor in order to determine the mass flow, wherein in order to eliminate noise signals from the measurement voltage (sen) recorded via the oscillation sensor computational technology means are provided for forming a complex conjugate spectrum (|sa1j|) from the spectrum of the excitation voltage (seD) as well as a vector product between this (|sa1j|) and the measurement voltage (sen) for the purpose of filtering, in order, by further computational technology means for inverse Fourier transformation, to obtain the signal relationship associated with the vector product between the excitation voltage (seD) and the measurement voltage (sen) so that the processed measurement voltage (sal) resulting therefrom then predominantly comprises a useful signal component.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

Further exemplary measures which improve the invention will be presented in more detail below together with the description of exemplary embodiments.

FIG. 1 shows a schematic block diagram of an exemplary mass flow meter having signal processing means,

FIG. 2 shows a graphical representation of an exemplary signal spectrum in the initial situation,

FIG. 3 shows a graphical representation of the exemplary signal spectrum after a first processing step,

FIG. 4 shows a graphical representation of the exemplary signal spectrum after a further processing step, and

FIG. 5 shows a graphical representation of the exemplary signal spectrum in the final situation.

DETAILED DESCRIPTION

According to FIG. 1, an exemplary mass flow meter comprises a measurement tube 1, through which a fluid medium to be measured flows. The measurement tube 1 is mounted so that it can oscillate, and it is set by means of an excitation unit 2 in mechanical oscillations which are influenced by the flow of substance to be measured. The measurement signal characterizing this influence is recorded via an oscillation sensor 3.

In order to eliminate noise signals from the measurement voltage se_n recorded via the oscillation sensor 3, this voltage is first processed by computational technology means 4 for forming a complex conjugate spectrum |sa_1j| from the spectrum of the input excitation voltage se_D as well as a vector product between |sa_1j| and the measurement voltage se_n for the purpose of filtering. Subsequent computational technology means 5 are used for inverse Fourier transformation of the signal relationship associated with the vector product between the input excitation voltage se_D and the measurement voltage se_n. The measurement voltage |sa_1j| resulting therefrom and available as output for further signal processing then predominantly contains a useful signal component.

Desired measurement values can be obtained therefrom by further signal processing. For instance, the phase relation between the excitation voltage se_D and the filtered measurement voltage se_n is to be regarded in terms of signal processing technology as a measure of the noise signal-free flow value, which can accordingly be determined and output.

The computational technology means 4 and 5 for carrying out the signal processing are arranged inside an electronic computation unit 6 which, in this exemplary embodiment, is part of a personal computer having a microcontroller and memory units. The method, on which the signal processing is based, is configured as software and stored in the electronic computation unit 6.

In detail, the method on which the signal processing is based may be described as follows:

FIG. 2 shows an exemplary spectrum of the excitation voltage se_D, which is sinusoidal here, together with the measurement signal voltage se_n recorded by the oscillation sensor. Although the reception signal can still be seen from the signal waveform of the measurement signal voltage se_n, noise signals are however very significantly superimposed on the useful signal per se.

Referring to FIG. 3, said noise signals impact on the Fourier spectrum of the recorded measurement voltage se_n, e.g., in the low frequency range. The Fourier spectrum of the excitation voltage se_D has a signal with a frequency which corresponds to the excitation voltage, i.e. 100 Hz in this exemplary embodiment. The further signal processing for the purpose of noise signal elimination, or at least reduction, is based on initially forming a complex conjugate spectrum of the excitation voltage se_D.

According to FIG. 4, the vector product can be formed between the complex conjugate spectrum |sa_1j| and the measurement voltage se_n. This signal processing step is used for filtering, and reflects the relationship between the excitation voltage se_D and the measurement voltage se_n. In the spectrum after filtering, as can be seen, there is then only a similarity with the excitation voltage se_D.

If an inverse Fourier transformation is subsequently carried out according to FIG. 5, the processed measurement voltage sa₁ resulting therefrom then predominantly contains a useful signal component and is phase-shifted relative to the excitation voltage se_D. The flow value is revealed by the phase shift between the two signals; the density of the medium is revealed by the frequency.

The invention is not restricted to the exemplary embodiment described above. Variants thereof, which are covered by the protective scope of the appended claims, are furthermore conceivable. For instance, relationships with the flow rate and the density of the medium may be established noise-free by further signal processing steps. Furthermore, the excitation voltage se_D is not restricted to a sinusoidal signal waveform, since the nature of the excitation is of secondary importance in respect of the solution according to the invention. Although a unequivocal signal processing result can be obtained with a frequency of 100 Hz for the excitation voltage, other frequency values in the low frequency range are nevertheless also suitable.

It will be appreciated by those skilled in the art that the present invention can be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The presently disclosed embodiments are therefore considered in all respects to be illustrative and not restricted. The scope of the invention is indicated by the appended claims rather than the foregoing description and all changes that come within the meaning and range and equivalence thereof are intended to be embraced therein.

LIST OF REFERENCES

• 1 measurement tube
• 2 excitation unit
• 3 oscillation sensor
• 4 computational technology means
• 5 computational technology means
• 6 electronic computation unit
• se_D excitation voltage
• se_n recorded measurement voltage
• sa₁ processed measurement voltage
• |sa_1j| complex conjugate spectrum

Claims

1. A mass flow meter having at least one measurement tube, through which mass flows, as an oscillation body which can be set in mechanical oscillation by means of an excitation unit, the oscillation behavior of which varying as a function of the mass flow can be recorded via at least one oscillation sensor in order to determine the mass flow, wherein in order to eliminate noise signals from the measurement voltage (sen) recorded via the oscillation sensor computational technology means are provided for forming a complex conjugate spectrum (|sa1j|) from the spectrum of the excitation voltage (seD) as well as a vector product between this (|sa1j|) and the measurement voltage (sen) for the purpose of filtering, in order, by further computational technology means for inverse Fourier transformation, to obtain the signal relationship associated with the vector product between the excitation voltage (seD) and the measurement voltage (sen) so that the processed measurement voltage (sa1) resulting therefrom then predominantly comprises useful signal component.

2. The mass flow meter as claimed in claim 1, wherein the phase relation between the excitation voltage (seD) and the filtered measurement voltage (sen) is in terms of signal processing technology a measure of the flow value freed from the noise signal.

3. The mass flow meter as claimed in claim 1, wherein the frequency ratio between the excitation voltage (seD) and the filtered measurement voltage (sen) is a measure of the noise-free density of the medium to be measured.

4. The mass flow meter as claimed in claim 1, wherein computational technology means are provided for noise-free determination of the flow rate of the medium to be measured.

5. The mass flow meter as claimed in claim 1, wherein the excitation voltage (seD) driving the excitation unit for oscillatory excitation of the measurement tube is sinusoidal.

6. The mass flow meter as claimed in claim 5, wherein the excitation voltage (seD) driving the excitation unit has a frequency in the range of from 50 Hz to 200 Hz.

7. A method for eliminating noise signals from the measurement voltage (sen) of an oscillation sensor in flow meters, comprising the computation steps: forming the complex conjugate spectrum (|sa1j|) from the spectrum of the excitation voltage (seD),forming the vector product between the complex conjugate spectrum (|sa1j|) and the measurement voltage (sen) for the purpose of filtering,carrying out an inverse Fourier transformation in order to evaluate the signal relationship, associated with the vector product, between the excitation voltage (seD) and the measurement voltage (sen), so that the processed measurement voltage (sa1) resulting therefrom then comprises useful signal component.

8. An electronic computation unit for processing the measurement signal obtained via an oscillation sensor of a flow meter, having computational technology means for carrying out the method as claimed in claim 7.

9. The electronic computation unit as claimed in claim 8, wherein the computational technology means are designed in the form of at least one microcontroller having at least one assigned memory unit.

10. A computer program product for an electronic computation unit as claimed in claim 9, the method for eliminating the noise signals being implemented by corresponding control instructions contained in software.

11. The mass flow meter as claimed in claim 3, wherein computational technology means are provided for noise-free determination of the flow rate of the medium to be measured.

12. The mass flow meter as claimed in claim 4, wherein the excitation voltage (seD) driving the excitation unit for oscillatory excitation of the measurement tube is sinusoidal.

13. The mass flow meter as claimed in claim 6, wherein the excitation voltage (seD) driving the excitation unit has a frequency of about 100 Hz.

14. A computer program product for an electronic computation unit, which can be operated according to a method as claimed in claim 7, the method for eliminating the noise signals being implemented by corresponding control instructions contained in software for execution by at least one microcontroller having at least one assigned memory unit.

15. A mass flow meter comprising: at least one measurement tube, through which mass flows, as an oscillation body which can be set in mechanical oscillation by means of an excitation unit;an excitation unit which can be set in mechanical oscillation to the mass flow as an oscillation body;at least one oscillation sensor to record the oscillation behavior of the mass flow; andcomputational technology means provided to eliminate noise signals from a measurement voltage (sen) recorded via the oscillation sensor to form a complex conjugate spectrum (|sa1j|) from a spectrum of an excitation voltage (seD) as well as a vector product between this (|saij|) and the measurement voltage (sen) for the purpose of filtering to obtain the signal relationship associated with the vector product between the excitation voltage (seD) and the measurement voltage (sen) so that a processed measurement voltage (sa) results therefrom.