JP4126837B2 - Vortex flow measurement method and vortex flow meter - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a vortex flow measuring method and a vortex flow meter, and more particularly, to a vortex flow measuring method and a vortex flow meter improved by a method for measuring a Karman vortex generated in a fluid.
[0002]
[Prior art]
As shown in FIG. 10, the vortex flowmeter in the prior art includes a pair of first and second piezoelectric elements 10, 11, a first charge converter 13 connected to the first piezoelectric element 10, The second charge converter 12 connected to the two piezoelectric elements 11, the signal that has passed through the noise balance 14 that reduces the noise of the signal from the second charge converter 12, and the signal from the first charge converter 13. An adder 15 that adds the signals, an active filter 16 that passes the signal obtained by the adder 15 only through a predetermined frequency band, and a Schmitt trigger circuit 18 that converts the signal that has passed through the active filter 16 into a pulse signal. A noise discriminating circuit 19 for discriminating noise contained in the signal from the adder 15, a V-F circuit 20, and a Schmitt trigger circuit 1 The microprocessor 21 that receives the pulse signal from the VF circuit 20 and the signal from the V-F circuit 20 for signal processing, the display 22 that displays the signal from the microprocessor 21, and the pulse signal from the microprocessor 21 are insulated. A transformer 23 for transmission, an F-I circuit 27 for converting a signal insulated and transmitted by the transformer 23 into an analog signal, a changeover switch 26 for switching a signal from the microprocessor 21, and a communication signal superimposed on the analog signal. Transmission / reception modems 24 and 25, a changeover switch 28 for switching between an analog signal and a pulse signal, and a pulse output circuit 29.
[0003]
The vortex flowmeter having such a configuration converts the AC charge signals output from the first and second piezoelectric elements 10 and 11 into AC voltage signals by the first and second charge converters 12 and 13, respectively. The adder 15 performs addition amplification. Thereafter, the noise component is removed from the addition-amplified signal via the active filter 16 and converted into a pulse signal corresponding to the vortex frequency by the Schmitt trigger circuit 18. The active filter 16 is selected in the optimum frequency band by the microprocessor 21 from the diameter, fluid density and maximum flow rate through which the fluid flows. This pulse signal is taken into the microprocessor 21, the frequency is measured by an internal counter, the flow rate calculation and the correction calculation are performed, and then the pulse signal corresponding to the flow rate signal is output. This pulse signal passes through the transformer 23 and the FI circuit 27 and is then converted into a 4-20 mA current signal.
[0004]
[Problems to be solved by the invention]
However, in the signal processing by the active filter of the vortex flow meter described in the prior art, the noise component superimposed on the vortex signal component is (1) pipe vibration, (2) beat-like noise, (3) low frequency noise, (4) High frequency noise such as vortex generator resonance, (5) Spike-like noise, etc. are mixed. These noises can be considerably reduced by the active filter 16, but the noise component that remains without being reduced still adversely affects the signal component that has passed through the active filter 16. Specifically, the Schmitt trigger circuit 18 causes a double count or causes a pulse drop. As a result, there is a problem that a stable flow rate cannot be obtained and an error occurs.
[0005]
Therefore, there is a problem to be solved by a method and a vortex flowmeter for removing and attenuating a noise component superimposed on a signal component obtained from a Karman vortex generated in a fluid.
[0006]
[Means for Solving the Problems]
In order to solve the above problems, the vortex flow measuring method and the vortex flow meter according to the present invention are configured as follows.
[0007]
(1) In a vortex flowmeter that measures the flow rate of a fluid by detecting an alternating stress signal generated by a Karman vortex, the alternating stress signal is a low-pass filter having a frequency band determined by a diameter and a flow range through which the fluid flows. The signal intensity of the noise component measured in advance from the signal intensity of each signal that has passed through the band division filter and passed through the band division filter that has divided the signal that has passed through the low pass filter into a plurality of frequency bands An eddy flow rate measurement method characterized in that a vortex flow rate signal is obtained by taking the difference between the two.
(2) In the above (1), the band dividing filter starts with a frequency band through which the alternating stress signal passes., ThinningFormed with a filter consisting of multiple frequency bandsIsA method of measuring vortex flow rate.
(3) Above (2), The signal intensity of the noise component is zeroStateA method for measuring a vortex flow rate, wherein the signal intensity is a signal intensity of a signal that has passed through the band-splitting filter.
(4) In the above claim 1,A second-order low-pass filter having an input connected to the output of the low-pass filter, an output connected to the input of the band-splitting filter, and uniformizing the amplitude of the signal regardless of the frequencyA method for measuring vortex flow rate.
(5) Above (4)The secondary low-pass filter is a digital filter, and makes the amplitude of the vortex signal proportional to the square of the fluid density and the flow velocity of the fluid uniform regardless of the frequency.A method for measuring vortex flow rate.
[0008]
(6) In a vortex flowmeter that measures the flow rate of a fluid by detecting an alternating stress signal generated by a Karman vortex, the alternating stress signal is a low-pass filter having a frequency band determined by a diameter and a flow range through which the fluid flows. The signal that has passed through the low-pass filter is passed through a band-splitting filter that has been divided into a plurality of frequency bands that have been switched to a gain based on the signal intensity of a noise component that has been measured in advance, and has passed through the band-splitting filter. A vortex flow rate measuring method characterized in that a vortex flow rate signal is obtained from the signal intensity of each signal.
(7) In the above (6), the band division filter includes a frequency band filter that allows the alternating stress signal to pass through., ThinningFormed by a filter consisting ofRuA method for measuring vortex flow rate.
(8) Above (7), The signal intensity of the noise component is zeroStateA method for measuring a vortex flow rate, wherein the signal intensity is a signal intensity of a signal that has passed through the band-splitting filter.
(9) In (6) above,A second-order low-pass filter having an input connected to the output of the low-pass filter, an output connected to the input of the band-splitting filter, and uniformizing the amplitude of the signal regardless of the frequencyA method for measuring vortex flow rate.
(10) (9)The secondary low-pass filter is a digital filter, and makes the amplitude of the vortex signal proportional to the square of the fluid density and the flow velocity of the fluid uniform regardless of the frequency.A method for measuring vortex flow rate.
[0009]
(11) In a vortex flowmeter that measures the flow rate of a fluid by detecting an alternating stress signal generated by a Karman vortex, a low pass that allows the alternating stress signal to pass through a frequency band determined by a diameter and a flow range through which the fluid flows. A filter, an A / D converter that converts a signal that has passed through the low-pass filter into a digital signal, and the signal converted by the A / D converterDigital signalA vortex flow rate signal by taking a difference between the signal intensity of a noise component measured in advance from the signal intensity of each signal that has passed through the band division filter and a filter band division filter that passes a filter composed of a plurality of frequency bands And signal calculation means to obtainCharacterized by comprisingVortex flow meter.
(12) In the above (11), the band dividing filter includes a filter in a frequency band that allows the alternating stress signal to pass., ThinningFormed by a filter consisting ofRuVortex flowmeter characterized by that.
(13) (12), The signal intensity of the noise component is zeroStateThe vortex flowmeter according to claim 1, wherein the vortex flowmeter has a signal intensity of a signal that has passed through the band dividing filter.
(14) In the above (11),An input is connected to the output of the low-pass filter, an output is connected to the input of the band-splitting filter, and a second-order low-pass filter that equalizes the amplitude of the signal regardless of the frequency is provided. It is a filter, and the amplitude of the vortex signal proportional to the square of the fluid density and the flow velocity of the fluid is made uniform regardless of the frequency.Vortex flowmeter characterized by that.
[0010]
(15) In a vortex flowmeter that measures the flow rate of a fluid by detecting an alternating stress signal generated by a Karman vortex, a low-pass filter that passes the alternating stress signal in a frequency band determined by a diameter and a flow range through which the fluid flows; An A / D converter that converts the signal that has passed through the low-pass filter into a digital signal, and the signal converted by the A / D converterDigital signalAre switched based on the signal strength of the noise component measured in advance and obtained by the switched gain. Signal calculation means for obtaining a vortex flow rate signal from the obtained signal intensity;Characterized by comprisingVortex flow meter.
(16) In the above (15), the band division filter includes a frequency band filter that allows the alternating stress signal to pass through., ThinningFormed by a filter consisting ofRuVortex flowmeter characterized by that.
(17) Above (16), The signal intensity of the noise component is zeroStateThe vortex flowmeter according to claim 1, wherein the vortex flowmeter has a signal intensity of a signal that has passed through the band dividing filter.
(18) In the above (15),An input is connected to the output of the low-pass filter, an output is connected to the input of the band-splitting filter, and a second-order low-pass filter that equalizes the amplitude of the signal regardless of the frequency is provided. It is a filter, and the amplitude of the vortex signal proportional to the square of the fluid density and the flow velocity of the fluid is made uniform regardless of the frequency.Vortex flowmeter characterized by that.
[0011]
Thus, by subtracting the signal strength of the noise component from the signal strength of the signal generated by the band division filter or controlling the gain based on the signal strength of the noise component, the vortex flow signal is generated accurately. In addition, a stable signal component can be obtained.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, various embodiments of a vortex flow rate measuring method and a vortex flow meter according to the present invention will be described with reference to the drawings.
[0013]
As shown in FIG. 1, a vortex flow meter embodying a vortex flow measuring method according to a first embodiment of the present invention includes a piezoelectric element 30 for detecting Karman vortices generated in a fluid, and the piezoelectric element 30. The charge converter 31 that converts the alternating current charge signal (alternating stress signal) obtained in step 1 into an alternating voltage signal, the anti-aliasing filter (LPF) 32 that passes only the eddy frequency component, and the anti-aliasing filter (LPF) 32 A / D converter 33 that converts the vortex signal of the analog signal into a digital signal, a low-pass filter (LPF0) 34 that removes frequencies above the maximum vortex frequency obtained by the aperture, fluid density, and maximum flow rate, A second-order low-pass filter (LPF) 35a, which is a digital filter that uniforms the amplitude, and a vortex signal that is converted into a uniform amplitude signal A microprocessor 40 having a band division filter 35 composed of a digital filter that performs subband decomposition, a signal calculation means for measuring and calculating the intensity of a signal that has passed through the band division filter 35, and storing noise component information and A memory ROM 41, RAM 42, EEPROM 43 for storing and reading signals including signals processed by the microprocessor 40, a D / A converter 44 for converting a digital signal obtained by the microprocessor 40 into an analog signal, an analog And an output stage 45 for outputting a signal.
[0014]
The low-pass filter (LPF0) 34 sets a so-called corner frequency, and is set to a frequency band that can pass through the maximum vortex frequency obtained by the diameter, fluid density, and maximum flow rate through which the fluid flows. This makes it possible to apparently remove background noise that is high-frequency component noise having a frequency component equal to or higher than the vortex frequency.
[0015]
The secondary low-pass filter (LPF0) 35a is a digital filter, and makes the amplitude of the vortex signal proportional to the square of the fluid density and the vortex frequency (flow velocity) uniform regardless of the frequency. Thereby, in the microprocessor 40, the signal comparison reference at the reference value TLA for comparing the vortex signals can be set based on the unified data.
[0016]
As shown in FIG. 2, the band-splitting filter 35 is composed of filters of frequency bands W1 to Wn. In the case of the embodiment, the band-splitting filter 35 is composed of filters composed of five frequency bands of frequency bands W1 to W5. Yes. First, a frequency band W1 that allows a signal that has passed through the anti-alien filter (LPF) 32 to pass through a signal that has passed through the first high-pass filter (HPF1) is further down-sampled by half (referred to as half-decimation); A frequency band W2 for passing the signal that has passed through the first low-pass filter (LPF1) 36b by half down-sampling and further passing the signal that has passed through the second high-pass filter (HPF2) 37a by down-sampling by half. The signal that has been passed through the second low-pass filter (LPF2) 37b is further down-sampled by one-half the third low-pass filter (LPF2) 37b. The signal that has passed through the high-pass filter (HPF3) 38a is further divided into two The frequency band W3 to be down-sampled and passed, and the signal that has passed through the second low-pass filter (LPF2) 37b by half down-sampling and further passed through the third low-pass filter (LPF3) 38b are further divided into two. The frequency band W4 that passes the signal that has passed through the fourth high-pass filter (HPF4) 39a down-sampled by one half and further down-sampled by one-half, and the signal that has passed through the third low-pass filter (LPF3) 38b. The frequency band W5 is down-sampled by half and passed through the fourth low-pass filter (LPF4) 39b, and further down-sampled by half and passed. The signals that have passed through these frequency bands W1 to W5 are input to the microprocessor 40, the intensity measurement 40a is performed, the signal reference value (TLA) 40b is determined, and the signal with the maximum intensity signal is calculated according to the calculation method described later. Extracted as
[0017]
In the vortex flow measuring method and the vortex flow meter having such a configuration, first, an AC charge signal (alternating stress signal) detected from the piezoelectric element 30 is converted into an AC voltage signal by the charge converter 31, and an anti-aliasing filter ( The vortex signal of the vortex frequency component that has passed through the LPF) 32 is converted into a digital signal by the A / D converter 33. Note that the present invention can also be applied to a vortex flowmeter using two piezoelectric elements as in the prior art. In that case, the same processing may be applied to the adder output signal. As shown in FIG. 3, the vortex signal converted into the digital signal has a uniform amplitude after passing through the second-order low-pass filter (LPF) 35a. Here, as shown in FIG. 4, since the gain is set smaller as the frequency band is higher, the gain of the signal component in the lower direction is higher and the gain of the signal component in the higher frequency band is lower. The digital signal composed of the signal components having the uniform characteristics (frequency and amplitude) captured in this way is extracted by the frequency division band which is captured and divided by the band division filter 35 constituted by the digital filter. Subband decomposition is performed. Hereinafter, the operation of subband decomposition by the band division filter will be described.
[0018]
When the vortex signal having a uniform amplitude that has passed through the second-order low-pass filter (LPF) 35a is allowed to pass through the high-pass filter and the low-pass filter, the subband decomposition function has the respective frequency bands (W1 to W5 in the embodiment). Limited by range. Since the output signal that has passed through the high-pass filter and the low-pass filter is down-sampled by one-half (1/2 decimation; thinning-out), the number of signal samples is kept constant.
[0019]
By repeating this operation N times, the frequency band can be divided into (N + 1). The signal component and the noise component are classified in each of the frequency bands W1 to Wn divided into (N + 1). The example of FIG. 3 is a case where n = 4 in accordance with the embodiment shown in FIG. 2, and is divided into five frequency bands W1 to W5. The frequency band W1 is a frequency band of the first high-pass filter (HPF1) 36a that allows the signal from which background noise has been removed to pass, and the frequency band W2 passes the signal that has passed through the first low-pass filter (LPF1) 36b. The frequency band of the second high-pass filter (HPF2) 37a down-sampled by half is the frequency band W3 is the third high-pass that down-samples the signal that has passed through the second low-pass filter (LPF2) 37b by half. The frequency band of the filter (HPF3) 38a, and the frequency band W4 is the frequency band of the fourth high-pass filter (HPF4) 39a obtained by down-sampling the signal that has passed through the third low-pass filter (LPF3) 38b by half. The frequency band W5 is a third low-pass filter (LP 3) a fourth low-pass filter a signal that has passed through 38b and one-half downsampling (LPF 3) is a frequency band of 39 b.
[0020]
  FIG. 4 shows the overall filter characteristics after combining the one shown in FIG. 3 with a secondary low-pass filter..Pass the band-splitting filterByClassify signal components and noise components in each frequency band W1 to W5.
[0021]
The maximum signal intensity obtained from each of the band division filters divided into such frequency bands W1 to Wn corresponds to the vortex frequency component. The calculation of the strongest signal of the eddy frequency component is calculated in the microprocessor 40, and each band division filter (first to fourth high-pass filters 36a to 39a and first to fourth low-pass filters 36b to 36b) is calculated. 39b) is calculated by a value obtained by subtracting the signal with the maximum signal intensity passed through 39b) and the signal intensity of noise at the time of zero flow or self-diagnosis, and outputs as a vortex flow signal. Hereinafter, a method for calculating the vortex flow rate signal will be described with reference to FIG.
[0022]
FIG. 5 (A) shows signal components in each frequency band W1 to W5 with respect to a vortex signal having a vortex frequency component that is passed through an anti-aliasing filter (LPF) 32 at zero flow rate and self-diagnosis to remove background noise. Since the signal strength in this case is a measurement at a flow rate of zero, the signal component should not be measured, so the signal component measured here is the noise component. The signal intensity An (i) is obtained. For example, the signal strength An (1) of the noise in the frequency band W1 is the largest, the signal strengths An (2) and An (4) in the frequency bands W2 and W4 are the smallest, and the signal strength An in the frequency bands W3 and W5. (3), An (5) is the signal strength at the intermediate position. These noise signal intensities An (i) are accumulated in the memory EEPROM and used for calculation of signal components and calculation of the reference value TLA in actual flow measurement.
[0023]
FIG. 5B shows the signal strength As (i) of the signal component detected in each frequency band at the time of the flow rate measurement. The signal strength As (3) in the frequency band W3 is the largest, The signal strengths As (1), As (5), As (2), As (4) in the order of the frequency bands W1, W5, W2, and W4.
[0024]
In FIG. 5C, the signal strength An (i) (see FIG. 5A) of the noise component described above is subtracted from the measured signal component signal strength (see FIG. 5B) and set in advance. Reference signal TLA, and the signal strength As (1) of the signal component in the frequency band W1 is large, but the signal strength As (1) of the noise is also large, so the signal strength As ( 1) 'has a small signal strength. The signal strength As ′, which is obtained in the same manner for the other frequency bands W2 to W5 and as a result is equal to or greater than the reference value TLA, becomes the signal strength As (3) ′ of the frequency band W3. It becomes a flow signal.
[0025]
Here, with respect to spike-like noise, there is a moving average or the like as a method for measuring the signal strength of each band division filter for each calculation cycle. According to this method, it can be determined whether the noise is stationary or sudden. The moving average is a so-called moving average filter, which is composed of an adder and a delay element such as a register, and normally decimates each time it passes through a low-pass filter.
[0026]
In this way, the signal component and the noise component are separated, and the background noise of each frequency band of the band dividing filter is measured in advance and stored in the memory RAM and EEPROM, and each divided band filter at the time of measuring the flow rate. By taking the difference from the signal intensity in each of the frequency bands W1 to Wn, it is possible to detect an accurate eddy flow signal that is not affected by the background noise with the flow meter attached.
[0027]
Next, the overall operation of the vortex flowmeter having the above function will be described with reference to the flowchart shown in FIG.
[0028]
First, when the power is turned on, various data and flags are initialized (step ST10). Specifically, the diagnosis execution flag is set, the flow rate presence / absence diagnosis flag is cleared, and the flow rate is zero. And, when the power is turned on for the first time, make sure to make a diagnosis. This is because the fluid does not normally flow before the power is turned on with the vortex flowmeter attached. Therefore, the state of these flags is not stored in the EEPROM.
[0029]
Next, the signal strength An (i) of the noise data is read from the EEPROM (step ST11). That is, the data measured in advance when measuring the background noise is read from the EEPROM.
[0030]
Next, a self-diagnosis is performed. The self-diagnosis will be described with reference to the flowchart shown in FIG. When the self-diagnosis starts, it is first determined whether the fluid is liquid, gas, or gas such as steam (step ST20).
[0031]
In the case of liquid, it is determined whether or not the detected signal band is in a low frequency region (step ST21). If a signal in the low frequency region is detected, it is determined that the flow is unstable (step ST22).
[0032]
If the fluid is gas or steam, it is determined whether or not the band of the detected signal is high (step ST23). If the band of the detection signal is a high frequency, it is determined that there is a possibility of valve resonance and header noise (step ST24). In step ST23, when the band of the detection signal is low, it is determined that there is vibration (step ST25).
[0033]
In this way, in self-diagnosis, various phenomena that appear to occur in the fluid, such as unstable flow, vibration, valve resonance, header noise, etc., occur depending on the frequency band of the detection signal. Thus, the self-diagnosis is completed in consideration of the subsequent flow rate measurement.
[0034]
Next, returning to FIG. 6, after the self-diagnosis is performed in step ST15, the diagnosis execution flag is cleared and the flow rate presence / absence determination flag is set (steps ST16 and ST17).
[0035]
In step ST13 and step ST17, when the flow rate presence / absence determination flag set and the flow rate presence / absence determination flag are set, each band obtained by measuring the background noise from the signal intensity As (i) obtained by the measurement. A value As (i) ′ obtained by subtracting the noise intensity An (i) of the divided filter is calculated to calculate a vortex flow rate signal, and the process returns to step ST12 to repeat this operation (steps ST18 and ST19).
[0036]
Next, a vortex flow rate measuring method and a vortex flow meter according to a second embodiment will be described with reference to the drawings.
[0037]
The vortex flowmeter embodying the vortex flow measurement method of the second embodiment has the same configuration as the vortex flowmeter of the first embodiment shown in FIGS. 1 and 2 and is different. The calculation method for calculating the vortex flow rate signal by the microprocessor is different.
[0038]
First, an AC charge signal (alternating stress signal) detected from the piezoelectric element 30 is converted into an AC voltage signal by the charge converter 31 and A / D conversion is performed on the vortex signal of the vortex frequency component that has passed through the anti-aliasing filter (LPF) 32. The digital signal is converted by the device 33. As shown in FIG. 8B, the vortex signal converted into the digital signal is converted into a vortex signal having a uniform amplitude after passing through the secondary low-pass filter (LPF) 35a. The digital signal composed of the signal components having the uniform characteristics (frequency and amplitude) captured in this way is extracted by the frequency division band which is captured and divided by the band division filter 35 constituted by the digital filter. Subband decomposition is performed.
[0039]
When the vortex signal having a uniform amplitude that has passed through the second-order low-pass filter (LPF) 35a is allowed to pass through the high-pass filter and the low-pass filter, the subband decomposition function has the respective frequency bands (W1 to W5 in the embodiment). Limited by range. Since the output signal that has passed through the high-pass filter and the low-pass filter is down-sampled by half (1/2 decimation; thinning), the number of signal samples is kept constant.
[0040]
The example of FIG. 8 shows a band division filter that is divided into five frequency bands W1 to W5. The frequency band W1 includes a first high-pass filter (HPF1) 36a that passes a signal from which background noise has been removed. The frequency band W2 is a frequency band of a second high-pass filter (HPF2) 37a down-sampled by half that passes the signal that has passed through the first low-pass filter (LPF1) 36b, and the frequency band W3 is A frequency band of a third high-pass filter (HPF3) 38a obtained by down-sampling the signal that has passed through the second low-pass filter (LPF2) 37b by half, and the frequency band W4 passes through the third low-pass filter (LPF3) 38b. A fourth high-pass filter (HP 4) 39a is a frequency band of the frequency band W5 is the frequency band of the fourth low-pass filter (LPF 3) 39 b which is one-half downsampling the signal passed through the third low-pass filter (LPF 3) 38b.
[0041]
The maximum signal intensity obtained from each of the band division filters divided into such frequency bands W1 to Wn (in the embodiment, n = 5 and frequency bands W1 to W5) is the eddy frequency component. It corresponds to. The calculation of the strongest signal of the eddy frequency component is performed by the microprocessor 40, and each band division filter (first to fourth high-pass filters 36a to 39a and first to fourth low-pass filters 36b to 39b) is calculated. ) And the respective signal strengths of the maximum signal strength passed through and the signal strength of noise at the time of zero flow rate or self-diagnosis, that is, by changing each gain of the band division filter according to the amplitude of the background noise, Output as a vortex flow signal. Specifically, in the band division filter, the signal strength is reduced to (1/2) by being decimated by half.ⁿTherefore, when comparing the signal strength in each band, the obtained signal strength is 2ⁿIt is necessary to double. Here, by changing the value of n in accordance with the noise signal intensity, the frequency characteristics shown in FIG. 8 to be described later can be realized. If the gain is lowered, the original signal may not be detected, but the signal component is larger than the square of the fluid density and the flow velocity in the signal band, so that the high frequency band Even if the gain is reduced to some extent, no problem occurs. That is, when the noise signal strength is “high” and the band filter gain is “low”, the value of n is small. When the noise signal strength is “low” and the band filter gain is “high”, the value of n is Become “big”. Hereinafter, a method for calculating the vortex flow rate signal will be described with reference to FIG.
[0042]
FIG. 8A shows the signal component in each frequency band W1 to W5 with respect to the vortex signal of the vortex frequency component that has been passed through the anti-aliasing filter (LPF) 32 at the time of zero flow rate and self-diagnosis to remove background noise. This shows the signal strength. In this case, since the signal strength is measured at a flow rate of zero, the signal component should not be measured. Therefore, the measured signal component is a noise component. Signal strength, that is, noise signal strength An (i). For example, the noise signal intensity An (1) in the frequency band W1 is the largest, the noise signal intensity An (2), An (4) in the frequency bands W2 and W4 is the smallest, and the noise signal intensity in the frequency bands W3 and W5. An (3) and An (5) are intensities at intermediate positions. These noise signal strengths An (1) to An (5) are stored in the memory EEPROM, and are used for changing the gain of each frequency band and calculating the reference value TLA in actual flow measurement.
[0043]
FIG. 8B shows the gain for the signal component detected in each frequency band at the time of flow rate measurement. Since the noise signal intensity An (1) in the frequency band W1 is the highest, the gain is the highest. Next, the frequency bands W3, W5, W4, and W2 increase in order.
[0044]
FIG. 8C shows the signal intensity As (i) measured by changing the gain of each filter of the band division filter based on the noise signal intensity An (i) described above and a preset reference value TLA. The signal component As (1) ′ of the signal component before passing through the filter in the frequency band W1 is large, but the noise signal strength An (1) (see FIG. 8A) is also large. The signal intensity As (1) of the signal component is a small signal intensity. The other frequency bands W2 to W5 are obtained in the same manner. As a result, the signal in the frequency band W3 has a small noise signal strength An (3) and a high signal strength As (3) ′ before passing through the filter. The signal intensity As (3) after passing through the filter becomes a large value and becomes the eddy flow signal measured this time.
[0045]
In this way, the signal component and the noise component are separated, and the background noise of each frequency band of the band dividing filter is measured in advance and stored in the memory RAM and EEPROM, and each divided band filter at the time of measuring the flow rate. By changing the gain of each frequency band W1 to Wn and measuring the signal intensity, it becomes possible to detect an accurate eddy flow signal that is not affected by the background noise.
[0046]
Next, the overall operation of the vortex flowmeter having the above function will be described with reference to the flowchart shown in FIG.
[0047]
First, when the power is turned on, various data and flags are initialized (step ST30). Specifically, the diagnosis execution flag is set, the flow rate presence / absence diagnosis flag is cleared, and the flow rate is zero. And, when the power is turned on for the first time, make sure to make a diagnosis. This is because the fluid does not normally flow before the power is turned on with the vortex flowmeter attached. Therefore, the state of these flags is not stored in the EEPROM.
[0048]
Next, the noise signal intensity An (i) of the noise data is read from the EEPROM (step ST31). That is, the data measured in advance during the background noise measurement is read from the EEPROM. If it is not set here, the ROM initial value and the factory default data are used.
[0049]
Next, when the diagnosis execution flag is set, self-diagnosis is performed (steps ST32 and ST34). As for self-diagnosis, self-diagnosis is performed based on the flowchart shown in FIG. 7 described in the first embodiment. If the detected signal is in the low frequency region, the flow is unstable. If a signal is detected in a low frequency band, vibration is detected, and if a signal is detected in a low frequency band, the presence of a noise source due to valve vibration or the like is specified. Since details have been described in the first embodiment, a description thereof will be omitted.
[0050]
Next, after the self-diagnosis is performed, the diagnosis execution flag is cleared and the flow rate presence / absence determination flag is set (steps ST36 and ST37).
[0051]
In step ST32, when the diagnosis execution flag is not set, it is determined whether the flow rate presence / absence determination flag is set or cleared (step ST33).
When the flow rate presence / absence determination flag is cleared, the flow rate is zero, and self-diagnosis (steps ST34 to ST37) is performed. When the flow rate presence / absence determination flag is set, the gain of the divided band filter is switched in accordance with the noise signal intensity An (i), and the flow rate signal is measured (steps ST38 and 39). Thereafter, the flow returns to step ST32 to repeatedly measure the flow rate.
[0052]
【The invention's effect】
As described above, the present invention divides the detected vortex signal into a frequency band to accurately separate the signal component and the noise component, and removes the signal of the noise component measured in advance at the time of measurement, thereby removing the signal component. As a result, it is possible to obtain a stable vortex flow rate signal.
[0053]
Further, at the time of measurement, it is possible to obtain a stable vortex flow rate signal by switching the gain of each frequency band of the band division filter based on a noise component measured in advance.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration of a vortex flowmeter of the present invention.
FIG. 2 is a block diagram showing a term system of the band division filter in FIG. 1;
FIG. 3 is a graph showing frequency characteristics in the band division filter in FIG. 2;
4 is a graph showing a frequency characteristic example in which a signal of a low-pass filter (LPF) is combined with the frequency characteristic of the band division filter in FIG.
FIG. 5 shows the signal intensity in each frequency band at the same flow rate zero or in self-diagnosis, FIG. 5 (A) is a graph showing the noise signal intensity, and FIG. 5 (B) is a flow measurement. FIG. 5C is a graph showing the signal intensity obtained by subtracting the noise signal intensity at the time of flow velocity measurement.
FIG. 6 is a flowchart showing an overall operation in the first embodiment.
FIG. 7 is a flowchart showing the self-diagnosis operation when the flow rate is zero.
FIG. 8 shows the signal strength in each frequency band when the flow rate is zero or during self-diagnosis according to the second embodiment, and FIG. 8 (A) is a graph showing the noise signal strength. FIG. 8B is a graph showing how the gain of each frequency band is switched based on the noise signal intensity, and FIG. 8C is a graph showing the signal intensity in the frequency band where the gain is switched. .
FIG. 9 is a flowchart showing an overall operation according to the second embodiment.
FIG. 10 is a block diagram showing a configuration of a vortex flowmeter in the prior art.
[Explanation of symbols]
30 Piezoelectric elements
31 Charge converter
32 Low-pass filter (LPF)
33 A / D converter
34 Low-pass filter
35 Band division filter
35a Secondary low-pass filter (LPF)
36a First high-pass filter (HPF1)
36b First low-pass filter (LPF1)
37a Second high-pass filter (HPF2)
37b Second low-pass filter (LPF2)
38a Third high-pass filter (HPF3)
38b Third low-pass filter (LPF3)
39a Fourth high-pass filter (HPF4)
39b Fourth low-pass filter (LPF4)
40 microprocessor
41 Memory (ROM)
42 Memory (RAM)
43 Memory (EEPROM)
44 D / A converter
45 Output stage

Claims (18)

In the vortex flowmeter that measures the flow rate of fluid by detecting the alternating stress signal generated by Karman vortex,
The alternating stress signal is passed through a low-pass filter having a frequency band determined by the diameter and flow range through which the fluid flows,
Pass the signal that has passed through the low-pass filter through a band dividing filter that is divided into a plurality of frequency bands,
A vortex flow rate measurement method characterized in that a vortex flow rate signal is obtained by taking a difference in signal strength of a noise component measured in advance from the signal strength of each signal that has passed through the band division filter. In claim 1 above,
The band division filter, including the frequency band to pass the alternating stress signal, vortex flow measurement method comprising Rukoto represented by a filter consisting of decimation frequency bands. In the above claim 2 ,
The signal intensity of the noise component is the signal intensity of the signal that has passed through the band dividing filter in a state where the flow rate is zero. In claim 1 above,
A second-order low-pass filter that has an input connected to the output of the low-pass filter, an output connected to the input of the band-splitting filter, and uniformizes the amplitude of the signal regardless of the frequency. Eddy flow measurement method. In claim 4 above
The vortex flow rate measurement is characterized in that the secondary low-pass filter is a digital filter, and makes the amplitude of the vortex signal proportional to the square of the fluid density and flow velocity of the fluid uniform regardless of the frequency. Method. In the vortex flowmeter that measures the flow rate of fluid by detecting the alternating stress signal generated by Karman vortex,
The alternating stress signal is passed through a low-pass filter having a frequency band determined by the diameter and flow range through which the fluid flows,
The signal that has passed through the low-pass filter is passed through a band-splitting filter that is divided into a plurality of frequency bands that are switched to a gain based on the signal intensity of a noise component that has been measured in advance.
A vortex flow rate measurement method characterized in that a vortex flow rate signal is obtained from the signal intensity of each signal that has passed through the band division filter. In the above claim 6,
The band division filter, including the filter of the frequency band for passing the alternating stress signal, vortex flow measurement method comprising <br/> that that will be formed by a filter made of decimation frequency bands. In claim 7 ,
The signal intensity of the noise component is a signal intensity of a signal that has passed through the band dividing filter in a state where the flow rate is zero. In the above claim 6,
A second-order low-pass filter that has an input connected to the output of the low-pass filter, an output connected to the input of the band-splitting filter, and uniformizes the amplitude of the signal regardless of the frequency. Eddy flow measurement method. In claim 9 ,
The vortex flow rate measurement is characterized in that the secondary low-pass filter is a digital filter, and makes the amplitude of the vortex signal proportional to the square of the fluid density and flow velocity of the fluid uniform regardless of the frequency. Method. In the vortex flowmeter that measures the flow rate of fluid by detecting the alternating stress signal generated by Karman vortex,
A low-pass filter that allows the alternating stress signal to pass through a frequency band determined by the diameter and flow rate range through which the fluid flows;
An A / D converter that converts a signal that has passed through the low-pass filter into a digital signal;
A filter band dividing filter for passing the digital signal converted by the A / D converter through a filter composed of a plurality of frequency bands;
And a signal operation unit for obtaining a vortex flow signal by taking the difference in signal intensity of the noise component is previously measured from the signal intensity of each signal passing through the band-splitting filter
Vortex flowmeter characterized by that . In claim 11 above
The band division filter, including the filter of the frequency band for passing the alternating stress signal, the vortex flowmeter, wherein <br/> that that will be formed by a filter made of decimation frequency bands. In claim 12 ,
The signal intensity of the noise component is a signal intensity of a signal that has passed through the band dividing filter in a state where the flow rate is zero. In claim 11 above
A second-order low-pass filter having an input connected to the output of the low-pass filter, an output connected to the input of the band-splitting filter, and uniformizing the amplitude of the signal regardless of the frequency;
The vortex flowmeter is characterized in that the secondary low-pass filter is a digital filter and makes the amplitude of the vortex signal proportional to the square of the fluid density and flow velocity of the fluid uniform regardless of the frequency. . In the vortex flowmeter that measures the flow rate of fluid by detecting the alternating stress signal generated by Karman vortex,
A low-pass filter that allows the alternating stress signal to pass through a frequency band determined by the diameter and flow rate range through which the fluid flows;
An A / D converter that converts a signal that has passed through the low-pass filter into a digital signal;
A band division filter for passing the digital signal converted by the A / D converter through a filter composed of a plurality of frequency bands;
Signal calculation means for switching the gain of each filter constituting the band division filter based on the signal intensity of a noise component measured in advance and obtaining a vortex flow rate signal from the signal intensity obtained by the switched gain provided with a door
Vortex flowmeter characterized by that . In claim 15 above,
The band division filter, including the filter of the frequency band for passing the alternating stress signal, the vortex flowmeter, wherein <br/> that that will be formed by a filter made of decimation frequency bands. In claim 16 ,
The signal intensity of the noise component is a signal intensity of a signal that has passed through the band dividing filter in a state where the flow rate is zero. In claim 15 above,
A second-order low-pass filter having an input connected to the output of the low-pass filter, an output connected to the input of the band-splitting filter, and uniformizing the amplitude of the signal regardless of the frequency;
The vortex flowmeter is characterized in that the secondary low-pass filter is a digital filter and makes the amplitude of the vortex signal proportional to the square of the fluid density and flow velocity of the fluid uniform regardless of the frequency. .

Images