JPH032247B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention determines the flow rate and flow rate of fluid flowing in the pipe from the propagation time of ultrasonic waves transmitted and received between ultrasonic transducers arranged diagonally opposite to the pipe. It relates to an ultrasonic flowmeter, in particular, modulating the transmission signal that drives the ultrasonic transducer with a predetermined modulation signal, and calculating the autocorrelation function of this modulation signal based on the reception signal obtained in response to this and the modulation signal. The present invention relates to an ultrasonic flowmeter in which the autocorrelation function is used to determine the flow velocity and flow rate of the fluid.

FIG. 1 is a diagram showing an example of the basic configuration of an ultrasonic flowmeter. As shown in this figure, ultrasonic pulses emitted by one ultrasonic transducer 1a propagate through the fluid 3 in the conduit 2. The received signal is received by the other ultrasonic transducer 1b, which is disposed diagonally opposite the ultrasonic transducer 1a. Here, if the pipe diameter of the conduit 2 is D, the incident angle of the ultrasonic pulse is θ, the sound velocity in the fluid 3 is C, and the flow velocity of the fluid 3 is V, then the ultrasonic transducer 1a transmits the ultrasonic pulse. The time required for the ultrasonic transducer 1b to receive this ultrasonic pulse after transmitting it T ₁
is given by T ₁ =D/cosθ/C+V・sinθ (1), and conversely, the time required for the ultrasonic pulse emitted by the ultrasonic transducer 1b to be received by the ultrasonic transducer 1a is The time T ₂ is given by T ₂ =D/cosθ/C−V·sinθ (2). Then, by eliminating the sound speed C from these equations (1) and (2) and finding the flow velocity V of fluid 3, we get: V=D/sin2θ・T ₂ −T ₁ /T ₁・T ₂ ……(3) Become. Here, the values normally used in such an ultrasonic flowmeter are D = 100 mm, θ =
Substituting 22° into equation (3), and assuming that the sound velocity C of the fluid 3 measured by this ultrasonic flowmeter is 1450 m/s and the flow velocity V is 1 cm/s, the propagation time difference of the ultrasonic pulses T ₁ −T ₂ is approximately 8 ps (8×10 ⁻¹² seconds). As described above, in an ultrasonic flowmeter, the difference in propagation time of ultrasonic pulses T ₁ -T ₂ is a very small value, so accurate measurement of propagation time of ultrasonic pulses is essential.

By the way, conventionally known methods for measuring the propagation time of such ultrasonic pulses include the sing-around method (or PLL method) and the pulse propagation time difference method. Both the sing-around method and the pulse propagation time difference method calculate T ₁ and T ₂ individually from the propagation time of the ultrasonic pulse, but they have the following drawbacks.

() Since the ultrasonic waves used for measurement are in the form of pulses, if there are bubbles or dust in the propagation path of the ultrasonic pulses, the ultrasonic pulses will be attenuated and errors will occur in the measurement of propagation time.

() Since an ultrasonic transducer that generates ultrasonic waves does not vibrate immediately even when a voltage is applied, the ultrasonic pulse obtained by such an ultrasonic transducer has its envelope as shown in Figure 2. a becomes gradual. For this reason,
When measuring the propagation time, a comparison level l is set, and the time when the ultrasonic pulse exceeds this comparison level l is measured to measure the propagation time. If the amplitude of the pulse (specific wave) changes beyond , a measurement error will occur.

() Furthermore, if the comparison level l is set low in order to eliminate measurement errors due to such envelope changes, etc., not only will it be more susceptible to the effects of noise, etc., but the acoustic energy of the ultrasonic pulse with level h will also be wasted. .

() Also, in order to eliminate measurement errors due to envelope changes using ultrasonic pulses with a sharp rise, it is necessary to instantaneously apply a high voltage to the ultrasonic transducer, but in order to do this, the circuit design must be devised. There is a need. Furthermore, in this case, the circuit becomes complicated and the reliability of the circuit elements is lowered because a high voltage is applied to the circuit elements, and since a high voltage is used, it is not preferable in terms of explosion protection.

On the other hand, 1 measurement method different from the measurement method described above
A phase difference method using continuous waves is also known, but such a phase difference method has the following drawbacks.

() Since the propagation time is measured using a continuous wave, a carrier wave is required to obtain the propagation time difference T ₁ − _{T 2} shown in equation (3) above, and measurement errors occur due to frequency fluctuations of this carrier wave. arise.

() Furthermore, the propagation times T ₁ and T ₂ of the ultrasonic pulses as shown in equation (3) above cannot be determined individually.

In view of the above points, the present invention has been made in the above () to
The present invention provides an ultrasonic flowmeter that can eliminate the drawbacks shown in () and also that can determine the propagation times T ₁ and T ₂ of ultrasonic pulses with high accuracy.
One side of the ultrasonic transducer is driven by a modulated signal obtained by modulating the carrier signal with a predetermined modulation signal, and the resulting ultrasonic wave is received by the other side of the ultrasonic transducer to obtain a reception signal. The present invention is characterized in that an autocorrelation function of the modulated signal is determined from the signal and the modulated signal, and the flow rate and flow rate of the fluid are determined based on this autocorrelation function.

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 3 is a block diagram showing a first embodiment of an ultrasonic flowmeter according to the present invention. In this diagram,
Reference numeral 4 denotes an oscillator that generates a carrier signal of a constant frequency, and the output of this oscillator 4 (carrier signal S1) is supplied to a modulator 5. Modulator 5 is modulation signal generator 1
7 based on the modulated signal S2 from the carrier signal S
1 (for example, FM modulation),
The output of this modulator 5 (drive signal S3) is supplied to a transmitting circuit 6. The transmission circuit 6 receives the drive signal S3.
The drive voltage obtained by this amplification operation is supplied to one of the ultrasonic transducers 1a via the switch 7, and the ultrasonic transducers 1a transmit ultrasonic waves. On the other hand, ultrasonic transducer 1
The ultrasonic waves received by b are sent to the same ultrasonic transducer 1.
After being converted into a corresponding reception signal by b, the received signal is supplied to the reception circuit 8 via the switch 7. The receiving circuit 8 amplifies the received signal, and the output (signal S4) of this receiving circuit 8 is supplied to a demodulator 9. The demodulator 9 demodulates the signal S4, and the output of this demodulator 9 (demodulated signal S5)
is supplied to the arithmetic unit 10. The arithmetic unit 10 multiplies the modulated signal S2 and the demodulated signal S5, and calculates the average value of the multiplication results to obtain the autocorrelation function of the modulated signal S2.The arithmetic result (flow velocity) of this arithmetic unit 10 is It is output from the output terminal 11. That is, here, if the modulated signal S2 is φ(t), the demodulated signal S5 is expressed as φ(t-τ), so the autocorrelation function Ψ(τ) obtained by the calculator 10 is
is, Ψ(τ)=lim T→∞1/T∫ ^T ₀ φ(t)・φ(t−τ)dt
...(4) becomes. Here, if the modulation signal S2 is, for example, an M-sequence signal whose clock period (parameter indicating a pseudo-random number) is t ₀ and whose signal period indicating repetition is N, the autocorrelation function Ψ( τ), the time difference τ is (k・N−1)・t ₀ τ
When in the range of (k・N+1)・t ₀ , Ψ(τ)=1−(1−1/N)・|τ−k・N・t ₀ |/
t ₀ ...(5) However, k is an integer.

When the time difference τ is outside the above range, Ψ(τ)=-1/N...(6). Therefore, if the modulation signal S2 is selected in this way, the autocorrelation function Ψ(τ) of this modulation signal S2
exhibits characteristics as shown in FIG. That is, if the clock period t ₀ and signal period N are selected to satisfy the following equation (N-1) t ₀ < T ₁ < Nt ₀ ...(7), the self One ultrasonic propagation time T ₁ can be determined from the value of the correlation function Ψ(τ) and the above equation (5).

Furthermore, in this example, the ultrasound propagation time T ₁ is
If it is within the range shown by equation (7), as is clear from equation (5), the autocorrelation function Ψ(T ₁ ) at this time is the variable Nt ₀ −T ₁
Since it changes linearly with respect to Ψ, the autocorrelation function Ψ
Once (T ₁ ) is determined, the ultrasonic propagation time T ₁ can be easily determined from this autocorrelation function Ψ(T ₁ ). Also, the autocorrelation function Ψ(τ) has a time difference τ of 0.
When the value is in the range of ~(N-1) _t0 , the value is constant at -1/N, and when it is in the range of (N-1) _t0 ~ _Nt0 , it has a large gain. In this way, since the autocorrelation function Ψ(T ₁ ) has an offset of (N-1)t ₀ , the accuracy of time measurement can be improved, and the ultrasonic propagation time T ₁ can be determined with high precision. Can be done. Furthermore, by switching the switch ₇ , the ultrasonic propagation time _T2 in the opposite direction can be determined in the same manner as the ultrasonic propagation time T1 described above.

Next, the integration period T in the ultrasonic flowmeter according to the present invention will be discussed using specific numerical values. First, in this invention, the autocorrelation function of the modulated signal S2 is determined by setting the integration period T to infinity as shown in equation (4) above, but the actual integration is performed over a finite interval (a finite period )T ₀ , the autocorrelation function Ψ'(τ) in this case has an error with respect to the true value Ψ(τ).

The relationship between this error and the integration period T ₀ will be described below.

First, the expected value E of the autocorrelation function Ψ′(τ)
[Ψ′(τ)] is given by the formula E[Ψ′(τ)]T ₀ −t ₀ /T ₀・Ψ(τ) ...(8), and the variance Var
[Ψ′(τ)] is Var[Ψ′(τ)]1/T ₀ ∫ ^Nt0 ₀ Ψ ² (τ)dτ 2N ² +5N−4/3N ²・t ₀ /T ₀ ……(9) It is given by Eq. Therefore, here, |Ψ(τ)−E[Ψ′(τ)]|,√[′(
)〕＜
7×10 ^-4 /2, t ₀ /T ₀ <3.5×10 ^-4 is obtained from the above equation (8), and t ₀ /T ₀ <8×10 ^-8 is obtained from the above equation (9). can get. Therefore, the integration period T ₀ is T ₀ > t ₀ /8×10 ^-8 ……(10), and especially when the signal period N is N=127, T ₀ > 10 ⁵・T _c ……( 11) However, T _c is the ultrasonic propagation time. Therefore, if the pipe diameter of pipe line 2 (see Fig. 1) is set to D = 25 mm, the integration period T ₀ becomes T ₀ >1.8 (SEC) (12). As is clear from this equation (12)C, the integration period
T ₀ is a sufficiently small and practical value.

Next, using specific numerical values, we will consider the difference between the resolution required for the ultrasonic flowmeter according to the present invention and the resolution required for a conventional ultrasonic flowmeter. First, we will discuss the resolution required for conventional ultrasonic flowmeters. The ultrasonic propagation time T _c in the conventional ultrasonic flowmeter is calculated from the above equations (1) and (2) as T _c ≈D/C·cosθ (13) where V/C≪1. Furthermore, in the above equation (3), if T ₂ −T ₁ = δT, then δT = sin2θ/D・V・T ₁・T ₂ ≒sin2θ/D・V・T ² _c ……(14) is obtained. . Therefore, the ultrasound propagation time T ₁ ,
The resolution δT/T _c required when determining T ₂ directly is δT/T _c 2V/C·sinθ (15). Here, the incident angle θ of the ultrasonic wave is 22°, and the fluid 3
The sound velocity C inside is 1450 m/s, and the flow velocity V of fluid 3 is 1 cm.
Then, the resolution δT/T _c shown in equation (15) becomes δT/T _c 5×10 ^-6 ...(16).

On the other hand, in the ultrasonic flowmeter according to the present invention, the ultrasonic propagation time T _c is given by T _c Nt ₀ (17). Therefore, from this equation (17) and the above equations (13) and (14), the ultrasonic propagation time T ₁ ,
The resolution δT/t ₀ required to obtain T ₂ is δT/t ₀ 2N·V/C·sinθ (18). That is, in this ultrasonic flowmeter, the required resolution can be lowered by the signal period N. For example, if the signal period N is N=127, the resolution δT/t ₀ at this time is δT/t ₀ 7×10 ^-4 (19).

In this way, in this ultrasonic flowmeter, by decomposing the time base (clock period) t ₀ to the required accuracy, the ultrasonic propagation times T ₁ and T ₂ can be determined with the required accuracy relatively easily. .

Note that in this embodiment, the propagation time T ₁ is (N
-1) Set the clock period t 0 and the signal period N so that t ₀ < T ₁ < N・t ₀ , that is, _the propagation time T ₁ matches the period of the modulation signal S2. However, the clock period t ₀ and the signal period N are adjusted so that this propagation time T ₁ is around an arbitrary integer multiple of Nt ₀ .
You may also set each value.

FIG. 5 is a block diagram showing a second embodiment of the ultrasonic flowmeter according to the present invention. In this diagram,
12 is a modulator that modulates the carrier signal S1 obtained by the oscillator 4, and this modulator 12 is connected to the terminal 13.
The carrier signal S1 is FM modulated based on the modulation signal S2a supplied via the drive signal S3a (this drive signal S3a has a frequency _{of high}
and a signal with a frequency of _LOW are alternately combined) to the intermodulation circuit 14, and also to one ultrasonic transducer 1a via the transmitting circuit 6 and the switch 7 in sequence, Ultrasonic waves are transmitted from the ultrasonic transducer 1a. On the other hand, the ultrasonic wave received by the ultrasonic wave transducer/receiver 1b is converted into a corresponding reception signal and then supplied to the intermodulation circuit 14 via the switch 7 and the reception circuit 8 in sequence. The cross-modulation circuit 14 mixes the driving signal S3a and the signal S4a supplied via the receiving circuit 8, and supplies the beat signal S6 obtained by the mixing to the counter 15. Here, if the drive signal S3a is, for example, as shown in FIG. 6A, and the signal S4a is as shown in FIG.
The beat signal S6 supplied to the counter 15 is as shown in FIG. 6C, since it is zero when the frequency of the signal S4a and the frequency of the signal S4a are the same, and _high - _LOW when they are different. That is, the counter 15 is supplied with a pulse train beat signal S6 proportional to the phase difference (ultrasonic propagation time) _T1 between the drive signal S3a and the signal S4a and the frequency difference _high - _LOW of the drive signal S3. Therefore, the counting result of the counter 15 is proportional to 1-Ψ(T ₁ ),
The count result of this counter 15 is supplied to the arithmetic unit 10a. The computing unit 10a determines the ultrasonic propagation time T ₁ from the counting result, and also determines the flow velocity V from the ultrasonic propagation time T ₁ and the ultrasonic propagation time T ₂ obtained by switching the switch 7.

In this way, even when two frequencies, _high and _low, are used, the autocorrelation function Ψ(τ) can be equivalently determined from the counting results of the counter 15. Furthermore, in this case, similar measurement results can be obtained by using a frequency comparator instead of the intermodulation circuit 14 and integrating the output of this frequency comparator.

FIG. 7 is a block diagram showing a third embodiment of the ultrasonic flowmeter according to the present invention. In this figure, parts corresponding to those in FIG. 3 are given the same reference numerals. In this figure, 16 is a switch for circuit calibration, and the common contact 16a and the contact 16b of this switch 16 are connected at predetermined time intervals or at arbitrary timing. As a result, the modulated signal S is transmitted via this switch 16.
2 is supplied to the arithmetic unit 10, and the autocorrelation function Ψ(O) at this time is determined in the arithmetic unit 10.
Here, this modulation signal S2 is, for example, the above-mentioned M
In the case of a serial signal, the autocorrelation function Ψ(O) becomes Ψ(O)=1, so this Ψ(O)=1 is used to calculate the autocorrelation function calculation circuit and the integral Circuits etc. can be calibrated.

FIG. 8 is a block diagram showing a fourth embodiment of the ultrasonic flowmeter according to the present invention, and in this figure, parts corresponding to those in FIG. 5 are given the same reference numerals. As shown in this figure, in this ultrasonic flowmeter, by switching the switch 18, the intermodulation circuit 13 to the arithmetic unit 10a can be calibrated.

Furthermore, in the first to fourth embodiments described above,
An M-sequence signal is used as a modulation signal, but this can be used, for example, as an average remainder sequence (L sequence), a twin prime number sequence,
A periodic pseudo-random signal such as a Berger sequence or an I sequence or a sine wave signal may be used. In this case as well, since the autocorrelation functions of all these modulated signals have periodic sharp peaks, it is possible to apply an offset to the time measurement as in the case described above.

As explained above, the ultrasonic flowmeter according to the present invention drives one of the ultrasonic transducers with a modulated signal obtained by modulating a carrier signal with a predetermined modulation signal, and transmits the ultrasonic wave obtained thereby. The ultrasonic wave is received by the other side of the ultrasonic transducer, and an autocorrelation function of the modulation signal is determined from the received signal obtained thereby and the modulation signal, and the flow rate of the fluid is determined based on this autocorrelation function. The following effects can be obtained.

() Since the signals in the circuit, including the ultrasonic transducer, are continuous waves, AGC (auto gain control) is easy to apply, and they are not affected by attenuation in the fluid.

(), () If FM modulation or the like is used as the modulation, it is possible to eliminate detection errors caused by the rise delay of the ultrasonic transducer.

() Since the signal in the circuit is a continuous wave, large acoustic power can be obtained even if the drive voltage of the ultrasonic transducer is lowered, and the transmitter circuit is also simpler.
Circuit reliability can be improved. Furthermore, in this case, since the voltage is low, it is advantageous in terms of explosion protection.

() Even if the frequency of the carrier signal that drives the ultrasonic transducer changes, the measured value does not change, so the oscillator that generates the carrier signal can be simplified. In this case, since the clock period t ₀ of the modulation signal directly affects the measurement accuracy, it is necessary to stabilize the clock period t ₀ , but in this case, a crystal oscillator or the like may be used.

() The ultrasonic propagation times T ₁ and T ₂ can be determined individually.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of the basic configuration of a general ultrasonic flowmeter, FIG. 2 is a waveform diagram for explaining FIG. 1, and FIG.
A block diagram showing an embodiment, FIG. 4 is a characteristic diagram for explaining FIG. 5, FIG. 5 is a block diagram showing a second embodiment of the ultrasonic flowmeter according to the present invention, and FIG. FIG. 7 is a block diagram showing a third embodiment of the ultrasonic flowmeter according to the present invention, and FIG. 8 is a block diagram showing a fourth embodiment of the ultrasonic flowmeter according to the present invention. be. 1a, 1b...Ultrasonic transducer, 2...Pipe line,
3... Fluid, 5, 5a... Modulator, 9... Demodulator, 10, 10a... Arithmetic unit, 14... Cross modulation circuit, 15... Counter.

Claims (1)

[Claims] 1. In an ultrasonic flowmeter that calculates the flow rate and flow rate of fluid flowing in a pipe from the propagation time of ultrasonic waves transmitted and received between ultrasonic transducers arranged diagonally opposite to the pipe, the carrier signal is One of the ultrasonic transducers is driven by a modulated signal modulated with a predetermined modulation signal, and the resulting ultrasonic wave is received by the other of the ultrasonic transducers to obtain a received signal. and an ultrasonic flowmeter characterized in that an autocorrelation function of the modulation signal is determined from the modulation signal, and a flow velocity and a flow rate of the fluid are determined based on this autocorrelation function.