JP2013088322A - Method for measuring flow velocity and flow volume - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a flow velocity measuring method by a correlation method, which does not need a temperature sensor.SOLUTION: The flow velocity measuring method for calculating the flow velocity of liquid of a flow channel by using a correlation method includes a process for acquiring waveform data of a first ultrasonic signal propagating forward between two points of the flow channel, a process for acquiring waveform data of a second ultrasonic signal propagating backward between the two points of the flow channel, a process for calculating propagation time difference from a mutual correlation function of the waveform data of the first and second ultrasonic signals, a process for extracting a waveform feature point from the waveform data of the first ultrasonic signal to calculate a forward propagation time of the ultrasonic signal, a process for extracting a waveform feature point from the waveform data of the second ultrasonic signal to calculate a backward propagation time of the ultrasonic signal, a process for calculating a sound velocity of the ultrasonic signal on the basis of a distance between the two points of the flow channel as well as the forward propagation time and backward propagation time of the ultrasonic signal, and a process for calculating a flow velocity of the liquid on the basis of the distance between the two points of the flow channel, the propagation time difference, the sound velocity, or the like.

Description

  The present invention relates to an improvement of a measurement method using ultrasonic waves that can be applied to measurement of fluid flow velocity, flow rate, and the like. In particular, the present invention relates to an improvement in a flow velocity (or flow rate) measurement method that directly obtains a propagation time difference between an ultrasonic signal propagating in a forward direction through a predetermined flow path and an ultrasonic signal propagating in the reverse direction through the flow path.

  In order to measure the flow rate of the fluid flowing through a flow path such as a pipe, for example, an ultrasonic flow meter or the like is used. There are several types of ultrasonic flowmeters, but with ultrasonic flowmeters that use a method that focuses on the difference in propagation time of ultrasonic signals, ultrasonic signals propagate in the fluid from upstream to downstream (forward direction). The flow velocity of the fluid is calculated from the propagation time difference Δt between the ultrasonic signal propagating from the downstream side to the upstream side (reverse direction), the propagation speed C of the ultrasonic signal, and the like (propagation time difference method). The flow rate is calculated from the flow velocity and the cross-sectional area of the flow path.

However, it is not easy to detect the propagation time of the ultrasonic signal to the upstream side and the propagation time to the downstream side with a required time accuracy.
Thus, for example, paragraphs 0003 and 0014 and Japanese Patent Laid-Open No. 2005-181268 (Patent Document 1) of the prior example describe directly obtaining the propagation time difference Δt by the correlation method. In the correlation method, the cross-correlation value between the entire reception signal waveform of the forward ultrasonic signal and the entire reception signal waveform of the reverse ultrasonic signal is obtained, and the time difference Δt is directly obtained from the peak of the correlation value. From this time difference Δt and the sound velocity C of the ultrasonic signal, the flow velocity V is calculated by the equation (V = (C ² / 2L cos θ) Δt) described in detail later. Here, L is a distance between two points on the flow path, and θ is an angle formed by the flow direction of the fluid and the propagation direction of the ultrasonic signal. The speed of sound C is obtained as a function of the temperature T as follows, for example, when the fluid to be measured is atmospheric pressure air.
C = 331.5 + 0.61T [m / S]

JP 2005-181268 A

  However, in the propagation time difference method using the above-described correlation method, in order to calculate the flow velocity of the fluid, the sound speed C in the fluid must be obtained in addition to the propagation time difference Δt of the ultrasonic signal. In order to calculate the speed of sound C, a temperature sensor that detects the temperature T of the fluid is required. This increases the cost of the velocimeter (or flow meter) and increases the size of the velocimeter in order to secure a space for installing the temperature sensor in the meter. Further, the speed of sound tends to change not only with the fluid temperature but also with the component and pressure of the fluid. In this case, an error occurs in the calculation formula C.

  Accordingly, an object of the present invention is to provide a flow velocity measurement method that does not require a temperature sensor (or temperature measurement) in a flow velocity measurement method based on a propagation time difference method (correlation method).

  Another object of the present invention is to provide a flow rate measurement method that does not require a temperature sensor in the flow rate measurement method based on the propagation time difference method (correlation method).

  One aspect of the present invention that solves the above-described problem is a method for measuring a flow velocity of a fluid in a flow path by using a correlation method. The flow velocity is measured in a forward direction between two points in the flow path. The process of acquiring the waveform data of the ultrasonic signal, the process of acquiring the waveform data of the second ultrasonic signal propagating in the reverse direction between the two points of the flow path, and the first and second ultrasonic signals The process of obtaining the propagation time difference from the correlation function (cross-correlation calculation) of the waveform data of the waveform, the feature point of the waveform is extracted from the waveform data of the first ultrasonic signal, and the ultrasonic signal between the two points of the flow path The process of obtaining the forward propagation time of the second, extracting the feature point of the waveform from the waveform data of the second ultrasonic signal, and obtaining the backward propagation time of the ultrasonic signal between the two points of the flow path, A distance between two points of the flow path, a forward propagation time of the ultrasonic signal, and Comprising the steps of calculating the sound speed of the ultrasound signal based on the direction of propagation time, the distance between two points of the flow path, the propagation time difference, and a process of calculating the flow rate of the fluid based on the speed of sound.

  With such a configuration, the propagation time difference is directly obtained by the cross-correlation calculation (described later) of the signal data of the two ultrasonic signals. Moreover, the signal propagation time in the forward direction and the reverse direction can be obtained from the signal data, and the sound speed can be calculated. The flow velocity is calculated from the propagation time difference, sound velocity, etc. As a result, there is an advantage that a thermometer is not required.

  The propagation time of the ultrasonic signal between the sensors is very short, and it is virtually impossible to A / D convert the signal with a time resolution that can directly detect the propagation point of the ultrasonic signal. Although a detection circuit is required, in the correlation method, since the entire signal is compared, the signal sampling interval can be set long enough not to lose the characteristics of the signal waveform, so a commercially available A / D converter should be used. I can do it. Then, the propagation time difference is directly obtained by the cross-correlation calculation of the signal data of the two ultrasonic signals.

  Furthermore, as a result of various studies by the inventor, the signal data acquired for the correlation method has a relatively low time resolution in the A / D conversion. It has been found that it is not usable in the case of a detection method that detects and “calculates the flow velocity”, but can be used when “propagation time” for “calculating the speed of sound” is obtained. As a result, the velocity of sound can be obtained by calculation using the signal data acquired for the correlation method, and the flow velocity can be measured without the need for a temperature sensor or the like.

Desirably, the process of obtaining the sound speed is as follows: C is the sound speed, L is the distance between the two points, t _{1 is} the forward propagation time, and t _{2 is the} backward propagation time t ₂ .
C = (L / 2) {(1 / t ₁ ) + (1 / t ₂ )}
To calculate the speed of sound C.

Desirably, the process of calculating the flow velocity of the fluid includes the flow velocity of the fluid as V, the sound velocity as C, the propagation time difference as Δt, and the angle formed between the flow direction of the fluid and the propagation direction of the ultrasonic signal as θ. The flow velocity V is obtained by calculating V≈ (C ² / 2L cos θ) Δt.

  Preferably, the process of obtaining the forward propagation time and the backward propagation time corrects a predetermined value from the position on the time axis of the feature point of the detected waveform of the ultrasonic signal.

  In the flow rate measurement method of the present invention, the flow rate Q is obtained by multiplying the flow velocity V obtained by the flow velocity calculation method by the cross-sectional area S of the flow path.

  One aspect of the present invention is an ultrasonic anemometer that obtains a flow velocity of a fluid in a flow path by using a correlation method, and a waveform of a first ultrasonic signal that propagates between two points of the flow path in a forward direction. Means for acquiring data, means for acquiring waveform data of the second ultrasonic signal propagating in the opposite direction between two points of the flow path, and mutual waveform data of the first and second ultrasonic signals. Means for obtaining a propagation time difference from the correlation function, means for extracting a characteristic point of the waveform from the waveform data of the first ultrasonic signal, and obtaining a forward propagation time of the ultrasonic signal between the two points of the flow path; Means for extracting a characteristic point from the waveform of the second ultrasonic signal and obtaining a backward propagation time of the ultrasonic signal between the two points of the flow path; a distance between the two points of the flow path; and the ultrasonic wave The sound speed of the ultrasonic signal is obtained based on the forward propagation time and backward propagation time of the signal. And means that the distance between two points of the flow path, the propagation time difference, and means for calculating the flow velocity of the fluid based on the speed of sound, a.

  With such a configuration, it is possible to obtain an anemometer that does not require a temperature sensor. The inventor found that the signal data obtained by the correlation method has a relatively low time resolution in the A / D conversion, so that the flow rate detection (calculation) in the detection method in which the propagation time of the ultrasonic signal is directly detected is used. Although it cannot be used, it has been found that it can be used to calculate the speed of sound. By reusing the signal data and calculating the sound speed, it is possible to configure an anemometer that does not require a temperature sensor or the like for calculating the sound speed.

  The flow meter according to one aspect of the present invention is an ultrasonic flow meter for obtaining a flow velocity of a fluid in a flow path by using a correlation method, and the first ultrasonic signal propagating forward between two points of the flow path. Means for obtaining the waveform data of the second ultrasonic signal, means for obtaining the waveform data of the second ultrasonic signal propagating in the opposite direction between the two points of the flow path, and waveform data of the first and second ultrasonic signals. , A means for determining a propagation time difference from the correlation function between the waveform data of the first and second ultrasonic signals, and a feature point of the waveform from the detected waveform data of the first ultrasonic signal. Means for obtaining the forward propagation time of the ultrasonic signal between the two points of the flow path, and extracting characteristic points of the waveform from the waveform data of the second ultrasonic signal. Between the two points of the channel and the means for obtaining the backward propagation time of the ultrasonic signal between Based on the distance, the means for obtaining the sound speed of the ultrasonic signal based on the forward propagation time and the reverse propagation time of the ultrasonic signal, the distance between the two points of the flow path, the propagation time difference, and the sound speed. Means for obtaining a flow rate of the fluid, and means for obtaining a flow rate by multiplying the obtained flow rate by the cross-sectional area of the flow path.

  By adopting such a configuration, it becomes possible to obtain a correlation method flowmeter that does not require a temperature sensor.

Desirably, the means for obtaining the speed of sound in the flowmeter or flowmeter is such that the speed of sound is C, the distance between the two points (between the transmitter and the receiver) is L, the forward propagation time is t ₁ , and the reverse By calculating C = (L / 2) {(1 / t ₁ ) + (1 / t ₂ )} as the direction propagation time t ₂ , the sound velocity C of the ultrasonic signal is obtained. Thereby, the speed of sound can be calculated from the measured waveform data. In addition, a temperature sensor is not required for calculating the speed of sound.

Desirably, the means for obtaining the flow velocity of the fluid includes the flow velocity of the fluid V, the sound velocity C, the time difference between the forward propagation time and the backward propagation time Δt, and the fluid flow direction and the ultrasonic signal. The flow velocity V is obtained by calculating V≈ (C ² / 2L cos θ) Δt, where θ is the angle formed with the propagation direction of.

  Thereby, since the sound velocity C is calculated by the above equation using the measured signal waveform data, the flow velocity (flow rate) that does not require the temperature sensor to calculate the flow velocity V can be measured.

  Preferably, the means for acquiring the forward propagation time and the backward propagation time discriminates a feature point of the waveform from the signal waveform data of the ultrasonic signal acquired in the correlation method, and the position of the feature point on the time axis From this, a predetermined value is corrected. Since the time difference from the propagation time (reception time) of the ultrasonic signal to the feature point is substantially constant, the propagation time can be obtained from the feature point by measuring the correction value in advance.

  Desirably, the characteristic point of the signal waveform of the ultrasonic signal is a first zero cross point after a predetermined ratio of the maximum amplitude value of the waveform is set as a threshold and the waveform exceeds the threshold. Thereby, it is possible to set a threshold value corresponding to the waveform pattern and extract a zero cross point that is less affected by noise.

  Desirably, two points of the flow path are set by two transceivers. Thereby, the occupied space in the measuring device of the transmitter and the receiver can be reduced.

According to the present invention, in detecting the flow velocity V in the predetermined flow path, the propagation time difference Δt between the two ultrasonic signals in the forward direction and the reverse direction is directly calculated from the waveform data of the two ultrasonic signals by the correlation method (cross-correlation calculation). calculate. Further, the propagation times t ₁ and t ₂ in the forward direction and the reverse direction of the ultrasonic signals are obtained from the waveform data of the two ultrasonic signals, and the sound velocity C is calculated from these propagation times. Since the flow velocity V is calculated based on the obtained sound velocity C and propagation time difference Δt, there is no need to measure temperature, fluid component, pressure, and the like. This is also convenient in terms of circuit cost reduction and device size reduction.

It is explanatory drawing explaining the apparatus structure which measures the flow velocity (or flow volume) of the fluid. It is explanatory drawing explaining the example which measures signal propagation time in a forward direction and a reverse direction with a set of transmitter / receivers arrange | positioned at the flow path. It is explanatory drawing explaining the example which calculates | requires propagation time difference by the correlation method. It is a graph explaining the relationship between temperature and sound velocity (when a medium is water). It is explanatory drawing explaining the 1st example which calculates propagation time from the feature point of a received signal. It is explanatory drawing explaining the 2nd example which calculates propagation time from the feature point of a received signal. It is explanatory drawing explaining the 3rd example which calculates propagation time from the feature point of a received signal. It is a flowchart explaining the procedure which measures the flow velocity (or flow volume) of a fluid.

Embodiments of the present invention will be described below with reference to the drawings.
In the embodiment of the present invention, two ultrasonic signals propagating in the forward and reverse directions in the same channel by a set of ultrasonic transmitters / receivers arranged at a predetermined distance in the channel are used as electrical signals, and A / Data is converted into data by the D converter and the signal data of the entire signal waveform is taken into the memory. Then, a cross-correlation operation (correlation method) between the waveform data of the ultrasonic signal in the forward direction and the waveform data of the ultrasonic signal in the reverse direction is performed, and the propagation time difference Δt is directly obtained from the correlation function.

Further, the MPU (microcomputer) extracts the characteristic points of the signal waveform from the signal data of the forward and reverse ultrasonic signals captured for the cross-correlation calculation, and the propagation time t ₁ and reverse in the forward direction of the flow path. The propagation time t ₂ in the direction is obtained. The speed of sound C is calculated from the propagation times t ₁ and t ₂ . Based on the calculated sound velocity C and propagation time difference Δt, the fluid flow velocity V is calculated. Further, the flow rate Q is calculated based on the flow velocity V and the cross-sectional area S of the flow path. As a result, the sound speed can be obtained without using the temperature parameter in the correlation method.

For convenience of explanation, explanation will be given in the following order.
(Configuration of Example)
(Explanation of propagation time difference method and correlation method)
(thermometer)
(Sound velocity calculation)
(Calculation of propagation time from received signal waveform)
(Zero cross method)
(Maximum amplitude method)
(Combination of threshold and zero cross method)
(Measurement procedure)
(Explanation of effect of embodiment)
(Comparative example)
(Time accuracy when calculating flow velocity)
(Time accuracy when calculating sound speed)

(Configuration of Example)
FIG. 1 is an explanatory view for explaining an example of an ultrasonic current meter (or a flow meter) in which the flow velocity measuring method of the present invention is used.
As shown in the figure, the ultrasonic current meter is roughly constituted by a controller 10, a transmission system 20, a sensor system 30, and a reception system 40.

The transmission system 20 includes an amplifier 21 that amplifies the transmission signal and a signal switch 22 that switches a transmission signal supply destination.
The sensor system 30 includes ultrasonic transceivers 32 and 33 provided on the pipe wall surface of a pipe (flow path) 31 through which a fluid such as water or air flows. As will be described later, the transceiver 32 is provided on the upstream side of the flow path, and the transceiver 33 is provided on the downstream side of the flow path. The distance between the sensors between the transceivers 32 and 33 is set to L [m], and the angle formed between the line segment connecting the sensors and the pipe axis of the pipe is set to θ.

  The ultrasonic transmitters / receivers 32 and 33 are configured by electromechanical transducers (for example, piezoelectric elements) that generate mechanical vibrations when a high-frequency voltage is applied and generate a voltage according to an applied pressure. Note that the ultrasonic transmitter / receiver 32 (or 33) may be configured separately by a transmitter that converts high-frequency signals into mechanical vibrations and a receiver that converts ultrasonic vibrations into electrical signals. A transceiver and a transmitter or receiver have substantially the same meaning.

  The reception system 40 includes a switch 41 for selecting reception signals of the ultrasonic transceivers 32 and 33, an amplifier 42 for level-amplifying the reception signals, and sampling the reception signals at a predetermined cycle to obtain reception data (waveform data of the reception signals). An A / D converter 43 to be obtained, a memory 44 for storing received data, and the like are included. The A / D converter 43 and the memory 44 operate in synchronization with the clock signal supplied from the controller 10.

  The amplifier 42 may change the amplification factor according to the level of the received signal (for example, automatic gain adjustment). In this case, since the phase of the output signal may change by changing the amplification factor, the phase change with respect to the amplification factor is grasped in advance and corrected. Alternatively, the data may be supplemented to the data interval of discrete received data stored in the memory 44 by signal processing (a suitable complementing method or the like) by a separate MPU (microprocessor) to obtain a smoother waveform.

  The controller 10 includes a known computer system such as a high-speed MPU, a DSP (digital signal processor), a memory, a timer, and an interface. The controller 10 is connected to a display 11 that displays process parameters, process states, and the like.

The controller 10 operates according to a control program (not shown). For example, the controller 10 controls the switch 22 to select a transmission signal destination, and controls the switch 41 to select a reception source of the reception signal. Thereby, the transmitter / receiver 32 transmits a transmission signal, and the transmitter / receiver 33 receives the signal to detect propagation of the ultrasonic signal in the forward direction of the flow path. In addition, the transmitter / receiver 33 transmits a transmission signal, and the transmitter / receiver 32 receives the signal to detect the propagation of the ultrasonic signal in the reverse direction of the flow path. As described above, each detected reception signal is stored in the memory 44.
Further, the controller 10 calculates a correlation value from the waveform data of the reception signal (ultrasonic signal) in the forward direction and the waveform data of the reception signal (ultrasonic signal) in the reverse direction by a correlation method program (cross-correlation calculation), The propagation time difference is obtained from the change (waveform) of the correlation value. Further, a waveform analysis program (signal processing program) for extracting feature points in the waveform data of the received signal is executed. Further, the flow rate and the flow rate are calculated based on the parameters detected by executing the mathematical formula processing program or the like.

(Explanation of propagation time difference method and correlation method)
Next, with reference to FIG. 2, detection of ultrasonic flow velocity by the propagation time difference method and detection of the propagation time difference by the correlation method will be described.
FIG. 2A shows an example (forward measurement) in which an ultrasonic signal is transmitted from the upstream transceiver 32 toward the downstream transceiver 33. FIG. 2B shows an example (measurement in the reverse direction) in which an ultrasonic signal is transmitted from the downstream transmitter / receiver 33 to the upstream transmitter / receiver 32. In both figures, the flow velocity of the fluid to be obtained is V [m / S], the velocity of the ultrasonic signal is C [m / S], the distance between the transceivers 32 and 33 is L [m], the flow direction of the fluid The angle formed by the propagation direction of the sound wave signal is θ, the cross-sectional area of the flow path is S [m ² ], the propagation time t ₁ [S] to the downstream side, and the propagation time t ₂ [S] to the upstream side.

The propagation time t ₁ of the ultrasonic signal emitted from the upstream transmitter / receiver 32 and received by the downstream transmitter / receiver 33 is expressed as follows.
t ₁ = L / (C + V cos θ) Equation (1)

The propagation time t ₂ of the ultrasonic signal emitted from the downstream transmitter / receiver 33 and received by the upstream transmitter / receiver 32 is expressed as follows.
t ₂ = L / (C−V cos θ) Equation (2)

From the equations (1) and (2), the propagation time difference Δt is obtained.
Δt = t ₂ −t ₁
= L {(C + Vcosθ) − (C−Vcosθ)} / (C−Vcosθ) (C + Vcosθ)
In this equation, when the condition of C >> V is applied, the propagation time difference Δt is Δt≈2LV cos θ / C ²
From this, the flow velocity V of the fluid is obtained by the following equation (propagation time difference method).
V≈ (C ² / 2L cos θ) Δt Equation (3)

Further, the flow rate Q is expressed by the following equation, where S is the cross-sectional area of the flow path and k is the flow rate correction coefficient.
Q = kSV Formula (4)
The flow rate correction coefficient k is a ratio of the average flow velocity of the pipe cross section to the average flow velocity in the ultrasonic propagation path, and varies depending on the flow path shape, the ultrasonic propagation path, and the Reynolds number. When the flow path shape and the ultrasonic wave propagation path are fixed, it is a function of the Reynolds number.

By the way, it is difficult to detect the signal propagation times t ₁ and t ₂ with required numerical values by sampling the received signal with the current time resolution of the A / D converter. For this reason, in order to detect the propagation times t ₁ and t ₂ , for example, an analog detection circuit that operates at high speed is uniquely assembled, or the propagation method is used (without obtaining the propagation times t ₁ and t ₂ ). The time difference Δt is obtained directly. Examples of the former include ultrasonic flowmeters described in Japanese Patent Application Laid-Open Nos. 10-332452 and 2001-141537 (described in a comparative example described later). The present invention employs the latter correlation method.
In the embodiment of the present invention, in the calculation formula of the propagation time difference method expressed as the above equation (3), the propagation time difference Δt is obtained by the correlation method from the signal waveform data of two received signals. Further, the sound velocity C is obtained by calculation using the signal waveform data of the received signal acquired for the correlation method (equation (5) described later), and this is applied to equation (3) to calculate the flow velocity V. These calculations are performed by the controller 10.

In the embodiment, the propagation time difference Δt in the above equation (3) is obtained by a correlation method (cross-correlation calculation). For example, the cross-correlation function R _xy (t) is a function of the shift amount τ when the y (t) waveform of one of the two signals x (t) and y (t) is delayed by τ. Is defined as

When this is applied to the propagation waveform of the ultrasonic signal, as shown in FIG. 3, the entire waveform in a certain time range of the signal propagating in the forward direction shown in FIG. 3A (corresponding to x (t) in Equation 1). ) And the entire waveform (corresponding to y (t) in _Equation 1) of the signal propagating in the reverse direction shown in FIG. 5B (corresponding to R _xy (t) in _Equation 1) ) Is calculated. FIG. 6C is an enlarged view of the rising portion of the ultrasonic signal propagating in the forward direction indicated by the dotted line portion in FIG. FIG. 4D is an enlarged view of the rising portion of the ultrasonic signal propagating in the reverse direction indicated by the dotted line in FIG.

  FIG. (E) shows the change of the correlation value of the cross-correlation function of both signal waveforms (correlation value graph), and shows the cross-correlation function corresponding to the rise of both signals in an enlarged manner. In FIG. (E), the portion from the time difference 0.0 to the peak portion (time difference 0.5 [μS]) of the correlation function (cosine waveform) corresponds to the propagation time difference Δt. Thus, by using the correlation method, the propagation time difference Δt between the two propagation signals can be obtained directly from the correlation function.

  Since the correlation method correlates the entire waveform existing in a certain time width, generally, even if the waveform is disturbed, the influence on the entire waveform is small, and errors due to bubbles and the like are small. This means that the signal waveform data used for the correlation method may be sampling data with a relatively low time resolution. In the embodiment, a commercially available A / D converter is used.

(thermometer)
FIG. 4 shows an example of the relationship between sound speed and temperature in water. As is clear from the above equation (3), in order to obtain the flow velocity V, it is necessary to know the sound velocity C in the fluid in advance. The speed of sound C varies depending on the type and composition of fluid, temperature, pressure, and the like. Depending on the measurement application, the sound speed C may be considered constant, but the sound speed C may have to be changed.
For example, consider a flow meter for an air conditioner that performs heating and cooling using hot water or cold water. It is assumed that the fluid is water and the change in sound velocity C due to pressure is outside the required accuracy range. Under such conditions, the speed of sound C is a function of temperature.

  Therefore, if the relationship between the temperature and the sound speed shown in FIG. 4 is stored in the controller 10 in advance, the sound speed can be obtained from the temperature of the fluid. For example, the speed of sound can be obtained from the fluid temperature measured by placing a temperature sensor (not shown) in the flow path close to the flow meter.

However, the flow meter and the temperature sensor may have to be separated from each other due to the degree of freedom of the system (heating, plant, etc.) using the fluid and the replacement of parts. Further, the water may become dirty with rust or the like, or antifreeze may be mixed in, and the relationship between temperature and sound velocity may deviate from the curve shown in FIG. In this case, it is necessary to calculate the speed of sound by measuring the amount of dirt and antifreeze mixed, but such calculation is generally difficult because the types of dirt and antifreeze are not necessarily the same.
Therefore, in the embodiment, the speed of sound C in the fluid is obtained by calculation from the signal waveform data used in the previous correlation method without using temperature measurement (thermometer).

(Sound velocity calculation)
In the propagation time difference method described above, when the expression (1) of the propagation time in the forward direction and the expression (2) of the propagation time in the reverse direction are combined so as to cancel Vcosθ, the sound speed C is expressed by the following expression.
C = (L / 2) {(1 / t ₁ ) + (1 / t ₂ )} Equation (5)
In the embodiment, the speed of sound C is obtained by equation (5), and this sound velocity C is applied to equation (3) of the flow velocity.

(Calculation of propagation time from received signal waveform)
In the embodiment of the present invention, the sound velocity C is obtained by the above equation (5), and the fluid flow velocity is obtained by the equation (3). In the correlation method described above, the propagation times t ₁ and t ₂ used in the equation (5) are not obtained. Therefore, it is considered to determine the propagation times t ₁ and t ₂ from the ultrasonic signal data group (signal waveform) stored in the memory for the correlation method. The signal data can be linearly compensated or curve-compensated by the MPU if necessary, and the signal data can be increased to smooth the waveform.

(Zero cross method)
FIG. 5 is an explanatory diagram for explaining the zero cross method. FIG. 2A shows the received waveform of the ultrasonic signal in the forward direction of the flow path. FIG. 5B shows the reception waveform of the ultrasonic signal in the reverse direction. In each figure, the horizontal axis represents the time axis, and the vertical axis represents the signal (voltage) level.
As shown in FIG. 5A, for example, when counting the number of local maximum points on the + side of the ultrasonic signal waveform and paying attention to the fourth zero cross point (characteristic point) immediately after, the time t from the waveform to the zero cross point is shown. ₁ 'is found. There is a difference between the propagation time t ₁ and the zero crossing time t ₁ ′ by a fixed value d ₁ determined by the ultrasonic driving frequency. Similarly, the zero crossing time t ₂ ′ is obtained from the waveform for the signal propagation of the ultrasonic signal in the reverse direction shown in FIG. There is a difference between the propagation time t ₂ and the zero crossing time t ₂ ′ by a fixed value d ₂ determined from the ultrasonic driving frequency.

D ₁ and d ₂ are stored in the controller (or flow meter device) in advance at the time of factory shipment. At the time of measurement, the characteristic point is detected from the received waveform by the MPU, and the forward propagation time t ₁ can be obtained by subtracting d ₁ from the time t ₁ ′. Further, the backward propagation time t ₂ can be obtained by subtracting d ₂ from t ₂ ′.

(Maximum amplitude method)
FIG. 6 illustrates an example focusing on the maximum amplitude of the ultrasonic signal waveform. The figure shows only the waveform of the ultrasonic signal in forward propagation. Since the waveform of the ultrasonic signal in the backward propagation is the same, the illustration is omitted.
In general, since the shape of the waveform to the maximum amplitude of the ultrasonic signal it does not vary with temperature and flow rate, the time t ₁ a signal waveform in the forward direction takes the maximum value 'is a fixed value d ₁ to the propagation time t ₁ of the ultrasonic It is time to add. Similarly, the time t ₂ ′ in which the signal waveform in the reverse direction takes the maximum value is a time obtained by adding the fixed value d ₂ to the ultrasonic wave propagation time t ₂ .

The controller stores d ₁ and d ₂ in advance at the time of factory shipment. At the time of measurement, the characteristic point (maximum value) is detected from the received waveform by the MPU, and the forward propagation time t ₁ can be obtained by subtracting d ₁ from the time t ₁ ′. Further, the backward propagation time t ₂ can be obtained by subtracting d ₂ from the time t ₂ ′.

(Combination of threshold and zero cross method)
FIG. 7 shows an example (forward propagation waveform) focusing on the first zero-cross point after the waveform of the ultrasonic signal exceeds a predetermined threshold. In the example of FIG. 5 described above, attention is paid to “a time during which a signal waveform passes through a zero cross point after a specific wave number has elapsed”. In this example, attention is paid to “a time passing through the zero cross point immediately after the amplitude of the waveform exceeds the threshold value, which is a value 0.8 Max obtained by multiplying the maximum value Max of the waveform by a fixed value (for example, 0.8)”. doing. In the embodiment, since the entire waveform of the received signal is stored in the memory, the threshold value can be set based on the maximum amplitude value in the entire vibration waveform having a plurality of peaks. The percentage of the maximum amplitude value to be set may be determined as appropriate by observing the ultrasonic signal waveform, and is not limited to “0.8”. Setting the threshold based on the maximum value of the entire waveform makes it easier to distinguish from noise. In addition, even when the waveform pattern of the ultrasonic signal partially changes (attenuates) due to bubbles, foreign matter, disturbances, or the like in the fluid, it is possible to use a determination criterion with fewer errors.

Also in this embodiment, the time differences d ₁ and d ₂ are previously stored in the controller at the time of factory shipment. At the time of measurement, a characteristic point (zero cross point) is detected from the received waveform by the MPU, and the forward propagation time t ₁ can be obtained by subtracting d ₁ from time t ₁ ′. Similarly, the backward propagation time t ₂ can be obtained by subtracting d ₂ from the time t ₂ ′ (not shown).

(Measurement procedure)
Next, the control operation of the controller (MPU) in the ultrasonic current meter (flow meter) shown in FIG. 1 will be described with reference to FIG.
First, when an event of a measurement start command occurs due to an operation of a switch (not shown) or a periodic output of a timer, the MPU built in the controller 10 stores a control program for this control routine (subroutine) stored in the storage device in advance. Is executed (step S10).

  The MPU supplies a switching signal to the switching devices 22 and 41 in order to receive a forward ultrasonic signal. The switch 22 is caused to select the transmitter / receiver 32 so that the transmitter / receiver 32 functions as an ultrasonic signal transmitter. Further, the transmitter / receiver 33 is selected by the switch 41 so that the transmitter / receiver 33 functions as a receiver. The MPU outputs a transmission signal from the interface of the controller 10 and amplifies the power by the amplifier 21 to drive the transceiver 32. Thereby, an ultrasonic signal is transmitted from the transmitter / receiver 32 toward the transmitter / receiver 33. The MPU supplies a clock signal to the A / D converter 43 and the memory 44 simultaneously with the transmission of the transmission signal to operate these parts. The received signal (ultrasonic signal) received by the transceiver 32 is amplified by the amplifier 42 and sampled by the A / D converter 43. For example, the entire waveform of the received signal as shown in FIG. Is stored as data (step S12).

  Similarly, the MPU supplies a switching signal to the switching devices 22 and 41 in order to receive an ultrasonic signal in the reverse direction. The transmitter / receiver 33 is selected by the switch 22 so that the transmitter / receiver 33 functions as an ultrasonic signal transmitter. In addition, the transmitter / receiver 32 is selected by the switch 41 so that the transmitter / receiver 32 functions as a receiver. The MPU outputs a transmission signal from the interface of the controller 10 and amplifies the power by the amplifier 21 to drive the transceiver 33. Thereby, an ultrasonic signal is transmitted from the transceiver 33 toward the transceiver 32. The MPU operates by sending a clock signal to the A / D converter 43 and the memory 44 simultaneously with sending the transmission signal. The received signal (ultrasonic signal) received by the transmitter / receiver 32 is amplified by the amplifier 42 and sampled by the A / D converter 43. For example, the entire waveform of the ultrasonic signal as shown in FIG. (Step S14).

  The MPU calculates a cross-correlation function between the entire waveform data of the forward ultrasonic signal and the entire waveform data of the reverse ultrasonic signal stored in the memory 44 by the correlation method to obtain the propagation time difference Δt (step). S16).

The MPU extracts the time t ₁ ′ of the feature point such as the zero cross point from the waveform data of the ultrasonic signal in the forward direction stored in the memory 44, and subtracts d ₁ from this to reduce the propagation time t _1. Calculate Further, as described above, the time t ₂ ′ of the feature point such as the zero cross point is extracted from the waveform data of the measured ultrasonic signal in the opposite direction, and the propagation time t ₂ is calculated by subtracting d ₁ therefrom. Any of the above-described methods can be used for extracting feature points. In terms of noise handling and the like, “threshold value related to peak value + zero cross method” (FIG. 7) is desirable (step S18).

The MPU calculates the speed of sound C from the known distance L between sensors, the propagation time t ₁ , and the propagation time t _{2 according} to the above equation (5) (step S20).

  The MPU calculates the flow velocity V of the fluid from the calculated sound speed C, the known distance L, and the propagation time difference Δt according to the above equation (3) (step S22).

  The MPU calculates the flow rate Q from the obtained flow velocity V, coefficient k, and cross-sectional area S of the pipe according to the above-described equation (4) (step S24).

The MPU displays the calculated flow velocity V and flow rate Q on the display 11. Alternatively, the measured flow velocity V and flow rate Q are sent to a process controller (computer) or the like via a network (not shown) (step S26).
In this way, the MPU calculates and outputs the flow velocity and flow rate every time an event occurs.

(Explanation of effect of embodiment)
According to the above-described embodiment, the propagation speeds t ₁ and t ₂ are obtained from the ultrasonic signal waveforms stored in the memory, the sound velocity C is calculated, and the flow velocity V is calculated by applying the equation (3) of the propagation time difference method. For this reason, there is an advantage that a thermometer for obtaining the sound velocity C, a conversion map (FIG. 4), correction according to the material of the fluid, and the like are not required.

(Comparative example)
In the ultrasonic flowmeters (comparative examples) described in Japanese Patent Laid-Open Nos. 10-332452 and 2001-141537, the flux is obtained by the “reciprocal difference in propagation time” different from the present application. The inverse propagation time difference method calculates the flow velocity by the following formula. The meaning of the symbols in the formula is the same as in the formula (3).
V = (L / 2 cos θ) {(1 / t ₁ ) − (1 / t ₂ )} Equation (6)
When calculating by the propagation time reciprocal difference method, it is not necessary to calculate the sound velocity C as in this embodiment. The propagation times t ₁ and t ₂ are obtained, and the flow velocity V is obtained from the difference between these reciprocals.
However, the time accuracy required at this time is generally on the order of sub-nanoseconds (for example, about 0.3 nsec and an A / D converter cannot be used), and is characterized by using a specially configured analog detection circuit. Propagation times t ₁ and t ₂ are obtained by directly detecting the times t ₁ ′ and t ₂ ′ at the points and correcting them. In other words, in the comparative example, the main part of signal detection is configured by an analog circuit, a gate circuit, or the like, and is not configured to process received signal data digitally.

  On the other hand, in the correlation method of this embodiment, the correlation is evaluated for the entire signal waveforms of the two signals, so the sampling interval of the received signal is several tens of nS (for example, 20 nS), and the signal is output by the A / D converter. Data can be obtained and digitally processed. This is because the time resolution required for sampling is two orders of magnitude larger (sampling interval is longer) than the time resolution required for the “reciprocal propagation time difference method” in the correlation method.

  The present invention has an advantage that the flow velocity V and the sound velocity C can be calculated with sufficient accuracy by utilizing the waveform data of the ultrasonic signal sampled in order to obtain the propagation time difference Δt by the correlation calculation as compared with the comparative example.

Next, the sound speed is calculated by using the signal data acquired for the correlation method by the detection method according to the present invention. It will be described that the sound speed can be detected with sufficient accuracy even with the acquired signal data.
(Time accuracy when calculating flow velocity)
First, an anemometer with the following specifications is assumed.
“Ultrasonic frequency: 1 [MHz], propagation time difference at maximum flow velocity: 1 [μS] (one cycle of ultrasonic signal), minimum flow velocity: 1/100 of maximum flow velocity, tolerance: ± 3% RD”
Since the propagation time difference is proportional to the flow velocity (see Equation (3)), the propagation time difference at the minimum flow velocity is 10 nS, which is 1/100 of the propagation time difference at the maximum flow velocity (1 [μS]). Therefore, the error of ± 3% RD at the minimum flow rate is ± 0.3 nS in the propagation time difference. As in this example, the sub-nS accuracy is generally required for the propagation time difference. Even when using the propagation time reciprocal difference method, the measurement of propagation time requires an accuracy of about sub-nS.

(Time accuracy when calculating sound speed)
Differentiating equation (3) with the speed of sound C,
(DV / dC) = (C / Lcos θ) ΔT Equation (7)
It becomes. Note that ΔT is separately measured by the correlation method.
The change in flow velocity δV relative to the change in sound velocity δC is δV = (C / Lcosθ) ΔTδC
It is expressed. Dividing this equation by equation (3) gives
(ΔV / V) = 2 (δC / C) Equation (8)
It becomes.

Equation (5) for calculating the temperature is a function of the _two propagation times t ₁ and t ₂ , and attention is paid to the relationship between t ₁ and the speed of sound with t ₂ fixed.
When the equation (5) is differentiated with respect to t ₁ , (dC / dt ₁ ) = − (L / 2) (1 / t ₁ ² ), so the change in sound velocity δC with respect to the change in propagation time δt ₁ is
δC = − (L / 2) (1 / t ₁ ² ) δt ₁
It is. When the sonic velocity C is sufficiently faster than the flow velocity V, t ₁ and t ₂ are almost equal, and the equation (5) becomes C≈ (L / t ₁ ). Therefore, the above equation is divided by this (δC / C) = − (1/2) (δt ₁ / t ₁ ) Equation (9)

From Equation (8) and Equation (9), δV / V = − (δt ₁ / t ₁ ) or | δV / V | = | δt ₁ / t ₁ | Equation (10)
It becomes.

That is, with respect to “when calculating the speed of sound”, the rate of change in flow velocity and the rate of change in propagation time (absolute value) are equal. When t ₁ is fixed and t ₂ is changed, the same result can be obtained only by changing the sign.

Assuming that the accuracy of the flow velocity is ± 3% RD (allowable error) under the same assumption as “when calculating the flow velocity”,
| δV / V | = | δt ₁ / t ₁ | = 3/100.

For example, if the ultrasonic wave propagation path L is 10 [cm] and the sound velocity C is 1500 [m / S], the propagation time is 67 [μS]. At this time, the change in propagation time corresponding to the flow rate accuracy of 3% RD is | δt ₁ | = (3/100) × 67 [μS] = 2 [μS].

  That is, in order to obtain a flow velocity accuracy of ± 3% RD, it is sufficient that the variation in propagation time is within 2 [μS] with respect to the sound velocity. This is more than 6000 times the allowable variation required when calculating the flow velocity. Further, it can be said that sampling of 20 [nS] (= 0.02 [μS]) is sufficient for the time resolution.

As described above, according to the embodiment of the present invention, in the flow velocity (or flow rate) measurement method by the propagation time difference method (correlation method) in which the transmitter / receiver is arranged in the flow channel, in the forward direction and the reverse direction of the flow channel. The signal data of the entire received signal waveform is stored in the memory, the propagation time difference Δt is obtained from the cross-correlation calculation of the signal data of the two received signals, and the feature points are extracted from the waveform of the received signal signal data stored in the memory. Then, the propagation times t ₁ and t ₂ are obtained, the sound velocity C is obtained from the propagation times t ₁ and t ₂ , and the flow velocity V (flow rate Q) is calculated from the propagation time difference Δt and the sound velocity C by the propagation time difference method (3). Since it is configured to perform, it is possible to calculate the flow velocity and flow rate without using a thermometer.

  It should be noted that the examples described through the above-described embodiments of the present invention can be used in appropriate combination depending on the application, or can be used with modifications or improvements, and the present invention is limited to the description of the above-described embodiments. It is not a thing. The form which added such a combination or a change or improvement can also be contained in the technical scope of this invention.

  The present invention is advantageously used for measurement of fluid flow velocity, measurement of fluid flow rate, ultrasonic flowmeter, ultrasonic flowmeter and the like. Further, the present invention can be applied to a device or a control system using a velocimeter or a flow meter.

10 Controller 11 Display 20 Transmission System 21 Amplifier 22 Signal Switcher 31 Piping (Flow Path)
32 Transmitter / Receiver 33 Transmitter / Receiver 41 Signal Switcher 42 Amplifier 43 A / D Converter 44 Memory

Claims (7)

A flow velocity measurement method for obtaining a flow velocity of a fluid in a flow path using a correlation method,
Acquiring waveform data of a first ultrasonic signal propagating forward between two points of the flow path;
Acquiring waveform data of a second ultrasonic signal propagating in a reverse direction between two points of the flow path;
Obtaining a propagation time difference from a correlation function between the waveform data of the first and second ultrasonic signals;
Extracting a feature point of the waveform from the waveform data of the first ultrasonic signal, and obtaining a forward propagation time of the ultrasonic signal between the two points of the flow path;
Extracting a feature point of the waveform from the waveform data of the second ultrasonic signal, and obtaining a backward propagation time of the ultrasonic signal between the two points of the flow path;
Calculating the speed of sound of the ultrasonic signal based on the distance between two points of the flow path, the forward propagation time and the reverse propagation time of the ultrasonic signal;
Calculating the flow velocity of the fluid based on the distance between the two points of the flow path, the propagation time difference, and the speed of sound;
Flow velocity measurement method including The process of calculating the sound speed is as follows:
Assuming that the speed of sound is C, the distance between the two points is L, the forward propagation time is t ₁ , and the backward propagation time t ₂ ,
C = (L / 2) ((1 / t ₁ ) + (1 / t ₂ ))
Calculate the speed of sound C by calculating
The method of measuring a flow velocity according to claim 1. The process of calculating the fluid flow velocity is as follows:
V is the flow velocity of the fluid, C is the speed of sound, Δt is the propagation time difference, and θ is the angle between the flow direction of the fluid and the propagation direction of the ultrasonic signal,
V≈ (C ² / 2L cos θ) Δt
To obtain the flow velocity V by calculating
The method for measuring a flow velocity according to claim 1 or 2, wherein: The process of obtaining the forward propagation time and the backward propagation time,
4. A feature point of a waveform is discriminated from the acquired waveform data of the ultrasonic signal, and a predetermined value is corrected from a position on the time axis of the feature point. The flow velocity measuring method according to any one of the above.   A flow rate measurement method for obtaining a flow rate by multiplying a flow velocity obtained by the flow velocity calculation method according to claim 1 by a cross-sectional area of the flow path. An ultrasonic anemometer that determines a flow velocity of a fluid in a flow path using a correlation method,
Means for acquiring waveform data of a first ultrasonic signal propagating in a forward direction between two points of the flow path;
Means for acquiring waveform data of a second ultrasonic signal propagating in a reverse direction between two points of the flow path;
Storage means for storing waveform data of the first and second ultrasonic signals;
Means for obtaining a propagation time difference from a correlation function between waveform data of the first and second ultrasonic signals;
Means for extracting a characteristic point of the waveform from the waveform data of the first ultrasonic signal, and obtaining a forward propagation time of the ultrasonic signal between the two points of the flow path;
Means for extracting a characteristic point of the waveform from the waveform data of the second ultrasonic signal, and obtaining a backward propagation time of the ultrasonic signal between the two points of the flow path;
Means for calculating the speed of sound of the ultrasonic signal based on a distance between two points of the flow path, a forward propagation time and a backward propagation time of the ultrasonic signal;
Means for calculating the flow velocity of the fluid based on the distance between the two points of the flow path, the propagation time difference, and the speed of sound;
An anemometer. An ultrasonic flowmeter that uses a correlation method to determine the flow velocity of a fluid in a flow path,
Means for acquiring waveform data of a first ultrasonic signal propagating in a forward direction between two points of the flow path;
Means for acquiring waveform data of a second ultrasonic signal propagating in a reverse direction between two points of the flow path;
Storage means for storing waveform data of the first and second ultrasonic signals;
Means for obtaining a propagation time difference from a correlation function between waveform data of the first and second ultrasonic signals;
Means for extracting a characteristic point of the waveform from the waveform data of the first ultrasonic signal, and obtaining a forward propagation time of the ultrasonic signal between the two points of the flow path;
Means for extracting a characteristic point of the waveform from the waveform data of the second ultrasonic signal, and obtaining a backward propagation time of the ultrasonic signal between the two points of the flow path;
Means for calculating the speed of sound of the ultrasonic signal based on a distance between two points of the flow path, a forward propagation time and a backward propagation time of the ultrasonic signal;
Means for calculating the flow velocity of the fluid based on the distance between the two points of the flow path, the propagation time difference, and the speed of sound;
Means for obtaining the flow rate by multiplying the obtained flow velocity by the cross-sectional area of the flow path;
A flow meter comprising.