WO2006080182A1

WO2006080182A1 - Ultrasonic flowmeter and ultrasonic flowmeter employing two methods

Info

Publication number: WO2006080182A1
Application number: PCT/JP2006/300110
Authority: WO
Inventors: Sosuke Suzuki; Toshihiro Yamamoto; Yoshinori Ohmuro
Original assignee: Fuji Electric Systems Co., Ltd.; Yao, Hironobu
Priority date: 2005-01-31
Filing date: 2006-01-06
Publication date: 2006-08-03
Also published as: JP4548482B2; JPWO2006080182A1

Abstract

A flow velocity distribution measurement unit (11) acquires a flow velocity distribution along a measurement line according to a measurement result obtained by a sensor (1). According to a Doppler shift frequency obtained by the processing of the flow velocity distribution measurement unit (11), its average value, a Doppler spectrum, and the like, a power/standard deviation calculation unit (12) calculates a power and standard deviation for each channel and judges the state of the flowing body to be measured, according to the calculation result.

Description

Specification

Ultrasonic flow meter, 2 types combined ultrasonic flow meter

Technical field

[0001] The present invention relates to an ultrasonic flowmeter.

Background art

Conventionally, there is a clamp-on type ultrasonic flowmeter as a flow measurement technique using ultrasonic waves.

In this clamp-on type ultrasonic flow meter, a material for transmitting sound waves to the pipe, that is, a wedge, is installed on a part of the outer peripheral surface of the pipe through which the fluid to be measured passes. The flowmeter measures the flow rate of the fluid to be measured by transmitting sound waves into the pipe through the wedge. As a clamp-on type ultrasonic flowmeter, flow measurement technology using the pulse Doppler method (hereinafter referred to as pulse Doppler method) and flow measurement technology using the propagation time difference method (hereinafter referred to as propagation time difference method) are known. Yes.

[0003] The pulse Doppler method measures the flow rate of a fluid from the moving speed of suspended particles, assuming that suspended particles and bubbles contained in the fluid move at the same speed as the fluid. As described in Ref. 1, non-stationary fluids can be measured with high accuracy without contact. The flow rate measurement technology described in Patent Document 1 transmits ultrasonic pulses (group) to the fluid to be measured in the pipe at regular intervals, and is caused by foreign matters such as bubbles (reflectors on the measurement line) mixed in the fluid. When the frequency of the reflected ultrasonic echo wave changes by a magnitude proportional to the flow velocity, the principle of Doppler shift is applied. That is, the Doppler shift is calculated based on the echo wave, the flow velocity distribution of the fluid to be measured is obtained, and the flow rate is calculated by integration calculation based on the flow velocity distribution, thereby obtaining the flow rate of the fluid to be measured. .

[0004] The pulse Doppler method is more accurate and capable of high-speed response than the propagation time difference method, and is excellent in bubble resistance. Furthermore, by providing multiple measurement lines, high-precision measurement is possible even when there is a drift. There is a feature that makes possible. However, the pulse Doppler method enables measurement because a certain amount of reflector exists in the fluid to be measured to some extent, so that there is not enough reflector or the reflector is biased. Highly accurate measurement is performed on fluid There is a problem!

[0005] On the other hand, the propagation time difference method uses one or more pairs of transmission / reception integrated transducers to compare the ultrasonic propagation time from the upstream side to the downstream side and the ultrasonic propagation time from the downstream side to the upstream side. Then, by using the difference, the average flow velocity and flow rate of the non-measurement fluid are calculated. Compared with the pulse Doppler method, this method is suitable for measuring the flow rate of liquids with less impurities and pure water. Conversely, with the propagation time difference method, high-precision measurement may be difficult if there are many impurities such as bubbles in the fluid to be measured.

[0006] In conventional ultrasonic flowmeters, measurement is performed by either the pulse Doppler method or the propagation time difference method.

Conventionally, for example, an invention described in Patent Document 2 has been proposed.

[0007] The invention described in Patent Document 2 includes a time difference type ultrasonic flow meter and a Doppler type ultrasonic flow meter, and the time difference type ultrasonic flow meter is less than the threshold value below a predetermined flow rate threshold value. Above, it switches to a Doppler type ultrasonic flowmeter.

[0008] However, the Doppler method in Patent Document 2 is different from the pulse Doppler method. That is, the Doppler method in Patent Document 2 is a method generally called a continuous wave method, and the pulse Doppler method is a pulse wave method. The Doppler method (continuous wave method) in Patent Document 2 is described in, for example, Non-Patent Document 1 and the like. However, if briefly explained, the velocity of the fluid to be measured at one point on the central axis of the pipe is described. This is a method of calculating this fluid velocity force flow rate by using a Doppler shift. On the other hand, the pulse Doppler method is a method of calculating the flow rate using the obtained flow velocity distribution by obtaining the flow velocity at specific positions (plural) in the pipe. Therefore, Patent Document 2 has a description that the Doppler method has a problem that the accuracy is lower than that of the time difference method. On the other hand, the pulse Doppler method has higher measurement accuracy than the propagation time difference method as described above.

[0009] As described above, the flow measurement technique using the pulse Doppler method has many advantages over the propagation time difference method, but it depends on the state of the fluid to be measured (the amount of the reflector and the deviation, etc.). For this reason, it may be better to use the propagation time difference method. Here, the state of the fluid to be measured is not necessarily constant at a certain measurement point (installation point of the ultrasonic flowmeter). 匕Therefore, even if there is a certain amount of reflector in the measured fluid at a certain point in time, there is not enough reflector at another point! There may be a situation where ヽ or Z and reflectors are present in a biased manner.

[0010] Therefore, it is conceivable to install an ultrasonic flow meter in which the pulse Doppler method and the propagation time difference method are mixed for each measurement point, and switch between both in accordance with the state of the fluid to be measured.

However, conventionally, it has been difficult to automatically determine the state of the fluid to be measured by the apparatus. Conventionally, the flow velocity distribution obtained during the processing for the flow rate calculation is displayed, or the received signal of the reflected ultrasonic echo wave is displayed, so that a human can see the displayed contents. The state of the fluid to be measured was judged empirically.

[0011] As described above, it is desirable to automatically determine the state of the fluid to be measured by the apparatus. Then, it is desired that the flow rate measurement can be performed with higher accuracy by switching between the pulse Doppler method and the propagation time difference method based on the determination result.

On the other hand, in the method described in Patent Document 2, the state of the fluid to be measured such as the amount and uniformity of the reflector is merely changed to V based on the flow rate. Therefore, it can not be said that the criteria for switching is not adequately appropriate. Further, as described above, Patent Document 2 is not related to the pulse Doppler method in the first place. By using the Nords-Doppler method, it is possible to measure with higher accuracy even in the unsteady state.

Patent Document 1: JP 2000-97742 A

Patent Document 2: Japanese Unexamined Patent Application Publication No. 2004-184245

Non-patent document 1: “Flow measurement for instrumentation engineers A to Zj pages 128-131, author: Japan Metrology Equipment Industry Association, publisher: Industrial Technology Co., Ltd., November 1, 1995, first edition Disclosure of the invention

[0013] An object of the present invention is to determine the state of the fluid to be measured in a pulse Doppler type ultrasonic flowmeter so that the accuracy of the measurement result can be known, and further, a difference in propagation time from the pulse Doppler method. Constructing an ultrasonic flow meter that can be used in combination with the method, and switching the pulse Doppler method to the propagation time difference method according to the result of determining the state of the fluid to be measured. Deteriorating measurement accuracy It is to provide an ultrasonic flowmeter that can measure the flow without using it, and a two-type combined ultrasonic flowmeter.

The pulse Doppler ultrasonic flowmeter of the present invention calculates, for each channel, power indicating the intensity of the echo signal based on the Doppler spectrum obtained based on the measured echo signal. Power calculation means, standard deviation calculation means for calculating the standard deviation based on the Doppler shift frequency obtained based on the measured echo signal and the average value for each channel, and for each channel The power or Z and the standard deviation are respectively compared with a predetermined threshold value, so that a state determining means for determining the state of the fluid to be measured is provided.

[0015] The calculated power indicates the state of the reflector included in the fluid to be measured (proportional to the number, etc.), and the standard deviation indicates the flow stability. For example, when the power is larger than the threshold value and the standard deviation is smaller than the threshold value, the state of the fluid to be measured can be regarded as a flow containing a certain amount of reflectors and a uniform flow to some extent. In other words, it can be considered that it is in a state where high-precision measurement can be performed with the pulse doppler one-way system. In this way, since the state of the fluid to be measured can be automatically discriminated, for example, if the discrimination result is displayed, the supervisor or the like can know the accuracy of the measurement result.

[0016] Furthermore, by applying the method for determining the state of the fluid to be measured to an ultrasonic flowmeter that uses a combination of the pulse Doppler method and the propagation time difference method, the two measurement methods can be switched when appropriate. it can.

That is, the two-system combined ultrasonic flowmeter of the present invention is a combined ultrasonic flowmeter of the pulse Doppler method and the propagation time difference method, and normally uses the pulse Doppler method. Power calculation means for calculating the power indicating the intensity of the echo signal based on the Doppler spectrum obtained based on the measured echo signal for each channel, and the measured echo signal for each channel. The standard deviation calculation means for calculating the standard deviation based on the Doppler shift frequency obtained based on the average and the average value thereof, and for each channel, the power or Z and the standard deviation are respectively compared with a predetermined threshold value. Therefore, it is determined whether to switch to the propagation time difference method temporarily according to the determination result by the state determination unit and the determination result by the state determination unit. Switch And a step.

[0018] Since the state determination means can determine whether or not the force is in a state where high-precision measurement can be performed by the pulse Doppler method, if it is determined that the measurement accuracy is poor, it is temporarily propagated. By switching to the time difference method, it is possible to prevent a significant deterioration in measurement accuracy.

[0019] Then, for example, the state determining unit determines whether or not the measurement accuracy is deteriorated for each channel, and the switching unit has the measurement accuracy determined by the state determining unit. If the number of channels determined to have deteriorated exceeds a preset ratio, the measurement method is switched to the propagation time difference method.

[0020] Further, for example, the power is a total power obtained by integrating the Doppler spectrum or an averaged power obtained by dividing the total power by the number of acquisitions.

Further, for example, instead of the standard deviation calculated based on the Doppler shift frequency and the average value thereof, a value obtained by standardizing the calculated standard deviation with the average value of the Doppler shift frequency is used. It is.

[0021] Also, for example, when an arbitrary number of measurements or measurement time is set in advance and the switching means is temporarily switched to the propagation time difference method, the measurement by the propagation time difference method is performed by the set The measurement may be performed for the number of measurement times or the measurement time.

[0022] Further, for example, when the switching means is temporarily switched to the propagation time difference method, the measurement result by the propagation time difference method is corrected so that the measurement result force by the pulse Doppler method immediately before the switching changes gently. And then output it.

[0023] According to the ultrasonic flowmeter of the present invention, the two-type combined ultrasonic flowmeter, etc., the pulse doppler one-type ultrasonic flowmeter is used to determine the state of the fluid to be measured. In addition, an ultrasonic flow meter that combines the pulse Doppler method and the propagation time difference method is configured so that the pulse Doppler method can be propagated according to the determination result of the state of the fluid to be measured. By switching to the time difference method, it is possible to measure the flow rate without degrading the measurement accuracy while energizing high-precision measurement by the pulse Doppler method. In addition, since the discrimination accuracy is high, switching is performed at an appropriate time.

Brief Description of Drawings

FIG. 1 is a functional block diagram of a configuration of a pulse Doppler type ultrasonic flowmeter according to this example. FIG. 2A is a diagram showing a criterion for determining a state of a fluid to be measured based on an experimental result.

FIG. 2B is a diagram showing criteria for determining the state of the fluid to be measured based on the experimental results.

FIG. 2C is a diagram showing criteria for determining the state of the fluid to be measured based on the experimental results.

FIG. 3 is a functional block diagram of a configuration of a two-method combined ultrasonic flowmeter according to this example.

FIG. 4 is a flowchart of the flow rate measurement process of this example.

FIG. 5 is a modification of the process of FIG.

FIG. 6 is a diagram for explaining processing for discrimination using a success rate.

FIG. 7A is a diagram for explaining a handling method when both measurement methods are frequently switched in a short time.

FIG. 7B is a diagram for explaining a response method when both measurement methods are frequently switched in a short time.

FIG. 7C is a diagram for explaining a handling method when both measurement methods are frequently switched in a short time.

FIG. 7D is a diagram for explaining a response method when both measurement methods are frequently switched in a short time.

[FIG. 8] This is a diagram showing a channel!

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a functional block diagram of a configuration of a pulse Doppler type ultrasonic flowmeter according to this example, which can determine the state of a fluid to be measured.

The illustrated pulse Doppler type ultrasonic flowmeter includes a sensor 1, a transmitter 2, an emitter 3, an amplifier 4, an AZD converter 5, a display device 6, and a CPUZMPUIO. The CPUZMPUIO executes a predetermined application program stored in its built-in memory or an external memory (not shown), thereby allowing the flow velocity distribution measurement unit 11, the power standard deviation calculation processing unit 12, and the flow rate calculation processing unit 13 to The processing of each functional unit is executed.

[0027] The sensor 1 is an ultrasonic transducer and is attached to a wedge 8 installed on a part of the outer peripheral surface of a pipe wall 7 of a pipe through which a fluid to be measured flows. The sensor 1 outputs an ultrasonic pulse based on the electrical signal generated by the transmitter 2 and the emitter 3. This ultrasonic filter Luth is a straight beam having a diameter of about 5 mm, for example, enters the fluid flowing through the pipe via the wedge 8 and the tube wall 7 and is reflected by the reflector 9 (such as bubbles) contained in the fluid. The reflected echo is received by the sensor 1 and converted into an electrical signal, and then output to the amplifier 4.

The transmitter 2 generates an electrical signal having a fundamental frequency f0, and the emitter 3 outputs the electrical signal from the oscillator 2 in a pulse shape at predetermined time intervals (lZFprf). Thus, the ultrasonic pulse having the fundamental frequency f0 is output from the sensor 1 at predetermined time intervals. The basic frequency f0 is basically a required frequency determined in inverse proportion to the inner diameter of the pipe. The sensor 1 is attached to the wedge 8 so as to be inclined at a certain angle with respect to the pipe. The ultrasonic pulse travels in the fluid along the measurement line shown in the figure. is there. In addition, the reflector 9 includes particles such as fine powder, foreign matter having an acoustic impedance different from that of the fluid to be measured, in addition to the bubbles.

[0029] The reflected echo electric signal output from the sensor 1 is amplified by the amplifier 4, further digitized by the AZD converter 5, and input to the CPUZMPUIO.

In the CPU / MPU 10, the flow velocity distribution measurement unit 11 and the flow rate calculation processing unit 13 are existing processing units. When the digital signal is input, the flow velocity distribution measurement unit 11 first measures the measurement. The flow rate calculation processing unit 13 calculates the flow rate distribution of the fluid, and the flow rate calculation processing unit 13 calculates the flow rate of the fluid to be measured based on this flow rate distribution. The apparatus further includes a power standard deviation calculation processing unit 12. RU

[0030] As described above, in the measurement by the pulse Doppler method, transmission / reception of the ultrasonic pulse is repeatedly performed a number of times with a predetermined repetition period (lZFprf).

The flow velocity distribution measurement unit 11 calculates the flow velocity at each position on the measurement line based on the measurement results obtained many times, and obtains the flow velocity distribution based on the calculated flow velocity at each position (measurement point). When calculating the flow velocity, the Doppler shift (difference between the transmission pulse frequency and the reception echo frequency) component is extracted, and the flow velocity is calculated based on the Doppler shift frequency fd.

[0031] Here, each position (measurement point) does not mean one point, but each measurement area (for example, an area corresponding to the pipe radius in the direction perpendicular to the pipe axis) is divided into areas (channels). Is called). Naturally, it is predicted which area the reflector 9 will pass. Force that is not connected Further, for each area, it is unpredictable where the reflector 9 flows in the area. In other words, the pulse Doppler method handles probabilistic events and it is difficult to obtain true values. Therefore, for example, when 256 times of flow velocity measurements are performed for each channel, the average value is calculated and used as the flow velocity at each position (measurement point). For each channel, the Doppler spectrum is calculated based on the 256 measurement results. For each measurement execution, the channel at which the measurement point (the position at which the reflector 9 was present) is determined based on the transmission time of the above ultrasonic pulse and the time taken to receive the reflected echo. it can.

[0032] First, the power standard deviation calculation processing unit 12 is based on data (Doppler shift frequency fd, Doppler spectrum, etc.) obtained in the flow rate calculation process by the flow velocity distribution measurement unit 11 described above. For example, Reference 1 ("Basic Ultrasound Medicine", Junichi Yagi, Nobuyuki Endo, Yasuo Hirata, Kei Ito; Ishiyaku Shuppan Co., Ltd .; October 10, 2002, 1st edition, 3rd edition, pages 26-31 Page 2), Reference 2 (`` Ultrasound Medical Dictionary '' by Satsuki Texto, Shujunsha, September 8, 2000, 1st edition, 1st edition, pages 170-175), etc. Thus, power (P) and standard deviation σ are calculated for each channel.

[0033] For each channel, for example, as described in References 1 and 2 above, first, a power spectrum with the Doppler shift frequency fd as the horizontal axis and the power as the vertical axis, that is, the Doppler spectrum P ( fd) is obtained. Then, using this Doppler spectrum P (fd), the power (P) in the channel is calculated by the following equation (1).

[0034] [Equation 1]

P = jP (fi) df _d (1) The above power (p) indicates the intensity of the echo wave, and as described in Reference 1 above, the reflection of the reflector in the fluid to be measured It is proportional to the number.

[0035] The average Doppler shift frequency fd is an average value of the Doppler shift frequency fd. From (first moment), variance σ ² (second moment), which is a statistic that specifies the degree of variation in Doppler shift frequency fd, can be expressed by the following equation (2).

[0036] [Equation 2]

^ ^2- (fj-) ² = f-() ² · · · (2) The square root () of the above variance is the standard deviation σ. This standard deviation indicates the variation in the time domain of the measurement results for each channel (flow stability) for each channel. If this variation is within a practical range, the measurement results are valid. it is conceivable that.

[0037] As described above, the power (Ρ) indicates the state (number, etc.) of the reflector included in the fluid to be measured, and the standard deviation indicates the flow stability. As shown in the graph, when the capacity is large and the standard deviation is large, the state of the fluid to be measured indicates that the reflector is included in a large amount and the flow is uneven. When the power is large and the standard deviation is small, the state of the fluid to be measured indicates that the reflector is contained in a certain amount and has a uniform flow. Also, when the power is small and the standard deviation is large, the state of the fluid to be measured indicates that the reflector is few (or nothing!) And the flow is uneven.

[0038] Here, the inventor of the present application uses a pulse Doppler type ultrasonic flowmeter, the conditions (state of the fluid to be measured) for performing high-accuracy flow measurement, which is a feature of the flowmeter, an experiment, etc. Based on the above. In other words, as shown in Fig. 2 (b), when the condition that the power is large and the standard deviation is small is satisfied, it was confirmed that the flow rate can be measured with high accuracy without any problems.

[0039] As a specific example, for example, as shown in Fig. 2 (b) by an experiment using a SUS50A pipe, the average flow velocity of the fluid to be measured is 0.4 (m / _S ) to 2 (m / s). In the case of s), a large variation in the measurement results can be confirmed at a mean flow velocity of 0.2 (m / s) where the measurement accuracy is good and the variation in the measurement results is small. Then, the inventor further calculated the power (P) and the standard deviation by sequentially changing the average flow velocity of the fluid to be measured, and obtained the experimental result shown in FIG. 2C. [0040] As shown in Fig. 2C, each flow velocity within the range of 0.4 (m / s) to 2 (m / s), which is an average flow velocity that can be measured with high measurement accuracy, is shown in Fig. 2C in Fig. 2A. When applied to classification, it almost corresponds to the condition of “high power and small standard deviation”. And in this experimental result, the threshold value that distinguishes between “large” and “small” is about 10 7 to the 8th power and the standard deviation is about 1 to 1.5 (rad). It can be judged that there is.

[0041] Here, in this experiment, the power and standard deviation at each flow velocity are obtained by changing the flow velocity, but this does not mean that normality / abnormality is determined by the flow velocity. By changing the flow velocity, the state of the fluid to be measured is changed to some force, so the flow velocity is used as a parameter. In other words, in this experiment, the number of reflectors 9 that pass through the measurement line per unit time increases as the force flow rate that causes bubbles to be included in the fluid so as to be substantially uniform in a substantially constant amount. Thus, the flow velocity suggests the number of reflectors in the fluid. In this experiment, bubbles are used as the reflector. Naturally, bubbles in the water have the property of rising, but they are swept away as soon as the flow velocity is high, so the variation is small. On the other hand, if the flow rate is slow, bubbles will rise in the middle and the bubbles will be biased upward in the direction of the pipe's strength, resulting in large variations.

[0042] As described above, the above experimental result only reproduces the state of the fluid to be measured with a flow velocity that does not mean that it can always be measured normally when the flow velocity is 2 m / s, for example. ,

The power (P) and standard deviation are calculated according to the number of reflectors and variations in the fluid.

In other words, the power standard deviation calculation processing unit 12 indicates the intensity of the echo signal based on the Doppler spectrum obtained based on the measured echo signal for each channel. A power calculation unit for calculating power, a standard deviation calculation unit for calculating the standard deviation based on the Doppler shift frequency obtained based on the measured echo signal and the average value for each channel, and each channel It can be said that it comprises a state discriminating unit (both not shown) for discriminating the state of the fluid to be measured by comparing the power or Z and the standard deviation with a predetermined threshold value.

As described above, the power (P) calculated by the power / standard deviation calculation processing unit 12 and the standard deviation are calculated. The difference is displayed on the display device 6, for example, and the operator or the like determines the current state of the fluid to be measured based on the threshold value or the like based on the above experimental results, and confirms that the measurement accuracy has deteriorated. can do. Alternatively, the threshold value and the like are set in advance in the power / standard deviation calculation processing unit 12 and, for example, if the condition “power is large and standard deviation is small” is not satisfied, the measurement accuracy deteriorates. May be displayed on the display device 6. That is, these display contents serve as an index for knowing the certainty of the measurement result.

[0045] By using power and standard deviation as described above, the state of the fluid to be measured can be accurately represented, so the above "index for knowing the accuracy of the measurement result" is highly reliable. . In addition, by setting an appropriate threshold based on the experimental results as described above, the reliability of determining whether or not the measurement accuracy is bad can be improved. The flow meter will be switched when appropriate.

That is, by applying the method for determining the state of the fluid to be measured to the two-method combined ultrasonic flowmeter shown in FIG. 3 described below, while taking advantage of the high measurement accuracy of the pulse Doppler method, When the measurement accuracy deteriorates, switching to the propagation time difference method can suppress the measurement accuracy.

FIG. 3 is a functional block diagram of a “configuration of a two-method combined ultrasonic flowmeter according to this example”.

The two systems are the pulse Doppler system and the propagation time difference system as described above. The two-type ultrasonic flowmeter shown in the figure is sensor 1, sensor 21, sensor 22, switch 23, amplifier 24, amplifier 25, switching unit 26, AZD converter 27, CPU / MPU 28, transmission circuit 29, display device 30. , And an input device 31.

The sensor 1 is an ultrasonic transducer for the pulse Doppler method shown in FIG. In FIG. 3, a pair of ultrasonic transducers for the propagation time difference method, that is, a sensor 21 and a sensor 22 are further provided. The sensor 21 and the sensor 22 themselves may be the same as the sensor 1 and can be shared. That is, sensor 1 may be deleted and sensor 21 used for the pulse Doppler method.

[0049] The transmission circuit 29 includes, for example, the transmitter 2 and the emitter 3 shown in FIG.

Switch 23 is a switch for switching between using sensor 1 and using sensors 21 and 22. When switch 23 is switched to the sensor 1 side, The output electrical signal is input to the sensor 1 and an ultrasonic pulse is emitted, and this is reflected by the reflector 9 and the like, so that the echo electrical signal output from the sensor 1 is input to the amplifier 24. become. On the other hand, when the switch 23 is switched to the sensor 21 or 22 side, the electrical signal output from the transmission circuit 29 is input to the sensor 21 or 22, and an ultrasonic pulse is emitted, and this is output to the sensor 22 or 22. The electric signal (received wave) received and output by 21 is input to the amplifier 25.

The amplifiers 24 and 25 are mere signal amplification circuits like the amplifier 4, the amplifier 24 amplifies the echo electric signal, and the amplifier 25 amplifies the received wave.

The switching unit 26 inputs the amplified output signal of either one of the amplifiers 24 and 25 to the AZD conversion 27. Naturally, when the sensor 1 is used, the switching unit 26 is switched to the amplifier 24 side. The AZD converter 27 performs AZD conversion on any of the above amplified output signals, and outputs the digital signal to the CPUZMPU 28.

[0051] Like CPUZMPUIO in Fig. 1, CPUZMPU28 has the function units of flow velocity distribution measurement unit 11, noise standard deviation calculation processing unit 12, and flow rate calculation processing unit 13. A switching control unit (not shown) that switches the switching unit 26 and the switch 23 to the propagation time difference method side when the deviation calculation processing unit 12 determines that the measurement accuracy is deteriorated as described above. . It also has a processing function unit (not shown) that performs flow rate calculation processing using the propagation time difference method! / Speak.

When the CPUZMPU 28 calculates the flow rate by any of the above methods, the flow rate is displayed on the display device 30. In addition, it is possible to display on the display device 30 which method is currently used.

FIG. 4 shows a flowchart of the flow rate measurement process executed by the CPUZMPU 28.

In this example, the flow rate measurement is normally performed by the pulse Doppler method, and only when it is determined that the measurement accuracy has deteriorated, the switching to the propagation time difference method is performed.

[0054] From this, normally, the switch 23 and the switching unit 26 are switched to the sensor 1 side, and when the digital signal of the echo wave detected by the sensor 1 is input from the AZD converter 27, the flow velocity distribution is The flow velocity is calculated by the measurement unit 11 (step S11). As described above, using the Doppler spectrum P (fd) or the like obtained in the process, the power standard deviation calculation processing unit 12 calculates the power (P) and the standard deviation σ (step S12).

[0055] Since the threshold value X for the power (Ρ) and the threshold value に対する for the standard deviation σ are set and registered in advance as described above, the power (Ρ) and the standard deviation calculated in step S12 are Compared with the threshold values X and 、, respectively, for example, when power (Ρ) ≥ Χ and standard deviation ≤ 、, that is, “the power is large and the standard deviation is small!” (Step S13, YES), the Nord Doppler method is continued as is, the flow rate calculation unit 13 calculates the flow rate (Step S14), and the calculation result is output to the display device 30. (Step S19).

[0056] On the other hand, if the condition of "power (P) ≥ X and standard deviation ≤ Y" is not satisfied (step S13, NO), switch to the propagation time difference method. That is, first, after switching the switch 23 and the switching unit 26, an ultrasonic pulse is transmitted from the sensor 21 and received by the sensor 22, and an ultrasonic pulse is transmitted from the sensor 22 and received by the sensor 21 (step S15). If both are received normally (step S16, YES), the average flow velocity and flow rate of the fluid to be measured are calculated based on the propagation time difference between the two (step S17), and the result is displayed on the display device 30. Output to. If at least one of them is unable to receive normally (step S16, NO), an abnormal process is performed (step S18).

[0057] In the above description, the processing of step S13 is based on the condition "power (P) ≥ X and standard deviation ≤ Y". However, the present invention is not limited to this example, and only "power" ≥ Χ is determined. The condition may be a condition, or only “standard deviation ≦ Υ” may be a determination condition. According to the experimental results shown in Fig. 2C, it is possible to discriminate to some extent using only the standard deviation. Also, the power is proportional to the number of reflectors in the fluid to be measured, and the pulse Doppler method has a small number of reflectors! This is because even if only power is used, it can be distinguished to some extent.

FIG. 5 shows a modification of the process of FIG.

The processing in steps S21 to S26 shown in FIG. 5 is almost the same as the processing in steps S11 to S19 in FIG. 4 (the processing in step S25 is the processing in steps S15 to S18 in FIG. 4). The difference is that the user can set arbitrary threshold values Xs and Ys at any time in advance (Steps The determination in step S27) and step S23 is to use the threshold values Xs and Ys instead of the threshold values X and Y in step S13. In other words, the manufacturer of the above two-method combined ultrasonic flowmeters sets the above threshold values X and Y based on the above experimental results and other experiences' intuition. It depends on etc. Therefore, for each user, the thresholds Xs and Ys can be set arbitrarily as deemed appropriate by the user. In this case, a threshold setting screen (not shown) is displayed on the display device 30, and the user operates the input device 31 to input a desired threshold value. The input device 31 is a touch panel, a keyboard or the like.

[0059] The flow velocity is calculated for each position (channel) in the direction perpendicular to the pipe axis of the pipe (in the pipe cross-section direction) as described above. It did not explain whether to use only channels or all channels only. Although it is possible to use only one arbitrary channel or to use any part (multiple) channels, the method considered appropriate for determining with high accuracy is shown below. Will be described with reference to FIG.

In the processing example shown in FIG. 6, first, as in FIG. 5, the threshold values Xs and Ys are arbitrarily set on the user side, and the threshold value Zs of an arbitrary success rate is set in advance. The success rate threshold Zs can be set to an arbitrary value between 0% and 100%, but the present inventor believes that the experiment result isopower is about 70%. The threshold values for power and standard deviation may be the values X and Y set by the manufacturer as in Fig. 4. Similarly, the threshold set for the success rate may be a value set by the manufacturer.

In FIG. 6, the processing in steps S31 and S32 is the same as in steps Sl 1 and S12, but the power (P) and standard deviation are calculated for all channels.

[0062] Then, for each channel, the power (P) and standard deviation are compared with the threshold values Xs and Ys to determine whether or not the measurement accuracy is poor. Count the number of channels that have been determined not to hesitate. Then, the success rate = nZN is calculated from the count number n and the total channel number N. Then, the calculated success rate is compared with the success rate threshold Zs, and if the success rate ≥Zs (step S34, YES), the pulse Doppler method is used as it is, and the flow rate calculation processing unit 13 calculates the flow rate. (Step S 35 ) o On the other hand, if the success rate is Zs (step S34, NO), the process is switched to the propagation time difference method, and the processing of steps S15 to S18 described in FIG. 4 is executed (step S36).

Here, the calculation of the power (P) and the standard deviation described above is influenced by the set parameters. For example, power (P) is affected by the number of acquisitions. This number of acquisitions is, for example, 256 in the above description, and the power (P) increases as the number of acquisitions increases. Since it is affected by parameters in this way, normality / abnormality cannot be evaluated centrally by the above threshold.

[0064] Therefore, in order to suppress the power (P) and the standard deviation from fluctuating depending on the setting parameters, the method for calculating the power (P) and the standard deviation may be as follows.

Power) may be a value averaged by the number of acquisitions. That is, the power (P) calculated by the above equation (1) divided by the number of times of acquisition (power (P) Z number of times of acquisition) may be used as the power (P) calculation result.

[0065] Alternatively, the standard deviation is calculated by replacing the standard deviation σ obtained based on the above equation (2) with the average Doppler shift frequency.

Value normalized by Fd

It is good. This is because even if the value of σ is the same, the meaning differs depending on whether the average Doppler shift frequency Fd is large or small. In other words, even if the values of σ are the same, the average Doppler shift frequency

[0067]

When the value of fd is small, the noise is larger than when it is large. Of course, this In these cases, the threshold values X, Y, etc. are also determined based on averaged or standardized values.

[0068] Although the standard deviation is the standard deviation of the Doppler shift frequency fd, it may instead be the standard deviation of the Doppler shift phase angle cod. The relationship between the Doppler shift frequency fd and the Doppler shift phase angle cod is expressed by the following equation (3). Accordingly, the standard deviation of the Doppler shift phase angle cod is obtained by multiplying the standard deviation of the Doppler shift frequency fd calculated from the square root of the calculation result of the above equation (2) by 2πΤ. T is the pulse repetition period lZFprfC.

[0069] [Equation 3] fd = cod x

• · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · only when the measurement accuracy is determined to be poor based on the above threshold. However, depending on the state of the fluid being measured, for example, the calculation result of the above-mentioned unity and standard deviation may be almost the same value as the above-mentioned threshold value. As a result, the above switching is frequently performed because the force exceeds or exceeds the threshold. However, the measurement result by the pulse Doppler method and the measurement result by the propagation time difference method are slightly different, as shown in Fig.7A, for example.If both measurement methods are frequently switched in a short time, the measurement result is It will fluctuate frequently as shown in 7B.

[0070] In the pulse Doppler method, ultrasonic pulse transmission / reception is performed, for example, 256 times to measure the flow velocity of one channel, and an average value is obtained. On the other hand, with regard to the propagation time difference, after processing 256 times, the calculation of correlation, propagation time, propagation time difference, temperature correction, etc. is performed, and finally the flow velocity and flow rate of the fluid to be measured are calculated. Therefore, since the processing time until the flow rate calculation in the propagation time difference method and the pulse Doppler method is on the same order, for example, the state shown in FIG. 7B can be obtained. [0071] The ultrasonic flowmeter measures and displays the flow rate, but also displays the flow velocity, and it is often the case that a monitor or the like refers to this to determine whether there is an abnormality. For this reason, if the flow velocity fluctuates frequently as described above, it may be misunderstood that an abnormality has occurred. In addition, the effect on the output flow rate value is considerable.

[0072] In this example, in order to deal with this problem, the power to propose the following two methods is not limited to this example.

The first method is to continue measurement using the propagation time difference method until a certain time t has elapsed or the number of measurements n has been executed when switching to the pulse Doppler force propagation time difference method. The value of t or n can be set by the user, for example, and is set arbitrarily in advance. For example, if 20 is set as n, after switching to the propagation time difference method, measurement by the propagation time difference method is continuously performed 20 times, and then the pulse Doppler method is returned to determine the switching of the measurement method. . In this way, it is possible to prevent frequent switching between both measurement methods in a short time. Even in the situation shown in Fig. 7B, the measurement results are shown in Fig. 7C, for example.

[0073] In the second method, the flow velocity measurement result by the pulse Doppler method immediately before switching to the pulse Doppler force propagation time difference method is vl, and each flow velocity measurement is performed up to the number of measurements n by the propagation time difference method after switching. The result is v2 (k; an integer from l to n, where v2 is the kth

k k

The flow velocity V after switching is calculated by the following formula every time measurement is performed using the propagation time difference method.

[0074] v = (a-l) vl + a X v2

k k k

(Where n is the number of measurements set in the same way as in the first method, and a = (1 / n) X k

k

is there).

[0075] According to the above calculation formula, the value of the flow velocity V gradually approaches the value of the measurement result by the propagation time difference method from the value force close to vl, so that the output flow velocity value does not change abruptly. As shown in Fig. 7D, it will change gradually over time.

[0076] Then, when the measurement is performed n times, the pulse Doppler method is switched.

The second method is particularly effective when the difference between vl and v2 is large.

By the above first or second method, output disturbance due to switching of the measurement method can be suppressed. Togashi.

Finally, the positions (channels) on the measurement line are shown in Fig. 8 for the time being. CH1, CH2, CH3, etc. shown in Fig. 8 correspond to the positions (channels) on the measurement line. According to the ultrasonic flowmeter of the present invention, the two-type ultrasonic flowmeter, etc., the pulse doppler unilateral ultrasonic flowmeter is used to determine the state of the fluid to be measured, thereby ensuring the accuracy of the measurement result. In addition, an ultrasonic flow meter that combines the pulse Doppler method and the propagation time difference method is configured, and the pulse Doppler method is changed to the propagation time difference method according to the determination result of the state of the fluid to be measured. By switching, it is possible to measure the flow rate without degrading the measurement accuracy while energizing high-precision measurement by the pulse Doppler method. In addition, since the discrimination accuracy is high, switching is performed at an appropriate time.

Claims

The scope of the claims

[1] Visit a Nords-Doppler ultrasonic flow meter,

Based on the measured echo signal for each channel, based on the Doppler spectrum obtained, power calculation means for calculating the power indicating the intensity of the echo signal, and for each channel based on the measured echo signal The standard deviation calculating means for calculating the standard deviation based on the obtained Doppler shift frequency and the average value thereof, and for each channel, the power or Z and the standard deviation are respectively compared with a predetermined threshold value. And a state determining means for determining the state of the fluid to be measured,

A pulse Doppler type ultrasonic flowmeter characterized by comprising:

[2] An ultrasonic flow meter that combines the pulse Doppler method and the propagation time difference method, and normally uses the pulse Doppler method.

Based on the measured echo signal for each channel, power calculation means for calculating the power indicating the intensity of the echo signal based on the obtained Doppler spectrum, and for each channel based on the measured echo signal The standard deviation calculating means for calculating the standard deviation based on the Doppler shift frequency obtained and the average value thereof, and comparing the power or Z and the standard deviation with a predetermined threshold value for each channel. And a state determining means for determining the state of the fluid to be measured,

Switching means for deciding whether or not to temporarily switch to the propagation time difference method according to the determination result by the state determination means;

2 type combined ultrasonic flowmeter characterized by having

[3] The state determination means determines whether or not the measurement accuracy deteriorates for each channel!

The switching means switches the measurement method to the propagation time difference method when the number of channels determined that the measurement accuracy is deteriorated by the state determination means is equal to or greater than a preset ratio. The two-method combined ultrasonic flowmeter according to claim 2,

[4] The power according to claim 2 or 3, wherein the power is a total power obtained by integrating the Doppler spectrum or an averaged power obtained by dividing the total noise by the number of acquisitions. Type combined ultrasonic flowmeter.

[5] The standard deviation calculated based on the average value of the Doppler shift frequency is used instead of the standard deviation calculated based on the Doppler shift frequency and the average value thereof. The two-type combined ultrasonic flowmeter according to claim 2 or 3.

[6] The standard deviation calculation means calculates the standard deviation of the Doppler shift phase angle using the Doppler shift phase angle and the average value instead of the Doppler shift frequency and the average value thereof. The two-type combined ultrasonic flowmeter according to claim 2 or 3.

7. The two-type combined ultrasonic flowmeter according to claim 2, further comprising input means for setting and changing the threshold value.

[8] Set any number of measurements or measurement time in advance,

3. The method according to claim 2, wherein when the switching means is temporarily switched to the propagation time difference method, the measurement by the propagation time difference method is performed for the set number of measurements or measurement time. Type combined ultrasonic flowmeter.

[9] When the switching means is temporarily switched to the propagation time difference method, the measurement result by the propagation time difference method is corrected and output so as to gradually change from the measurement result by the pulse Doppler method immediately before the switching. The two-method combined ultrasonic flowmeter according to claim 2, wherein the two-method ultrasonic flowmeter is used.

[10] The computer in the Nords Doppler ultrasonic flow meter

A function for calculating the power indicating the intensity of the echo signal based on the Doppler spectrum obtained based on the measured echo signal for each channel;

A function for calculating the standard deviation for each channel based on the measured Doppler shift frequency and the average value based on the measured echo signal,

A function for determining the state of the fluid to be measured by comparing the power or Z and standard deviation with a predetermined threshold value for each channel; and

A program to realize

[11] For the computer in the ultrasonic flowmeter that combines the pulse Doppler method and the propagation time difference method,

In normal times, the pulse Doppler method shall be used.

For each channel, the Doppler spectrum obtained based on the measured echo signal Based on the function of calculating the power indicating the intensity of the echo signal;

A function for determining whether or not to switch to the propagation time difference method according to the determination result by the state determination means;

A program to realize