JP4548485B2 - Pulse doppler type ultrasonic flow meter, its program - Google Patents

Description

  The present invention relates to a pulse Doppler ultrasonic flow meter.

  A clamp-on type ultrasonic flowmeter is equipped with an ultrasonic transducer (module that transmits and receives ultrasonic pulses of an arbitrary frequency) on a part of the outer peripheral surface of a tubular body such as a water pipe, and the fluid flowing inside the tubular body This is a flow meter that measures the flow velocity / flow rate from the outside of the tubular body. Clamp-on type ultrasonic flowmeters can be roughly classified into a propagation time difference type and a pulse Doppler type.

  The propagation time difference formula is based on the difference in time required for the ultrasonic wave to travel in the forward path and the return path by reciprocating the ultrasonic wave in a path that crosses the fluid flowing inside the tubular body diagonally. Is a method of measuring.

  On the other hand, the pulse Doppler method is a method of measuring the flow rate of fluid from the moving speed of suspended particles, bubbles, etc., assuming that suspended particles, bubbles, etc. contained in the fluid move at the same speed as the fluid. This is because the frequency of the ultrasonic wave reflected by suspended particles or the like changes due to the Doppler effect, and the flow velocity distribution of the fluid is calculated based on the frequency deviation. The flow rate of the fluid is calculated by integral calculation.

Such a pulse-Doppler flow measurement technique, for example, as disclosed in Japanese Patent Application Laid-Open No. H10-228867, enables non-contact and high-precision measurement with respect to a fluid in an unsteady state. In this technique, an ultrasonic pulse having a frequency f ₀ is transmitted to a fluid to be measured at a constant interval of a repetition frequency f _prf and an ultrasonic echo reflected by a reflector on a survey line is received. Based on this, the Doppler shift frequency is calculated to determine the flow velocity distribution of the fluid to be measured, and the flow rate can be measured by deriving the flow rate by integral calculation based on this flow velocity distribution.

  Regarding the flow rate measurement technique using the pulse Doppler method, Patent Document 2 discloses a flow rate switching method that doubles the flow rate measurement range by shifting the flow rate measurement range to either positive or negative. Presented.

  The pulse Doppler method is a method for obtaining a flow rate by obtaining Doppler shift frequencies at a plurality of points (hereinafter referred to as channels) in a pipe, calculating a flow velocity thereof, and integrating the cross section of the pipe. The flow velocity is calculated from the Doppler shift frequency by the following equation (1):

However, the Doppler shift frequency f _fd is the sampling theorem, can not be measured only within the range determined by the pulse repetition frequency _{f prf (-f prf / 2~f prf} / 2 range).
From this, in the technique of the above-mentioned patent document 2, what is conventionally assigned to the measurement range equally to positive and negative is assigned only to positive or only negative. Therefore, even if the measurement range is doubled, it does not really double the measurement range itself. That is, the measurement range and 0 to F _prf in the case of positive alone, but only has a -f _prf ~0 the measurement range in the case of a negative only. In other words, the measurement range is, after all, the value of the pulse repetition frequency f _prf itself, and cannot be increased further. The positive and negative mean the fluid flow direction (positive direction, negative direction).
JP 2000-97742 A JP, 2004-61109, A Even if a Doppler shift frequency exceeds the measuring range decided by repetition frequency, the subject of the present invention is correct Doppler shift frequency measurement, and correct Doppler shift frequency distribution by performing amendment suitably. In particular, the present invention provides a pulse Doppler type ultrasonic flowmeter that can cope with a case where the measurement range is greatly exceeded.

  The pulse Doppler type ultrasonic flowmeter of the present invention is a pulse Doppler type ultrasonic flowmeter that measures the flow rate of a fluid to be measured that flows in a pipe using ultrasonic Doppler shift. A Doppler shift frequency for transmitting a pulse to the fluid to be measured, receiving an ultrasonic echo thereof, and measuring a Doppler shift frequency of each channel from the first channel closest to the pipe wall of the pipe to the m-th channel in the center of the pipe In the Doppler shift frequency distribution from the first channel to the m-th channel obtained from the measurement result by the measurement means and the Doppler shift frequency measurement means, a channel range affected by the aliasing is obtained, and the obtained channel range The measured value of Doppler shift frequency of each channel in A Doppler shift frequency distribution correction unit that corrects using a correction value obtained based on a fixed range, and a flow rate calculation that determines the flow rate of the fluid under measurement using the Doppler shift frequency distribution corrected by the Doppler shift frequency distribution correction unit. Means.

As described above, in the pulse Doppler ultrasonic flow meter, the measurement range of the Doppler shift frequency is determined by the pulse repetition frequency f _prf . If the Doppler shift frequency exceeds this measurement range, the measured value is obtained as an incorrect value affected by aliasing.

  Thus, in the pulse Doppler type ultrasonic flowmeter of the present invention, the Doppler shift frequency distribution correction means discriminates all the channels whose Doppler shift frequency measurement values are affected by the aliasing. The determined Doppler shift frequency measurement value of each channel is corrected using the correction value.

  In the pulse Doppler type ultrasonic flowmeter having the above-described configuration, for example, the Doppler shift frequency distribution correction unit determines whether or not the channel where the aliasing has occurred, and whether or not the first channel has already been affected by the aliasing. If the first channel is not affected by the aliasing, the measured value of the Doppler shift frequency of each channel in the range from the first channel to the channel where the aliasing occurs is not corrected, The measurement value of the Doppler shift frequency of each channel in the range from the channel where the aliasing occurs to the m-th channel is corrected using the measurement range of the Doppler shift frequency as the correction value.

The flow rate of the fluid to be measured is not uniform, and the center of the tube is faster than the vicinity of the tube wall. Therefore, the absolute value of the Doppler shift frequency is larger in the center of the tube than in the vicinity of the tube wall. As a result, it is not normally possible that the channel near the tube wall is affected by folding, but the channel on the center side of the tube is not affected by folding. On the contrary, it is possible that the channel near the tube wall is not affected by folding, but the channel on the center side of the tube is affected by folding. Therefore, in the above configuration, if there is a turn-up point, the channel in the range from there to the center of the tube is assumed to be affected by the turn-back, and correction is performed using the measurement range f _prf as the correction value. I do.

  In the pulse Doppler type ultrasonic flowmeter having the above-described configuration, for example, the Doppler shift frequency distribution correction means determines the channel where the aliasing occurs, and is the first channel already affected by the aliasing? If the first channel is affected by aliasing, the measured value of the Doppler shift frequency of each channel in the range from the first channel to the channel where the aliasing occurs is The Doppler shift frequency measurement range is used as the correction value, the correction is performed, and the Doppler shift frequency measurement value of each channel in the range from the channel where the aliasing occurs to the m-th channel is measured. Two times the range is corrected as the correction value.

It is possible that all channels are affected by aliasing. In this case, if there is a place where folding occurs, the channel in the range from there to the center of the pipe is doubly affected by the folding. Therefore, twice the measurement range (2f _prf ) is used as a correction value for channels in this range. Of course, each channel in the range from the first channel to the place where the aliasing occurs is corrected using the measurement range f _prf as a correction value.

  In the pulse Doppler type ultrasonic flowmeter having the above-described configuration, for example, the Doppler shift frequency distribution correction unit determines the channel where the aliasing occurs and is already affected by the aliasing in the first channel. In the case where the first channel is not affected by the loopback and there are a plurality of loopback occurrence locations, the first loopback occurrence location from the first channel is determined. The measured value of the Doppler shift frequency of each channel in the range up to the channel is not corrected. For the range from the channel at the first aliasing occurrence point to the m-th channel, the even-numbered aliasing channel at the aliasing occurrence point To the odd-numbered channel where the fold occurs is not corrected and the odd-numbered For the range from the channel where the aliasing occurs to the even-numbered channel where the aliasing occurs, the measured value of the Doppler shift frequency of each channel in the range is the correction value and the measured range of the Doppler shift frequency is the correction value. Correct as

  As described above, since the absolute value of the Doppler shift frequency is larger in the center of the tube than in the vicinity of the tube wall, the Doppler shift frequency distribution from the vicinity of the tube wall to the center of the tube is, for example, when the flow is in the positive direction, Although it basically increases, there may be a partial decrease when there is a local flow velocity fluctuation. In this case, there may be a plurality of places where folding occurs. With the above configuration, even when there is a local variation in the flow velocity, the measured value of the Doppler shift frequency can be corrected and a correct Doppler shift frequency distribution can be obtained without any problem.

  In the pulse Doppler type ultrasonic flowmeter having the above-described configuration, for example, the Doppler shift frequency distribution correction unit determines the channel where the aliasing occurs and is already affected by the aliasing in the first channel. In the case where the first channel is affected by folding and there are a plurality of places where the folding occurs, the first occurrence of the folding from the first channel is determined. The measurement value of the Doppler shift frequency of each channel in the range up to the channel is corrected using the measurement range of the Doppler shift frequency as the correction value, and the range from the channel where the first aliasing occurs to the m-th channel Is the odd-numbered occurrence of the folding from the channel of the occurrence of the folding. For the range up to the number of channels, the measured value of the Doppler shift frequency of each channel within the range is corrected using the measured range of the Doppler shift frequency as the correction value, and the even-numbered channel from the occurrence of the aliasing is corrected. For the range up to the channel at the position where the folding occurs, the measured value of the Doppler shift frequency of each channel within the range is corrected using twice the measured range of the Doppler shift frequency as the correction value.

  Even in the above configuration, even when there is a local variation in the flow velocity, the correct Doppler shift frequency distribution can be obtained by correcting the measured value of the Doppler shift frequency without any problem.

  In the pulse Doppler type ultrasonic flowmeter having the above-described configuration, for example, the Doppler shift frequency distribution correction unit further determines the flow direction of the fluid to be measured, and the determined flow direction is a positive direction. The correction is a process of adding the correction value to the measured value of the Doppler shift frequency of each channel.

  Similarly, in the pulse Doppler type ultrasonic flowmeter having the above-described configuration, for example, the Doppler shift frequency distribution correction unit further determines the flow direction of the fluid to be measured, and the determined flow direction is a negative direction. The correction is a process of subtracting the correction value from the measured value of the Doppler shift frequency of each channel.

  The flow direction of the fluid to be measured is determined in the range from the first channel to the channel where the folding occurs or in the range from the first channel to the channel where the measured value of the Doppler shift frequency crosses zero. The cumulative addition value or average value of the measured values of the Doppler shift frequency of each channel is calculated. When the calculated cumulative addition value or average value is a positive value, the positive direction is obtained. When the calculated cumulative addition value or average value is a negative value, the negative direction is obtained. To do.

  Even in the case where the measurement target is the one in which the flow direction of the fluid to be measured is reversed depending on the time zone, for example, the flow direction is automatically determined without any manual intervention according to the above configuration, and the flow direction is determined according to the flow direction. Correction processing can be performed, and it can be handled by installing only one flow meter.

  According to the pulse Doppler type ultrasonic flowmeter of the present invention, even when the Doppler shift frequency exceeds the measurement range determined by the repetition frequency, it is possible to obtain a correct Doppler shift frequency distribution by performing correction processing, The measurable range of the Doppler shift frequency can be substantially larger than the measurement range determined by the repetition frequency. In particular, even if the Doppler shift frequency is a value that cannot be measured using the method of Patent Document 2, it can be measured. Furthermore, it is not only possible to deal with either positive / negative, but it can deal with both positive / negative.

It is a block diagram of a pulse Doppler type ultrasonic flowmeter. It is an example of the measurement result of the Doppler shift frequency exceeding a measurement range. It is an example of the measurement result of the Doppler shift frequency exceeding a measurement range. It is an example (the 1) of a measurement result of Doppler shift frequency exceeding a measurement range. It is a figure which shows the flow direction determination about the example of FIG. 3A. It is the example which frequency-corrected the measurement result of FIG. 3A. It is an example (the 2) of a measurement result of the Doppler shift frequency exceeding a measurement range. It is a figure which shows the flow direction determination about the example of FIG. 3D. It is the example which frequency-corrected the measurement result of FIG. 3D. It is an example (the 3) of a measurement result of the Doppler shift frequency exceeding a measurement range. It is a figure which shows the flow direction determination about the example of FIG. 4A. It is the example which frequency-corrected the measurement result of FIG. 4A. It is an example (the 4) of the measurement result of the Doppler shift frequency exceeding a measurement range. It is a figure which shows the flow direction determination about the example of FIG. 4D. It is the example which frequency-corrected the measurement result of FIG. 4D. It is a flowchart figure which shows the whole flow volume calculation process roughly. It is a detailed flowchart figure of the process of step S2 of FIG. It is a detailed flowchart figure of the process of step S3 of FIG. It is a detailed flowchart figure of the process of step S38 of FIG. It is a detailed flowchart figure of the process of step S39 of FIG. It is a detailed flowchart figure of the process of step S41 of FIG. It is a detailed flowchart figure of the process of step S42 of FIG. It is an example of a measurement result of a Doppler shift frequency that exceeds the measurement range and has a local variation. It is a figure which shows the flow direction determination about the example of FIG. 12A. It is the example which frequency-corrected the measurement result of FIG. 12A. It is an example of the to-be-measured object from which the flow direction of a fluid changes. It is an example of the to-be-measured object from which the flow direction of a fluid changes.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of a pulse Doppler type ultrasonic flowmeter of the present embodiment.

The illustrated pulse Doppler type ultrasonic flowmeter includes an ultrasonic transducer 1, an emitter 2, a transmitter 3, an amplifier 4, an A / D converter 5, a display device 6, and a CPU 10.
The periodic signal output from the transmitter 3 is input to the emitter 2, and the emitter 2 generates an electrical pulse for causing the ultrasonic transducer 1 to transmit the ultrasonic pulse from the periodic signal, and the ultrasonic transducer 1 input. The ultrasonic transducer 1 sends out an ultrasonic pulse into the fluid through the pipe wall by this electric pulse, and receives a reflected echo. The reflected echo is converted into an electric signal by the ultrasonic transducer 1 and amplified by the amplifier 4. The amplified electrical signal and the periodic signal of the transmitter 3 (that is, the signal of the transmission wave and the reception wave) are converted into a digital signal by the A / D converter 5 and input to the CPU 10.

  The ultrasonic transducer 1 is installed on the pipe wall 7 of the pipe, transmits an ultrasonic pulse toward the pipe wall 8 on the opposite side, and is in the middle (contained in the fluid, approximately equal to the velocity of the fluid). A reflected echo reflected by a reflector 9 such as a bubble (moving at the same speed) is received.

  The CPU 10 includes a Doppler frequency distribution measurement unit 11, a frequency distribution correction processing unit 12, and a flow rate calculation processing unit 13. The Doppler frequency distribution measurement unit 11 obtains a Doppler frequency distribution in a range from the tube wall 8 to the center of the tube as shown in FIG. That is, the frequency difference between the transmission wave and the reception wave output from the A / D converter 5 is obtained. The frequency distribution correction processing unit 12 is a range from the Doppler shift frequency of each channel obtained by the Doppler frequency distribution measuring unit 11 to the cross section of the pipe (actually, from the pipe center to the pipe wall 8 (opposite pipe wall). ) Distribution of Doppler shift frequency. The flow rate calculation processing unit 13 obtains a flow velocity distribution based on the Doppler shift frequency distribution, and calculates a flow rate based on the flow velocity distribution.

  The Doppler frequency distribution measurement unit 11 may be realized by a dedicated circuit. In this case, the output of the A / D converter 5 is input to this Doppler frequency distribution measuring circuit, and the Doppler frequency distribution obtained by this circuit is input to the CPU 10. In any case, the Doppler frequency distribution measurement unit 11, the circuit, and the flow rate calculation processing unit 13 are existing configurations and will not be described in detail.

A feature of the present invention resides in the frequency distribution correction processing unit 12. That is, in the conventional frequency distribution processing unit, exceeds if the flow velocity is fast, or "range determined by the repetition frequency _{f prf (-f prf / 2~f prf} / 2 range)" Doppler shift frequency is the moiety If there is, “folding” occurs, so that the distribution of the calculated Doppler shift frequency becomes wrong. Since the flow velocity distribution and the flow rate are calculated based on the Doppler shift frequency distribution, there is a problem that the calculation result of the flow velocity distribution and the flow rate is also wrong. The frequency distribution correction processing unit 12 of the present invention corrects the Doppler shift frequency distribution in which “folding” has occurred to a correct one. As a result, even when “turnback” occurs, the flow velocity distribution and the flow rate can be calculated correctly.

Here, the “turning back” will be described.
2A and 2B show Doppler shift frequency distributions when “folding” occurs.

  2A shows an example in which the flow is in the positive direction, and FIG. 2B shows an example in which the flow is in the negative direction. As shown in FIG. 1, the flow is normally “positive” when the transmission direction of the ultrasonic pulse is upstream of the flow, and the flow is opposite when the transmission direction is the opposite. Distinguishes from “negative direction”. Normally, it can be identified that “the flow is positive” if the Doppler shift frequency is a positive value, and “the flow is negative” if the Doppler shift frequency is a negative value. However, when “folding” occurs, as shown in FIGS. 2A and 2B, there are both a positive part and a negative part of the Doppler shift frequency. Although details will be described later, in the example of FIGS. 2A and 2B, the value of the Doppler shift frequency in the vicinity of the tube wall 8 (range of channel 0 to channel i) is correct (no aliasing occurs). In 2A, since the Doppler shift frequency near the tube wall 8 is positive, the flow can be determined as a positive direction, and in FIG. 2B, the flow can be determined as a negative direction. A method of determining the flow direction in the case where “folding” occurs will be described later.

In the example of FIG. 2A, since the flow is in the positive direction, the value of the Doppler shift frequency gradually increases from the channel 0 near the tube wall 8 toward the tube center (as described above, the flow velocity is higher in the tube center. This shows a case where the Doppler shift frequency in an arbitrary channel i shown in the figure is approximately f _prf / 2, and the Doppler shift frequency of the next channel i + 1 exceeds f _prf / 2. In this case, as shown in the figure, the Doppler shift frequency of the channel i + 1 becomes approximately −f _prf / 2 due to “folding”, and the value of the Doppler shift frequency after the channel i + 2 gradually increases again from −f _prf / 2. Go. That is, in this example, the channel range (i + 1 to m) from the channel i + 1 to the channel m at the center of the tube is affected by the “turnback”, and the value of the Doppler shift frequency is incorrect.

In the example of FIG. 2B, the value of the Doppler shift frequency gradually decreases from the channel near the tube wall 8 toward the center of the tube, contrary to FIG. 2A. When −f _prf / 2 is _exceeded , it becomes “f _prf / 2” by “folding” and decreases from there.

  FIG. 3 and FIG. 4 show examples of Doppler shift frequency distributions when the above “folding” occurs, and examples of Doppler shift frequency distributions corrected by the frequency distribution correction processing unit 12.

  The Doppler shift frequency distribution when “folding” occurs is roughly divided into four patterns. That is, there are four patterns shown in FIGS. 3A, 3D, 4A, and 4D. Among these, FIGS. 3A and 4A are the same as those shown in FIGS. 2A and 2B. That is, the channel in the vicinity of the tube wall 8 is not affected by the “turnback”, and the measured value of the Doppler shift frequency is correct, but somewhere in the range from the channel to the channel m at the center of the tube. , "Folding" occurs, and in the range from this "folding" generation channel i to the channel m at the center of the tube, the measured value of the Doppler shift frequency becomes an incorrect value due to the influence of "folding". It is what.

  On the other hand, FIGS. 3D and 4D show patterns in which “folding” has already occurred in a channel (channel 0 or the like) near the tube wall 8. Further, in this example, “folding” occurs immediately before channel i + 1. That is, in the range from channel i + 1 to channel m, the influence of “folding” is doubly affected. The same applies to FIG. 4D.

  FIG. 3C shows the frequency distribution of FIG. 3A corrected by the frequency distribution correction processing unit 12. Similarly, FIG. 3F for FIG. 3D, FIG. 4C for FIG. 4A, and FIG. 4F for FIG. 4D are the frequency distributions corrected by the frequency distribution correction processing unit 12. Naturally, the corrected frequency distribution is correct.

As shown in FIG. 3C, the frequency distribution shown in FIG. 3A is correct from channel 0 to channel i, but is incorrect after channel i + 1, and is actually over f _prf / 2 after channel i + 1. In the vicinity, it exceeds f _prf and is close to 3 / 2f _prf . Therefore, even if the method of Patent Document 2 is used, “folding” occurs in the portion _exceeding f _prf , the Doppler shift frequency distribution has an incorrect content, and the flow measurement result has an incorrect value.

  First, the frequency distribution correction processing unit 12 determines which of the above four patterns corresponds, and executes correction processing for the corresponding pattern, thereby correcting the wrong frequency distribution to the correct one.

In the case of correcting the frequency distribution shown in FIG. 3A, basically, a location where “folding” occurs (channel i + 1) is obtained, and all channels after this channel i + 1 (channel i + 1 to channel m at the center of the tube). ) may be added to f _prf the Doppler shift frequency (which is the same in the pattern of FIG. 4A. However, the in "addition" without "subtraction"). However, it is desirable to be able to cope with local fluctuations as in the processing example described later.

For example, as shown in FIG. 3B, the channel where the “folding” occurs can be obtained by the fact that the difference from the previous channel i is f _prf / 2 or more. As described above, the frequency difference between channel i and channel i + 1 is a very large value (approximately about f _prf ), and the frequency difference between adjacent channels is usually not more than f _prf / 2. (Only when “wrapping” occurs). Therefore, naturally, the threshold value for determining the occurrence of “folding” is not limited to “f _prf / 2 or more”, but may be any value that is not normally possible (for example, “2 f _prf / 3 or more”). The same applies to other patterns (as shown in FIGS. 3E, 4B, and 4E).

On the other hand, in the case of the patterns of FIGS. 3D and 4D, the channel 0 to channel i values are also affected by the “turnback”. Further, the channel i + 1 to the channel m are affected by “folding” twice. Thus, in the case of FIG. 3D, basically, f _prf is added to the Doppler shift frequency of the channel 0 to channel i portions, and 2 × f _prf is added to the Doppler shift frequency of the channel i + 1 to channel m portions. What is necessary is just to add. However, it is desirable to be able to cope with local fluctuations as in the processing example described later. Thereby, the Doppler shift frequency distribution shown in FIG. 3F is obtained. The same applies to the case of FIG. 4D, but “subtraction”, not “addition”.

As shown in FIG. 3F, the Doppler shift frequency of the channel m at the center of the tube exceeds 3 f _prf / 2 and cannot be measured by the conventional technique (even using the method of Patent Document 2).
As described above, according to the pulse Doppler type ultrasonic flowmeter of this example, even when the value of the Doppler shift frequency greatly exceeds the measurement range determined by the pulse repetition frequency f _prf because the flow velocity is very fast. Correcting the measured Doppler shift frequency distribution (2 times or more, 3 times or more) enables to obtain a correct flow velocity / flow rate calculation result.

  The processing by the frequency distribution correction processing unit 12 has been schematically described above, but will be described in detail below with reference to the flowcharts shown in FIGS. Note that the processing described schematically and the processing in the flowcharts shown in FIGS. 5 to 11 are performed by setting predetermined application programs and various setting values such as threshold values stored in a memory (not shown) (ROM, flash memory, etc.). This is realized by the CPU 10 reading and executing. The same applies to the calculation processing of the Doppler shift frequency distribution and the flow velocity / flow rate calculation processing.

FIG. 5 is a flowchart schematically showing the entire flow rate calculation process by the CPU 10.
As shown in FIG. 5, in the CPU 10, first, the Doppler frequency distribution measurement unit 11 calculates the Doppler shift frequency of each channel (step S1), and then the frequency distribution correction processing unit 12 first obtains the flow direction and the like. To determine which of the above four patterns corresponds (step S2), and by performing frequency shift processing using the processing flow for the determined pattern (step S3), the Doppler shift frequency distribution is “turned back”. It corrects to the right thing which excluded the influence of.

  After that, if the flow rate calculation processing unit 13 performs the conventional flow rate calculation process (step S4) and the flow rate calculation process (step S5) using the corrected Doppler shift frequency distribution, the flow rate of the fluid will be within the measurement range. Even under such circumstances, the flow velocity and flow rate can be calculated correctly.

FIG. 6 is a detailed flowchart of the flow direction determination process in step S2.
In the processing of FIG. 6, first, the wall nearest channel is set as the current channel. That is, first, the wall nearest channel is set as a target of processing in steps S12 to S15. The channel closest to the wall is the channel 0 closest to the tube wall 8. After that, for all channels from channel 0 closest to the tube wall 8 to channel m at the center of the tube, such as channel 1, channel 2,... The process is repeatedly executed (steps S16 and S17). That is, the channel to be processed is channel i (i = 0 to m), i = 0 at the beginning, and the processing of steps S12 to S15 is performed until i = m while performing the processing of i = i + 1 in step S17. Run repeatedly.

The process of steps S12 to S15 will be described. This process is basically a process of finding and recording the “folding” occurrence location. First, the difference p _i between the current channel and the next channel is calculated (the Doppler shift frequency of the next channel i + 1−the Doppler frequency of the current channel i), and this is set as the difference value p _i for the current channel i (step S12). ). If the absolute value of the difference value p _i is equal to or _larger than a predetermined threshold (as described above, here, f _prf / 2) (step S13, YES), the phase displacement flag corresponding to the next channel i + 1 is turned on. If it is less than the threshold (step S13, NO), the phase displacement flag corresponding to the next channel i + 1 is kept OFF (step S15). Although not particularly illustrated, a first table in which a phase displacement flag is associated with each channel is prepared in advance, and all phase displacement flags are OFF by default. The difference value p _i is temporarily stored. Further, as described above, the threshold value is not limited to f _prf / 2, and may be any value that cannot normally occur except when “folding” occurs.

The next channel is a channel adjacent to the current channel in the center of the tube.
By creating / updating the first table by the processing described above, it can be seen that the channel whose phase displacement flag is ON is the place where “folding” occurs.

Next, again, the channel 0 closest to the tube wall 8 is set as the current channel (step S18), and all channels from the channel 0 to the channel m at the center of the tube are sequentially processed (step S21, S18). S22), Steps S19 and S20 are repeatedly executed. Although not shown in the figure, in the process of step S18, a variable P indicating the cumulative addition value of the difference value p _i is initialized (P = 0).

In the processes of steps S19 and S20, first, referring to the first table, it is determined whether or not the phase displacement flag corresponding to the next channel i + 1 of the current processing target channel i is OFF (step S19). (Step S19, NO) No processing is performed, and when it is OFF (step S19, YES), the difference value p _i of the current channel i is added to the variable P (P = performs P + _{p i)} processing (step S20).

When the processing of steps S19 and S20 is executed for all channels from channel 0 to channel m (step S21, YES), the current flow direction of the fluid to be measured is determined depending on whether the variable P is positive or negative. That is, if the variable P (accumulated sum of the difference values p _i) is positive (step S23, YES), if the flow is determined to be positive (step S24), and the variable P is negative (Step S23, NO), it is determined that the flow is in the negative direction (Step S25).

When the determination in step S19 is NO, the difference value p _{i of the} channel is not added to the variable P. Therefore, the influence of “turnback” is eliminated, and the flow direction can be determined without any problem by the variable P. That is, as described above, the flow is faster in the center of the tube than near the tube wall. For example, when the flow is in the positive direction, basically, the flow is in the next channel rather than the current channel. Since the flow is fast (that is, the Doppler shift frequency is large), the difference value p _i is basically a positive value (this is basically a positive value even in the range affected by “folding”). Value). However, in some cases, as shown later in FIG. 12, there is a possibility that a portion where the difference value p _i becomes a negative value partially due to local flow velocity fluctuations. Positive / negative is judged by the value. Further, since the difference value p _i becomes a large minus (approximately −f _prf ) at the “turn-back” location, the difference value at the “turn-back” location is excluded in step S19. The value of is always a positive value.

When the flow is in the negative direction, it may be considered that the flow is opposite to that in the positive direction. That, of course also in this case, towards the tube center is faster flow than near the tube wall (Thus, the absolute value of the difference value p _i is the larger tube center) is, since the flow is in the negative direction Basically, since the Doppler shift frequency of each channel decreases from the position closest to the tube wall toward the center of the tube, the difference value p _i is a negative value (except for some of the fluctuation portions). Therefore, the variable P is also a negative value.

By the above process, the flow direction can be definitely determined even when “turnback” occurs or when a local flow velocity fluctuation occurs.
When the flow direction determination process in FIG. 6 is completed, the frequency shift process in step S3 in FIG. 5 is subsequently executed.

FIG. 7 shows a detailed flowchart of this frequency shift process.
Since the flow direction can be determined by the above processing, for example, when the flow is in the positive direction, it can be seen that the pattern is one of FIGS. 3A and 3D. As described above, FIG. 3A or 3D can be discriminated based on whether or not the channel 0 immediately adjacent to the tube wall 8 which is the processing start channel (the processing reference) is already affected by “turning back”. . In the process of FIG. 7, in addition to the result of determining the flow direction, it is determined whether any of the above four patterns is satisfied by determining whether or not the channel 0 closest to the tube wall 8 has already been affected by “folding”. 7 is executed, and one of steps S38, S39, S41, and S42 in FIG. 7 is executed according to the determination result. A detailed flowchart of step S38 is shown in FIG. 8, a detailed flowchart of step S39 is shown in FIG. 9, a detailed flowchart of step S41 is shown in FIG. 10, and a detailed flowchart of step S42 is shown in FIG.

  In the process of FIG. 7, the average value of the Doppler shift frequency in the range from the channel closest to the wall to the channel whose phase displacement flag is ON is obtained, or the Doppler shift frequency in the range from the closest wall to the Doppler shift frequency crosses zero. Find the average value of.

That is, first, the channel 0 closest to the tube wall 8 is set as the current channel (channel to be processed) (step S31), and the channels 1, 2, 3,... To the processing target channel (step S35), until the channel whose phase displacement flag is ON becomes the current channel, or until the Doppler shift frequency crosses 0 (that is, NO in step S32 or YES in step S34) Until Doppler shift frequency of each channel is cumulatively added (step S33). That is, _assuming that the Doppler shift frequency of channel i is f _fdi , the variable indicating the cumulative addition value of the Doppler shift frequency is Q, and the variable Q is initialized (Q = 0) in step S31, the process of step S33 is Q = Q + f _fdi .

  The determination method of whether the Doppler shift frequency has crossed 0 is, for example, by comparing the Doppler shift frequency of the current channel i with the Doppler shift frequency of the next channel i + 1, one being a positive value and the other being a negative value. If the value is, it is determined that the Doppler shift frequency has crossed zero.

  In the example of FIG. 3, the value of the variable (cumulative addition value) Q obtained by the above processing is the cumulative addition value Q of the Doppler shift frequency of each channel in the range A shown in the example of FIG. 3A. Will be required. That is, in this case, NO is determined in step S32, and the cumulative addition value Q of the Doppler shift frequencies of each channel in the range A from channel 0 to channel i immediately before channel i + 1 whose phase displacement flag is ON is obtained. Become. If the pattern shown in FIG. 3D is used, the cumulative addition value Q of the Doppler shift frequencies of each channel in the range E shown in the figure is obtained. That is, in this case, YES is obtained in step S34, and the cumulative addition value Q of the Doppler shift frequencies of each channel from the channel 0 to the channel immediately before the Doppler shift frequency crosses 0 is obtained.

  As is clear from FIGS. 3A and 3D, the cumulative added value Q is a positive value in the case of FIG. 3A, and the cumulative added value Q is a negative value in the case of FIG. 3D. In the case of the pattern as shown in FIG. 3A, as already described, the measured value of the Doppler shift frequency in the range A from the channel 0 closest to the tube wall 8 to the channel i immediately before the “folding” occurrence point is correct (“folding”). Since the value is not affected by (), it is a positive value. However, in consideration of the possibility that there is a channel in which the Doppler shift frequency is partially negative due to local fluctuations, the cumulative addition value of the entire range A is obtained (the variable Q is therefore set to the cumulative addition value). For example, it may be an average value).

  On the other hand, in the case of the pattern of FIG. 3D, the region of the range B shown in the figure (the range from the channel 0 immediately adjacent to the tube wall 8 to the channel i immediately before the “folding” occurrence point) is also affected by the “folding”, In such a case, the Doppler shift frequency of each channel in the region in the vicinity of the tube wall 8 (the range E in the figure) where the flow velocity is relatively low is usually positive as shown in the drawing. Nevertheless, it is known to be negative.

  4A and 4D, conversely, when the region near the tube wall 8 is not affected by “folding”, the variable Q is a negative value and is affected by “folding”. The variable Q has a positive value.

  Accordingly, when the flow is in the positive direction (step S36, YES) and the variable Q is a positive value (step S37, NO), the Doppler shift frequency correction process (process) corresponding to the pattern of FIG. a) is executed (step S38). Similarly, when the flow is in the positive direction (step S36, YES) and the variable Q is a negative value (step S37, YES), the Doppler shift frequency correction process (process) corresponding to the pattern of FIG. b) is executed (step S39). Similarly, when the flow is in the negative direction (step S36, NO) and the variable Q is a positive value (step S40, YES), a Doppler shift frequency correction process (process) corresponding to the pattern of FIG. 4D. d) is executed (step S42). Similarly, when the flow is in the negative direction (step S36, NO) and the variable Q is a negative value (step S40, NO), the Doppler shift frequency correction process (process) corresponding to the pattern of FIG. 4A is performed. c) is executed (step S41).

Hereinafter, the processes a, b, c, and d will be described in detail with reference to FIGS.
First, a detailed flowchart of process a shown in FIG. 8 will be described. In other words, the Doppler shift frequency correction process when the flow is in the forward direction and the region near the tube wall 8 is not affected by “folding” will be described.

In the process of FIG. 8, first, the range frequency R is initialized (R = 0) (step S51). The range frequency R means a correction amount of the Doppler shift frequency. In the case of the example of FIG. 3A, as described above, there is no need to correct the range A shown in the figure, so that R = 0, and for the range C shown in FIG. Since it is sufficient to add f _prf uniformly to the shift frequency, the range frequency R is set to f _prf in step S56.

In step S51, the phase-shifted flag is set to OFF. This phase-shifted flag and the processing of steps S54 to S56 are not necessary in the case of the example shown in FIG. 3A. However, the case of FIG. 3A may be in the state shown in FIG. 12A due to local fluctuations, for example. In the example of FIG. 12A, an arbitrary range F within the range C is again a value within the measurement range (here, less than f _prf / 2) because the value of the Doppler shift frequency is partially reduced due to local fluctuation. It has become. In this case, the value of the Doppler shift frequency in the range F is correct and does not need to be corrected. From this situation, the phase-shifted flag is used to distinguish whether or not the current channel needs to be corrected.

However, before describing the processing in FIG. 8, first, processing in the case where local variation is not considered will be described. This process is a process that does not use the phase-shifted flag in the process of FIG. That is, the process of setting the phase-shifted flag to OFF in step S51 is not necessary, and the process of simply setting the range frequency R to f _prf may be executed instead of the processes of steps S54 to S56. Thus, when the range frequency R is initialized in step S51, first, the current channel i is set to channel 0 (i = 0) closest to the tube wall 8, and the phase shift flag of the current channel is referred to by referring to the first table. Is determined to be ON (step S53). Since the flag is OFF in the case of channel 0 (step S53, NO), the process directly proceeds to step S57. In step S57, the Doppler shift frequency of the current channel is corrected to a value obtained by adding the range frequency R to the measured Doppler shift frequency. In the case of channel 0, since R = 0, the Doppler shift frequency remains a measured value (not corrected).

Then, until the current channel becomes the channel m at the center of the tube, the same processing as described above is repeatedly performed by sequentially setting each channel from channel 0 to channel m as the current channel. Even after channel 1, until the channel whose phase displacement flag is ON becomes the current channel, the range frequency R remains 0, so the Doppler shift frequency remains the measured value (not corrected). Then, when the phase shift flag of the current channel is ON (step S53, YES), the above-mentioned “processing for setting the range frequency R to f _prf ” is performed. As a result, for all channels from the main channel to channel m at the center of the tube, the processing in step S57 becomes “Doppler shift frequency = measured value of Doppler shift frequency + f _prf ”.

When the processing in FIG. 10 is a processing that does not consider local fluctuations, the processing is the same as the above processing except that the range frequency R is −f _prf .
Next, the difference between the process of FIG. 8 and the process corresponding to the example of FIG. 3A described above will be described with reference to the example of FIG. 12A.

In this case, first, the range frequency R is initialized in step S51 (R = 0), and the phase-displaced flag is turned off. In the above processing, when the phase shift flag of the current channel is ON (step S53, YES), the “processing to set the range frequency R to f _prf ” is simply performed, but instead of steps S54 to S56. The difference is in the processing.

In the example of FIG. 12A, the phase displacement flags of the three channels j, k, and l are turned on. That is, first, since the measurement range f _prf / 2 is once exceeded in channel j, “turnback” occurs. In this example, as shown in FIG. 12C, the Doppler shift frequency is partially reduced. The Doppler shift frequency in the range of channel l is again below the measurement range f _prf / 2. FIG. 12C shows the corrected Doppler shift frequency distribution, that is, the original (correct) Doppler shift frequency distribution.

  In this case, since the Doppler shift frequency in the range of channel k to channel l is correct, there is no need to correct it. However, in the above-described processing that does not consider the local variation, the entire range from channel j to channel m at the center of the tube These channels are uniformly corrected.

  In this case, as shown in FIG. 12A, “folding” occurs in the channels k and l, so that the phase displacement flag is turned on not only for the channel j but also for the channels k and l by the processing of FIG. .

From the above, by performing the processing in FIG. 8, the Doppler shift frequency distribution can be corrected without any problem even when local fluctuations as shown in FIG. 12 occur.
That is, according to the process of FIG. 8, in the example of FIG. 12, when the current channel is channel j, k, l, the determination in step S53 is YES, and the process of steps S54 to S56 is performed. When the phase-shifted flag is ON, the process of step S55 is executed, and when it is OFF, the process of step S56 is executed.

First, when the current channel is channel j, since the phase-shifted flag remains OFF in step S51, the process of step S56 is executed. In step S56, f _prf is added to the current range frequency (correction value) R (R = R + f _prf ), and the phase-shifted flag is _turned ON. Here, since the range frequency R remains in the state initialized in step S51, R = 0 + f _prf = f _prf . Thus, for each channel from channel j to channel k−1, the processing in step S57 is “Doppler shift frequency = measured value of Doppler shift frequency + f _prf ”, and the Doppler shift frequency is corrected.

When the current channel becomes channel k, the determination in step S53 is YES and the phase-shifted flag is ON (step S54, YES), so the process of step S55 is executed. In the process of step S55, f _prf is subtracted from the current range frequency R (R = R−f _prf ) and the phase-shifted flag is turned OFF. Here, since the current range frequency R is f _prf , R = f _prf −f _prf = 0. Thus, for each channel from channel k to channel 1-1, the process of step S57 is “Doppler shift frequency = measured value of Doppler shift frequency + 0”, and the Doppler shift frequency is not substantially corrected. It will be.

Thereafter, when the current channel becomes l, the same processing as that for the channel j is performed. For each channel from the channel l to the channel m, the correction processing in step S57 is “Doppler shift frequency = Doppler”. The measured value of the shift frequency + f _prf ”, and the Doppler shift frequency is corrected.

Next, the process b shown in FIG. 9 will be described. The process of FIG. 9 is the same as the process of FIG. 8 except that the initial value of the range frequency R is set to f _prf in step S61. However, since the initial value of the range frequency R is set to f _prf , for each channel in the range B shown in FIG. 3D, the processing in step S67 is “Doppler shift frequency = measured value of Doppler shift frequency + f _prf ”. For each channel in the range D, the process of step S67 is “Doppler shift frequency = measured value of Doppler shift frequency + 2f _prf ”. If there are a plurality of channels whose phase displacement flag is ON due to local fluctuations as shown in FIG. 12, the step is performed for each channel in the range corresponding to the range of channel k to channel l. The process of S67 is “Doppler shift frequency = measured value of Doppler shift frequency + f _prf ”.

In addition, regarding the flow of the process c shown in FIG. 10, the processes in steps S75 and S76 are “addition” in the processes in steps S55 and S56 in FIG. Therefore, the correction process in step S77 is the same as the process in FIG. 8 except that “the correction is not performed” or “the Doppler shift frequency = the measured value of the Doppler shift frequency−f _prf ”. To do. According to the processing of FIG. 10, the range frequency R is set to R = R−f _prf = −f _{prf in} step S76 for the channel corresponding to the channel j, and then R = R = R−f _prf = −f _prf for the channel corresponding to the channel k. By setting R + f _prf = 0, the state is returned to the “no correction” state, and the range frequency R is again set to R = R−f _prf = −f _{prf in} step S76 for the channel corresponding to the channel l.

Similarly, in the flow of the process d shown in FIG. 11, the initial value of the range frequency R is set to −f _prf in step S81, and the processes in steps S85 and S86 are “subtraction” of the processes in steps S55 and S56 in FIG. ”Is“ addition ”and“ addition ”is“ subtraction ”, the process of step S87 is“ Doppler shift frequency = measured value of Doppler shift frequency−f _prf ”or“ Doppler shift frequency = Doppler shift frequency. Except for the point that the measured value is -2f _prf ", the process is the same as the process of FIG.

When a local variation as shown in FIG. 12 occurs in the pattern as shown in FIG. 4A, the Doppler shift frequency measurement is performed for each channel corresponding to the channel j to the channel k-1 by the processing of FIG. Correction is performed by subtracting f _prf from the value, correction is not performed for each channel corresponding to channel k to channel 1-1, and Doppler shift frequency measurement is performed for each channel corresponding to channel 1 to channel m. Correction is performed by subtracting f _prf from the value. Of course, no correction is performed for each channel from channel 0 to a channel corresponding to channel j-1.

According to the processing of FIG. 11, the range frequency R is set to R = R−f _prf = −2f _{prf in} step S86 for the channel corresponding to the channel j, and then R = in step S85 for the channel corresponding to the channel k. R + f _prf = −f _prf, and the range frequency R is again set to R = R−f _prf = −2f _{prf in} step S86 for the channel corresponding to the channel l.

In this case, for each channel corresponding to channel j to channel k−1, correction is performed by subtracting 2f _prf from the Doppler shift frequency measurement value, and for each channel corresponding to channel k to channel l−1. Then, correction for subtracting f _prf from the Doppler shift frequency measurement value is performed, and correction for subtracting 2f _prf from the Doppler shift frequency measurement value is performed for each channel corresponding to channel 1 to channel m. Of course, for each channel from channel 0 to the channel corresponding to channel j-1, correction is performed by subtracting f _prf from the Doppler shift frequency measurement value.

  The pulse Doppler ultrasonic flowmeter of the present example considers local fluctuations, but does not consider the situation where the flow direction is partially reversed due to local fluctuations. The reason for determining the flow direction of the fluid to be measured in this example is to cope with a system in which the flow direction of the fluid becomes positive or negative depending on, for example, a time zone.

  That is, for example, as in the case of pumped-storage power generation, reservoirs are created above and below the power plant, and the water in the lower pond is pumped up to the upper pond using nighttime power that consumes less electricity. , When generating electricity using the head and bottom of the lower pond, the flow direction may be switched between night and day. In the case of a large-scale water purification plant, water is supplied from a water purification plant to multiple water stations, but it is necessary to adjust the amount of water between water stations depending on the downstream load. The direction may change.

Furthermore, in the piping shown in FIG.
(1) At a certain time, valves A and B are stopped and flow from valve C to valve D (FIG. 13A).
(2) Otherwise, stop valves C and D, and flow from valve A to valve B (Fig. 13B)
There is a case. At this time, at the point X of the pipe, the flow direction is reversed between the cases (1) and (2). For this reason, if the flow meter installed at the point X has a function of discriminating the flow direction, there is an advantage that measurement can be performed with one flow meter.

  According to the pulse Doppler type ultrasonic flowmeter of the present invention, even when the Doppler shift frequency exceeds the measurement range determined by the repetition frequency, it is possible to obtain a correct Doppler shift frequency distribution by performing correction processing, The measurable range of the Doppler shift frequency can be substantially larger than the measurement range determined by the repetition frequency. In particular, even if the Doppler shift frequency is a value that cannot be measured using the method of Patent Document 2, it can be measured. Furthermore, it is not only possible to deal with either positive / negative, but it can deal with both positive / negative.

Claims (10)

In the pulse Doppler ultrasonic flowmeter that measures the flow rate of the fluid to be measured flowing in the pipe using the ultrasonic Doppler shift,
An ultrasonic pulse is transmitted to the fluid to be measured at a predetermined pulse repetition frequency, the ultrasonic echo is received, and the Doppler of each channel from the first channel closest to the pipe wall of the pipe to the m-th channel in the center of the pipe A Doppler shift frequency measuring means for measuring the shift frequency;
In the Doppler shift frequency distribution from the first channel to the m-th channel obtained by the measurement result by the Doppler shift frequency measuring means, a channel range affected by aliasing is obtained, and each channel within the obtained channel range is obtained. Doppler shift frequency distribution correction means for correcting the measured value of the Doppler shift frequency of using a correction value obtained based on the measurement range of the Doppler shift frequency,
A flow rate calculating means for obtaining a flow rate of the fluid under measurement using the Doppler shift frequency distribution corrected by the Doppler shift frequency distribution correcting means;
A pulse Doppler ultrasonic flowmeter characterized by comprising: The Doppler shift frequency distribution correction means determines a channel where the aliasing occurs and determines whether or not the first channel is already affected by aliasing,
When the first channel is not affected by aliasing, the measured value of the Doppler shift frequency of each channel in the range from the first channel to the channel where the aliasing occurs is not corrected, and the occurrence of aliasing occurs. 2. The pulse Doppler method according to claim 1, wherein a measured value of a Doppler shift frequency of each channel in a range from a location channel to the m-th channel is corrected using the measured range of the Doppler shift frequency as the correction value. Ultrasonic flow meter. The Doppler shift frequency distribution correction means determines a channel where the aliasing occurs and determines whether or not the first channel is already affected by aliasing,
When the first channel is affected by the aliasing, the measured value of the Doppler shift frequency of each channel in the range from the first channel to the channel where the aliasing occurs is the measurement range of the Doppler shift frequency. The correction value is used as the correction value, and the measured value of the Doppler shift frequency of each channel in the range from the channel where the aliasing occurs to the m-th channel is set to twice the measurement range of the Doppler shift frequency. The pulse Doppler type ultrasonic flowmeter according to claim 1, wherein the pulse Doppler ultrasonic flowmeter is corrected as a correction value. The Doppler shift frequency distribution correction means determines a channel where the aliasing occurs and determines whether or not the first channel is already affected by aliasing,
When the first channel is not affected by folding and there are a plurality of places where the folding occurs,
The measured value of the Doppler shift frequency of each channel in the range from the first channel to the first channel where the aliasing occurs is not corrected,
The range from the channel at the first fold occurrence point to the m-th channel is not corrected for the range from the even-numbered fold occurrence channel to the odd-numbered fold occurrence channel. , For the range from the odd-numbered channel where the aliasing occurs to the even-numbered channel where the aliasing occurs, the measured value of the Doppler shift frequency of each channel within the range is measured The pulse Doppler ultrasonic flowmeter according to claim 1, wherein the correction value is corrected as the correction value. The Doppler shift frequency distribution correction means determines a channel where the aliasing occurs and determines whether or not the first channel is already affected by aliasing,
When the first channel is affected by aliasing and there are a plurality of aliasing locations,
Correcting the measured value of the Doppler shift frequency of each channel in the range from the first channel to the channel of the first occurrence of folding, with the measured range of the Doppler shift frequency as the correction value,
About the range from the channel where the first fold occurs to the m-th channel,
For the range from the even-numbered channel where the aliasing occurs to the odd-numbered channel where the aliasing occurs, the measured value of the Doppler shift frequency of each channel within the range is the measurement range of the Doppler shift frequency. Correct as the correction value,
For the range from the odd-numbered channel where the aliasing occurs to the even-numbered channel where the aliasing occurs, the measured value of the Doppler shift frequency of each channel within the range is the same as the measurement range of the Doppler shift frequency. 2. The pulse Doppler type ultrasonic flowmeter according to claim 1, wherein 2 times is corrected as the correction value. The Doppler shift frequency distribution correction means further determines the flow direction of the fluid to be measured,
6. The correction according to claim 1, wherein when the determined flow direction is a positive direction, the correction is a process of adding the correction value to a measured value of the Doppler shift frequency of each channel. The pulse Doppler type ultrasonic flowmeter according to any one of the above. The Doppler shift frequency distribution correction means further determines the flow direction of the fluid to be measured,
The said correction is a process which subtracts the said correction value from the measured value of the Doppler shift frequency of each said channel when this determined flow direction is a negative direction. The pulse Doppler type ultrasonic flowmeter according to any one of the above.   For each channel, an absolute value of a difference between the measured value of the Doppler shift frequency of the own channel and the measured value of the Doppler shift frequency of the adjacent channel is obtained, and a channel whose absolute value is equal to or greater than a predetermined threshold value, The pulse Doppler type ultrasonic flowmeter according to any one of claims 2 to 7, wherein the channel is a channel where the folding occurs. Determination of the flow direction of the fluid to be measured is
Accumulation of measured values of Doppler shift frequency of each channel in the range from the first channel to the channel where the aliasing occurs or the range from the first channel to the channel where the measured value of Doppler shift crosses 0 9. An addition value or an average value is calculated, and when the calculated cumulative addition value or average value is a positive value, a positive direction is set, and when it is a negative value, a negative direction is set. The pulse Doppler ultrasonic flowmeter according to any one of the above. To the computer of a pulse Doppler type ultrasonic flowmeter that measures the flow rate of the fluid to be measured flowing in the pipe using the ultrasonic Doppler shift,
An ultrasonic pulse is transmitted to the fluid to be measured at a predetermined pulse repetition frequency, the ultrasonic echo is received, and the Doppler of each channel from the first channel closest to the pipe wall of the pipe to the m-th channel in the center of the pipe A first function for measuring the shift frequency;
In the Doppler shift frequency distribution from the first channel to the m-th channel obtained by the measurement result by the first function, a channel range affected by the aliasing is obtained, and each channel within the obtained channel range is obtained. A second function for correcting the measured value of the Doppler shift frequency using a correction value obtained based on the measurement range of the Doppler shift frequency;
A third function for determining the flow rate of the fluid to be measured using the Doppler shift frequency distribution corrected by the second function;
A program to realize