CN107655533B - A kind of Ultrasonic Wave Flowmeter signal processing method and system based on backward energy integral - Google Patents
A kind of Ultrasonic Wave Flowmeter signal processing method and system based on backward energy integral Download PDFInfo
- Publication number
- CN107655533B CN107655533B CN201711043275.3A CN201711043275A CN107655533B CN 107655533 B CN107655533 B CN 107655533B CN 201711043275 A CN201711043275 A CN 201711043275A CN 107655533 B CN107655533 B CN 107655533B
- Authority
- CN
- China
- Prior art keywords
- point
- signal
- area
- value
- peak point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01F—MEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
- G01F1/00—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow
- G01F1/66—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow by measuring frequency, phase shift or propagation time of electromagnetic or other waves, e.g. using ultrasonic flowmeters
Abstract
It is proposed that the Ultrasonic Wave Flowmeter signal processing method integrated based on backward energy obtains the maximum positive peak point of backward energy signal according to the ultrasonic echo signal of sampling by comparing；Then determination section starting point forward finds integral starting point to be greater than 0.1 times of echo-signal normalization amplitude for Rule of judgment；From integral starting point to maximum positive peak point, a peak point is often found, will be normalized after the squared magnitude, calculate the trapezoidal area surrounded between adjacent peak point and added up；Judge the size of cumulative area value and preset area value；It is cumulative to stop integral after cumulative area value is more than preset area value, preset area value is subtracted by cumulative area value and obtains area difference；The ordinate of known area difference and adjacent peak point solves the intersection point of small trapezoidal upper bottom edge and X-axis, as characteristic point；It determines fair current, the adverse current propagation time of ultrasonic echo signal, and calculates gas flow.
Description
Technical field
The present invention relates to field of flow measurement, are a kind of Ultrasonic Wave Flowmeter transmitters, especially a kind of to be based on back
The Ultrasonic Wave Flowmeter signal processing method and system of wave energy integral.Using gas ultrasound wave current amount proposed by the present invention
It counts signal processing method and handles ultrasonic echo signal, realize the measurement of gas flow.
Background technique
Ultrasonic Wave Flowmeter is widely used in day due to the advantages that its measurement accuracy is high, range ratio is big, without crushing
The gas flow measurement field of the large diameter pipelines such as right gas.In measurement method, because propagation time difference method is caused by temperature change
Measurement error it is smaller, be the most frequently used, most effective measurement method.When using propagation time difference method measuring flow, it is necessary first to
It drives transmitting transducer to generate ultrasonic signal, obtained ultrasonic signal is then converted into echo letter by reception energy converter again
Number；Corresponding fair current, adverse current propagation time are calculated by the time difference between measurement transmitting and reception signal, obtains pipeline
The face velocity of cross section, and then calculate gas flow.The initial time of ultrasonic signal is known, and when echo-signal arrival
Quarter be it is unknown, need to be determined according to stable characteristic point a certain in echo-signal.Since ultrasonic signal is in gas
When propagation, energy attenuation is serious, and as the increase of gas flow rate can generate corresponding path offset, further weakens energy
Amount, so that echo-signal amplitude is faint, vulnerable to noise jamming.And the feature directly determined by straightforward procedures such as maximal peak points
Point can all generate biggish fluctuation, influence measurement accuracy.
Domestic and foreign scholars study gas ultrasonic flowmeter signal processing method, are mainly the following method.
(1) based on the method for fixed threshold
From the angle of echo-signal amplitude, the measurement method of fixed threshold be by set a fixed threshold value,
When the amplitude of ultrasonic echo signal is greater than this value, just using this moment as when the arrival of ultrasonic echo signal
Between, and acquire with this propagation time of ultrasonic echo signal.Tamotsu Kobayashi et al. (Kobayashi T.,
Shigeta K.,Fujii T.Transit-time difference type ultrasonic flowmeter[P].U.S.:
20020062690A1,2002-05-30.) describe the hardware realization of fixed threshold method, i.e., using comparator and with reference to electricity
Pressure value determines arrival time of ultrasonic echo signal.R.Bates et al. (Bates R., et al.Development of
a custom on-line ultrasonic vapour analyzer/flowmeter for the ATLAS inner
detector,with application to gaseous tracking and Cherenkov detectors[J]
.Journal of Instrumentation, 2013,8:C01002.) software realization of above-mentioned fixed threshold method is referred to,
Stop clock by starting clock while ultrasonic signal issues, and after echo-signal amplitude is more than preset value to count
Calculate the propagation time of ultrasonic signal.However, service life of noise pollution, drift or ultrasonic transducer etc. can all cause
The variation of ultrasonic echo signal amplitude obtains once ultrasonic echo signal amplitude fluctuations are larger according to fixed threshold method
The echo signal reception time will malfunction.
(2) based on the method for energy gradient
From the angle of echo-signal energy variation, Shen Ziwen et al. also proposed a kind of letter based on energy gradient
(Shen Ziwen, Xu Kejun, Fang Min wait Ultrasonic Wave Flowmeter signal processing method of the based on energy gradient for number processing method
[J] Chinese journal of scientific instrument, 2015,36 (9): 2138-2144.), pretreatment is filtered etc. to the echo-signal of sampling first,
Then each peak point of echo-signal, and the energy (energy of signal to the progress square of its amplitude, as the peak point are found
Squared magnitude with signal is in proportionate relationship, FANGM, XU K J, ZHU W J, et al.Energy transfer model
and its applications of ultrasonic gas flow-meter under static and dynamic
Flow rates. [J] .Review of Scientific Instruments, 2016,87 (1): 015107.), obtains each
The energy gradient of peak point, and envelope fitting is carried out to it, in order to reduce calculation amount, guarantee the real-time of signal processing,
The envelope of ultrasound echo signal energy is sought using the method for linear interpolation.Energy curves in the case where obtaining each flow
Afterwards, the threshold value of distinguishing characteristic wave is used for according to energy curves setting.In actual measurement, according to preset threshold value and
The energy curves for seeking echo-signal in real time, differentiate corresponding characteristic point.Since this method needs before measuring flow
Threshold parameter selection is carried out, is operated relative complex.
(3) based on the method for energy jump
Daniel company, the U.S. using detection energy jump method search characteristic point method (William Freund,
Winsor Letton,James Mc-Clellan,Baocang Jia,Anni Wey,Wen Chang.Method and
apparatus for measuring the time of flight of a signal,US patent NO.5983730,
Nov.16,1999).Due to echo-signal energy entirely be activated to propagation arrive received again during, from weak to strong, then from
Arrive by force weak, therefore backward energy change rate will first increase and reduce again, according to the threshold variations point of energy gradient, that is, can determine super
Acoustic transit time.Specifically, seeking the squared magnitude of echo-signal each point first；Then, it is sought often using sliding average
The average energy value of a bit, then the change curve for the ratio between drawing out forward and backward two o'clock average energy, i.e. energy gradient curve；Most
Afterwards, the critical point for finding energy gradient, determines ultrasonic propagation time.But the patent does not disclose the crucial skill of method
Art and seeks the time interval etc. when energy gradient between forward and backward two o'clock if the time interval of sliding window is chosen.
Summary of the invention
It is difficult to determine stable characteristic point for Ultrasonic Wave Flowmeter echo-signal, when judging the arrival of echo-signal
It carves, and the problem of instrument is required compared with high real-time, the invention proposes the gas ultrasound waves integrated based on backward energy
Flow meter signal processing method.Because the energy of signal and the squared magnitude of signal are in proportionate relationship, in order to simplify meter
It calculates, the present invention uses the energy of square characterization echo-signal of echo-signal amplitude.Using DSP and FPGA as the hardware of double-core
(Xu Kejun, Fang Min, Wang Wei, red legend is handsome, a kind of Ultrasonic Wave Flowmeter based on FPGA and DSP of Shen Ziwen, Shen on platform
Please Chinese invention patent, publication number CN104697593A, applying date 2015.06.10), write software program, real-time implementation base
In the signal processing method of backward energy integral.
The calculating step of Ultrasonic Wave Flowmeter signal processing method based on backward energy integral are as follows: basis is adopted first
The ultrasonic echo signal of sample obtains the maximum positive peak point of backward energy signal by comparing；Then determination section rises forward
Initial point finds integral starting point with 0.1 times greater than echo-signal normalization amplitude for Rule of judgment；From integral starting point (product
Point starting point is also a peak point) between maximum positive peak point, a peak point is often found, will be returned after the squared magnitude
One changes, and calculates the trapezoidal area surrounded between adjacent peak point and adds up；Judge the size of cumulative area value and preset area value；When
Cumulative area value is cumulative more than integral is stopped after preset area value, subtracts preset area value by cumulative area value and obtains difference in areas
It is worth (small trapezoidal area)；Ordinate (the small trapezoidal upper bottom of known area difference (small trapezoidal area) and adjacent peak point
Value), the intersection point (trapezoidal height) of small trapezoidal upper bottom edge and X-axis is solved, as characteristic point；Determine the suitable of ultrasonic echo signal
Stream, adverse current propagation time, and calculate gas flow.
Detailed description of the invention
Fig. 1 is the signal processing method flow chart based on backward energy integral.
Fig. 2 is 0m3The backward energy signal and its packet under backward energy signal and its envelope and different flow under/h
Winding thread figure.
Fig. 3 is determining integrating range starting point schematic diagram.
Fig. 4 is to determine to choose integral starting point schematic diagram.
Fig. 5 is the integral area schematic diagram based on backward energy integration method.
Fig. 6 is determining preset area value schematic diagram.
Fig. 7 is the feature point calculating method based on backward energy integral.
Fig. 8 is the hardware block diagram of Ultrasonic Wave Flowmeter.
Fig. 9 is gas ultrasonic flowmeter system master chip dsp software flow chart.
Specific embodiment
Fig. 1 is the signal processing method flow chart based on backward energy integral.After echo-signal such as is filtered at the processing,
Just start to calculate characteristic point.
(1) echo-signal maximum positive peak point and minimal negative peak point are found.In the echo signal data of acquisition, pass through
Point-by-point comparison obtains two peak points for meeting condition.
(2) it will be compared after two peak points square.It is in proportionate relationship according to the squared magnitude of the energy of signal and signal,
By maximum positive peak point and minimal negative peak point respectively square after compare size.To completely indicate backward energy signal, then
The amplitude for the echo-signal every bit for needing to acquire all carries out square, in this way the processing operand that will increase algorithm, influences journey
The whole efficiency of sequence.The present invention carries out the amplitude of each peak point using the ascent stage each peak point for finding echo-signal
Square processing, can not only make full use of information entrained by backward energy signal, but also the calculation amount of entire algorithm can be reduced.And
And after searching out a peak point, subsequent area integral calculating is just carried out, (cumulative area value is big when reaching preset condition
In preset area value SA) after calculating can be completed, be further reduced the repetitive cycling of calculating process.It is reduced using such process
The operand of data, helps to improve the real-time of system.
(3) backward energy signal maximum positive peak point is obtained.To determine that integral starting point is prepared, backward energy signal is most
The amplitude (ordinate) of big positive peak point uses AmaxIt indicates.
(4) integral starting point, squared magnitude normalization are determined.In the case where known maximum positive peak point, find first
The section starting point of integral finds and integrates then according to the requirement of setting (greater than 0.1 times of echo-signal normalization amplitude)
Since initial point integrate area integrating starting point.The requirement specifically set is described in detail in Fig. 3.
(5) it quadratures area.Start reference area value after finding integral starting point, specifically include 6 steps:
Quadrature area step 1: find integral starting point after first peak point.
Quadrature area step 2: to being normalized after the peak point squared magnitude.Peak point amplitude AnIt indicates, returns
Amplitude after one change is all usedIt indicates, the backward energy signal amplitude of the peak point is its square, uses An 2It indicates, n is indicated from product
Divide starting point to each peak point subscript between maximum positive peak point.Normalizing formula is
Quadrature area step 3: calculate trapezoidal area between two neighboring peak point, and add up.As shown in figure 5, adjacent
Two peak points can be surrounded with axis of abscissas one it is small trapezoidal, obtain the energy of echo-signal by the way that trapezoidal area is cumulative
Value.Two neighboring peak point (amplitude normalization) coordinate is used respectively(Xn+1,) indicate, n and n+1 are indicated
From integral starting point to each peak point subscript maximum positive peak point.Small trapezoidal area SnCalculation formula is
All small trapezoidal areas are added up, cumulative area value is obtained.The calculation formula of cumulative area S is
Quadrature area step 4: whether be greater than preset area value SAJudge cumulative area value and preset area value
Size:
If cumulative area value is not up to preset area value, normalized amplitude is (under trapezoidal after saving the peak point square
Bottom), find next peak point backward, then repeat second and third, four steps, continue to determine whether be greater than preset area value SA；
If cumulative area value has reached preset area value, start to carry out next step.
Quadrature area step 5: reference area difference S'.Cumulative area value subtracts preset area value and obtains difference in areas
Value S'.
It quadratures the step 6 of area: characteristic point abscissa is asked by area difference.The step describes in detail in Fig. 7.
(6) value of characteristic point is obtained.By the value then available fair current, adverse current propagation time of characteristic point, pipeline is obtained
The face velocity of cross section, and then calculate gas flow.With a certain sound channel, downstream propagation times tdFor, calculation formula are as follows:
td=xtIn/f (4) formula, xtFor the characteristic point abscissa value being calculated, f is the sample frequency of echo-signal；
The calculation formula of gas flow Q are as follows:
In formula,For the average speed in sound channel, S is the cross-sectional area of pipeline, and α indicates one and number of channels, sound channel
The related correction factor of factors such as distribution and Reynolds number, L indicate sound channel length, and θ indicates angle between energy converter and gas pipeline,
td、tuRespectively indicate ultrasonic wave fair current, the adverse current propagation time of same sound channel.
Fig. 2 is 0m3The backward energy signal and its packet under backward energy signal and its envelope and different flow under/h
Winding thread figure.The figure of top half is 0m3Backward energy signal and its envelope under/h, this is by by 50 groups of 0m3Returning under/h
The pretreatments such as wave signal is filtered, squared magnitude, averaging and normalization, using curve-fitting tool in MATLAB to echo
All peak points of energy signal carry out what envelope was fitted.The figure of lower half portion is the backward energy letter under different flow
Number and its envelope, by 0m3/h、200m3/h、400m3/h、600m3/h、800m3/h、1000m3Each 50 groups of echo-signals under/h
Carry out what same processing fitting obtained.It can be seen that by different flow backward energy signal envelope when flow increases,
The backward energy signal decline stage can be distorted, and (some researches show that return when detected fluid flow velocity increases to a certain range
Wave signal can generate serious distortion.SABATINI A M.A digital-signal-processing technique for
ultrasonic signal modeling and classification[J].IEEE Transactions on
Instrumentation Measurement, 2001,50 (1): 15-21.), the amplitude of echo-signal, which can be undergone, starts from scratch fastly
Speed increases, and reaches the process of slow-decay to zero after maximum value, i.e., during releasing energy, is slow and jiggly releases
It puts.But in the ascent stage of backward energy signal, since the ultrasonic wave of excited target energy converter transmitting directly acts on, substantially not
There are reflected acoustic waves etc. to influence, and energy is concentrated and stablized, thus, it can be seen that each represented by 1,2,3,4,5,6 in Fig. 2
The waveform profiles of ascent stage under a flow are basically unchanged, and with the increase of flow, the waveform profiles of ascent stage occur
Translation, but waveform profiles are substantially parallel under each flow.The energy as entrained by the ultrasonic wave emitted every time is identical
(frequency, amplitude, the phase of pumping signal are all identical), so, energy entrained by backward energy signal ascent stage is same
Be under flow it is stable, integrated by the area surrounded to the ascent stage, available stable characteristic point, and then acquire
The flow of tested gas.
Fig. 3 is determining integrating range starting point diagram.When the flow velocity of tested gas is more than certain range, i.e., in big flow
Under, the noise signal adulterated in echo-signal obviously becomes larger, it can be seen that the noise signal amplitude after normalizing in Fig. 3 is
Through more than 0.1, in Fig. 2, it is stable for having been described above the energy of same flow lower ascent stage, so the section end point of integral
◇ is assured that as maximal peak point, then, determine area start integral section starting point it is just very crucial.If section
Starting point is chosen at first point of sampling, then noise signal will be taken as echo-signal to calculate into area, not only increases
Add calculation amount, interference is introduced and is calculated, and cannot reflect the real information of echo-signal, reduces measurement accuracy.And by most
Determination section starting point can have big peak point although the position (abscissa) of peak-peak may be fluctuated forward
Effect removal noise jamming, can reduce the influence of positional fluctuation by algorithm process.Take into account big flow and small flow ascent stage
Peak value chooses maximal peak point 206 to 220 points (2048 points of each echo signal sample, 25 points of each period) forward
Position, as integrating range starting point, as shown in the o in Fig. 2.Integrate starting point ● be chosen in Fig. 4 and be described in detail.
Fig. 4 is to determine to choose integral starting point diagram.The present invention is finding area by many experiments with the method for exhaustion
Between after starting point, from section starting point to section end point (maximal peak point), setting is with the normalization of echo-signal amplitude
0.1 on the basis of (after square indicate backward energy signal, then amplitude is normalized to 0.01, the calculation process according to Fig. 1,
First find the peak point of echo-signal, then square be the peak point of backward energy signal), when normalized peak point amplitude is full
It is exactly required peak point when foot is more than or equal to 0.1, these peak points are best as subsequent areal calculation effect.Such as Fig. 4 institute
Show, in integral starting point ● the peak value between the end point ◇ of section is all the peak point for the condition that meets.It can from Fig. 4
Out, when backward energy signal normalization amplitude is 0.01, the area very little covered before the starting point o of section with X-axis, i.e.,
The noise signal of signal leading portion interferes very little, can be ignored.
Fig. 5 is the integral area schematic diagram based on backward energy integration method.The section starting point for determining area integral,
And then after finding integral starting point, specific areal calculation can be both carried out.In figure ● representative be exactly we
It finds and meets the integral starting point to impose a condition and ascent stage each peak point, including maximal peak point b shown in figure, secondary big peak
It is worth the peak point c of point a and interlude.The method that linear fit is used between the adjacent two o'clock of each peak point, structure are into a line
Section, then trapezoidal bottom edge adjacent representated by the ordinate with this two o'clock (solid line in figure), axis of abscissas collectively constitute one
It is a small trapezoidal, each small trapezoidal area is calculated and added up, while guaranteeing precision, simplifies calculating process, avoids
To calculation amount brought by signal node-by-node algorithm.To described in each small trapezoidal areal calculation such as above-mentioned formula (2).
Fig. 6 is determining preset area value schematic diagram.After integral starting point and ascent stage each peak point required for finding, I
Need to be arranged the preset value of an area, when cumulative area value is more than preset area value, just stops accumulation calculating area, open
The abscissa of the characteristic point to begin needed for solving.So, this area preset value is equally very crucial.When the area value that we choose
When smaller, corresponding peak value points are just few, i.e. for the covered area of institute the peak point c of interlude at, putting 3 is exactly our solutions
Obtained characteristic point；When the area value that we choose is larger, corresponding peak value points are just more, i.e. the covered area of institute is secondary
At big peak point a, point 1 is exactly that we solve obtained characteristic point.When backward energy signal waveform fluctuates, backward energy
Signal fluctuation causes peak point amplitude that can change, and then the cumulative area value being calculated can change, and fluctuation causes to add up
The part that area value increases or decreases needs to be compensated with the trapezoidal area of last part in cumulative area value, i.e., terraced in figure
The area value (two parts shaded area in figure) of shape S1 or trapezoidal S3 compensates because area preset value be it is certain,
Need to compensate the area value of part be it is equal, i.e., the area value of trapezoidal S1 is equal to trapezoidal S3 area value, and because trapezoidal S3
Upper bottom is significantly less than the upper bottom of trapezoidal S1, so, height (abscissa of the peak point c of interlude and the spy of trapezoidal S3
Sign point the distance between 3 abscissas) it is higher (between 1 abscissa of abscissa and characteristic point of a of second largest peak value than trapezoidal S1
Distance) it is big more, i.e., required by the obtained fluctuation of 1 abscissa of characteristic point of solution to be significantly less than the fluctuation of 3 abscissa of characteristic point, that
, the stability of 1 abscissa of characteristic point is higher than the stability of 3 abscissa of characteristic point.The stability of characteristic point is to guarantee gas
The key of flow measurement precision.So area preset value should be chosen at second largest peak value point a and maximal peak point b (shown in Fig. 5)
Between, and no more than maximal peak point b, area discrepancy brought by the peak value of big flow and small flow ascent stage is taken into account,
The area value after amplitude calculates after normalizing is chosen, determines preset area value SAWhen between 60 to 63, effect is best.
Fig. 7 is the feature point calculating method based on backward energy integral.Area preset value be chosen at second largest peak value point a and
Maximal peak point b, when cumulative area value is greater than preset area value SAWhen, stop cumulative, it is known that difference in areas (trapezoidal area) and phase
The ordinate (upper bottom value) of adjacent peak point, so that it may solve trapezoidal height.Second largest peak value point a and maximal peak point b are (such as
Shown in Fig. 5) coordinate be respectively (x1, y1), (x2, y2), wherein y1 and y2 is the width of normalized backward energy signal
Value, then, the slope k of linear fit is between this two o'clock
Cumulative area value and preset area value SABetween area difference S' be
The distance between 1 abscissa of characteristic point and maximal peak point b abscissa be x, i.e., the height of trapezoidal S', then, S''s
Area value is
Unknown number x (only one solution is eligible) can be found out according to formula (6):
Then the value of final solved 1 abscissa of characteristic point is (x2-x).
Fig. 8 is the hardware block diagram of Ultrasonic Wave Flowmeter.It mainly include primary instrument, DSP minimum system, FPGA minimum
System, driving signal generation and amplifying circuit, transmitting/receiving signal channel switching circuit, echo-signal conditioning and Acquisition Circuit,
Power management module.
The primary instrument is made of four energy converters and flowtube, and four energy converters are respectively and fixedly installed to flowtube
On；Four energy converters are respectively energy converter 1, energy converter 2, energy converter 3 and energy converter 4, set up two-channel direct-injection type structure；Transducing
Device is transceiver type, is controlled by transmitting/receiving signal channel switching circuit, and it is sharp for emitting to can not only be used for transmitting transducer
Signal is encouraged, is used for receives echo-signal but also as energy converter is received.
The FPGA minimum system is made of fpga chip, clock, reset circuit and FLASH (flash memory), wherein FPGA core
Piece model EP2C8Q208C8N.It include DAC (digital analog converter) control module, ROM (read-only memory), frequency dividing inside FPGA
Module, RAM_2SPORT.
The DSP minimum system is by dsp chip, clock, outer watchdog, FRAM (ferroelectric memory), man-machine interface, string
Port communications composition, wherein dsp chip model TMS320F28335.
The power management module provides the digital power and analog power of required amplitude for system.
Fig. 9 is gas ultrasonic flowmeter system master chip dsp software flow chart.After system electrification, DSP carries out initial
Change, completes aobvious to the distribution of GPIO (universal input/output) mouth, timer interruption initialization, liquid crystal display initialization, flowmeter
Show the initialization tasks such as initialization.DSP sends excitation initial signal to FPGA, and switches transmitting, receiving channel.Detect timer
Whether time tag position is 0, if 0, is then directly entered DSP and receives unloading initial signal wait state；If 1, then liquid is updated
Crystalline substance display, communicates with host computer, and by timer time mark position 0.If after DSP receives unloading initial signal, starting to read
The echo-signal deposited in FPGA is taken, integral area is calculated, acquires characteristic point abscissa, obtain the propagation time, seek each sound channel
Flow, and then export final flow rate Q.
Claims (4)
1. a kind of Ultrasonic Wave Flowmeter signal processing method based on backward energy integral, including primary instrument, DSP are minimum
System, FPGA minimum system, driving signal generate and amplifying circuit, transmitting/receiving signal channel switching circuit, echo-signal tune
Reason and Acquisition Circuit, power management module and the Ultrasonic Wave Flowmeter signal processing software based on backward energy integral；Its
It is characterized in that: being difficult to determine stable characteristic point for Ultrasonic Wave Flowmeter echo-signal, judge the arrival of echo-signal
Moment, and the problem of instrument is required compared with high real-time, propose the gas ultrasound wave current amount based on backward energy integral
Signal processing method is counted, because the squared magnitude of the energy of signal and signal is in proportionate relationship, calculate, make to simplify
With the energy of square characterization echo-signal of echo-signal amplitude；The committed step of this method are as follows: be filtered echo-signal
After processing, just start to calculate characteristic point；
(1) echo-signal maximum positive peak point and minimal negative peak point are found；In the echo signal data of acquisition, by point-by-point
Compare two peak points for obtaining the condition of satisfaction；
(2) it will be compared after two peak points square；It is in proportionate relationship according to the squared magnitude of the energy of signal and signal, it will most
Big positive peak point and minimal negative peak point respectively square after compare size；To completely indicate backward energy signal, then need
The amplitude of the echo-signal every bit of acquisition is all carried out square, in this way the processing operand that will increase algorithm, influences program
Whole efficiency；Using the ascent stage for finding echo-signal each peak point, a square processing is carried out to the amplitude of each peak point, both
Information entrained by backward energy signal can be made full use of, and the calculation amount of entire algorithm can be reduced；Also, it is searching out
After one peak point, subsequent area integral calculating is just carried out, when reaching preset condition, i.e., cumulative area value is greater than default face
Product value SAWhen, it completes to calculate；
(3) backward energy signal maximum positive peak point is obtained；To determine that integral starting point is prepared, backward energy signal maximum is just
The amplitude A of peak pointmaxIt indicates；
(4) integral starting point, squared magnitude normalization are determined；In the case where known maximum positive peak point, integral is found first
Section starting point be greater than echo-signal normalization 0.1 times of amplitude then according to the requirement of setting, find integral starting
Since point integrate area integrating starting point；
(5) it quadratures area；Start reference area value after finding integral starting point, specifically include 6 steps:
Quadrature area step 1: find integral starting point after first peak point；
Quadrature area step 2: to being normalized after the peak point squared magnitude；Peak point amplitude AnIt indicates, normalization
Amplitude afterwards is all usedIt indicates, the backward energy signal amplitude of the peak point is its square, uses An 2It indicates, n is indicated from integral
Initial point is to each peak point subscript between maximum positive peak point；Normalizing formula is
Quadrature area step 3: calculate trapezoidal area between two neighboring peak point, and add up；Two adjacent peak points are all
Can be surrounded with axis of abscissas one it is small trapezoidal, obtain the energy value of echo-signal by the way that trapezoidal area is cumulative；Two neighboring peak
Value point coordinate is used respectivelyWithIt indicates, n and n+1 are indicated from integral starting point to maximum positive peak point
Between each peak point subscript；Small trapezoidal area SnCalculation formula is
All small trapezoidal areas are added up, cumulative area value is obtained；The calculation formula of cumulative area S is
Quadrature area step 4: judging whether cumulative area value is greater than preset area value SA:
If cumulative area value is not up to preset area value, save normalized amplitude after the peak point square, i.e., trapezoidal bottom,
Find next peak point backward, then repeat second and third, four steps, continue to determine whether be greater than preset area value SA；
If cumulative area value has reached preset area value, start to carry out next step；
Quadrature area step 5: reference area difference S'；Cumulative area value subtracts preset area value and obtains area difference S'；
Quadrature area step 6: asking characteristic point abscissa, i.e. known trapezoidal area value and upper bottom by area difference
The position of characteristic point can be obtained in value, the height for solving it；
(6) value of characteristic point is obtained；By the value then available fair current, adverse current propagation time of characteristic point, conduit cross-sectional is obtained
The face velocity in face, and then calculate gas flow；With a certain sound channel, downstream propagation times tdFor, calculation formula are as follows:
td=xt/f (4)
In formula, xtFor the characteristic point abscissa value being calculated, f is the sample frequency of echo-signal；
The calculation formula of gas flow Q are as follows:
In formula,For the average speed in sound channel, S is the cross-sectional area of pipeline, α indicate one with number of channels, sound channel distribution and
The related correction factor of the factors such as Reynolds number, L indicate sound channel length, and θ indicates angle between energy converter and gas pipeline, td、tuPoint
Ultrasonic wave fair current, the adverse current propagation time of same sound channel are not indicated.
2. a kind of Ultrasonic Wave Flowmeter signal processing method based on backward energy integral as described in claim 1,
It is characterized in that: in the case where known maximum positive peak point, when 25 2048 points of each echo signal sample, points of each period
When, the position of maximal peak point 206 to 220 points forward is chosen, as integrating range starting point；After finding section starting point,
From section starting point to the end point of section, set on the basis of echo-signal amplitude normalized 0.1, when normalized peak
It is exactly required peak point when value point amplitude is met more than or equal to 0.1.
3. a kind of Ultrasonic Wave Flowmeter signal processing method based on backward energy integral as described in claim 1,
Be characterized in that: area preset value should be chosen between second largest peak value point a and maximal peak point b, and no more than maximal peak point
B chooses the area value after amplitude calculates after normalizing, determines preset area value SABetween 60 to 63.
4. a kind of Ultrasonic Wave Flowmeter signal processing method based on backward energy integral as described in claim 1,
When being characterized by area difference S' and seeking characteristic point abscissa, by the abscissa of characteristic point and maximal peak point b abscissa it
Between distance be set as x；The coordinate of second largest peak value point a and maximal peak point b after known normalization be respectively (x1, y1), (x2,
Y2), the abscissa of characteristic point can be solved；Wherein, the slope of a, b point-to-point transmission fitting a straight line are as follows:
The upper bottom of known trapezoidal and height, then S' are as follows:
Find out unknown number x:
Then the value of the abscissa of final solved characteristic point is (x2-x).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711043275.3A CN107655533B (en) | 2017-10-31 | 2017-10-31 | A kind of Ultrasonic Wave Flowmeter signal processing method and system based on backward energy integral |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711043275.3A CN107655533B (en) | 2017-10-31 | 2017-10-31 | A kind of Ultrasonic Wave Flowmeter signal processing method and system based on backward energy integral |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107655533A CN107655533A (en) | 2018-02-02 |
| CN107655533B true CN107655533B (en) | 2019-09-10 |
Family
ID=61095941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711043275.3A Active CN107655533B (en) | 2017-10-31 | 2017-10-31 | A kind of Ultrasonic Wave Flowmeter signal processing method and system based on backward energy integral |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107655533B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3936697A1 (en) * | 2018-11-29 | 2022-01-12 | BP Exploration Operating Company Limited | Event detection using das features with machine learning |
| CN110554378B (en) * | 2019-09-10 | 2021-05-11 | 深圳锐越微技术有限公司 | Single-channel Doppler radar radial motion direction identification method and device |
| CN112890859B (en) * | 2021-01-19 | 2024-01-09 | 广州多浦乐电子科技股份有限公司 | Data processing method and system based on ultrasonic signal feature vector |
| CN115078909B (en) * | 2022-06-16 | 2024-03-15 | 国网江苏省电力有限公司宿迁供电分公司 | Non-ideal pulse leading edge detection method and device based on energy inflection point |
| CN115655396B (en) * | 2022-12-26 | 2023-06-02 | 水发航宇星物联科技(辽宁)有限公司 | Ultrasonic wave error prevention method and ultrasonic fluid detection device |
| CN116211453A (en) * | 2023-05-09 | 2023-06-06 | 浙江伽奈维医疗科技有限公司 | Temperature control method and system for varicose vein radio frequency ablation catheter |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4670238B2 (en) * | 2003-12-10 | 2011-04-13 | パナソニック株式会社 | Gas shut-off device |
| CN102113899A (en) * | 2009-12-30 | 2011-07-06 | 东软飞利浦医疗设备系统有限责任公司 | Method for extracting two-dimensional dynamic blood vessel information based on ultrasonic echo in real time |
| CN102323592A (en) * | 2011-08-12 | 2012-01-18 | 中国科学院光电技术研究所 | Normalization method for target echo signal |
| CN106643940A (en) * | 2017-02-20 | 2017-05-10 | 重庆川仪自动化股份有限公司 | Method for computing propagation time of ultrasonic flow meters on basis of echo energy |
| CN106768109A (en) * | 2017-02-21 | 2017-05-31 | 合肥工业大学 | Ultrasonic Wave Flowmeter signal processing method based on echo ascent stage peak fitting and based on backward energy point location |
| CN106871980A (en) * | 2017-02-13 | 2017-06-20 | 合肥工业大学 | The excitation of gas ultrasonic flowmeter batch (-type) and signal processing method and system based on adjacent peak maximum difference |
-
2017
- 2017-10-31 CN CN201711043275.3A patent/CN107655533B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4670238B2 (en) * | 2003-12-10 | 2011-04-13 | パナソニック株式会社 | Gas shut-off device |
| CN102113899A (en) * | 2009-12-30 | 2011-07-06 | 东软飞利浦医疗设备系统有限责任公司 | Method for extracting two-dimensional dynamic blood vessel information based on ultrasonic echo in real time |
| CN102323592A (en) * | 2011-08-12 | 2012-01-18 | 中国科学院光电技术研究所 | Normalization method for target echo signal |
| CN106871980A (en) * | 2017-02-13 | 2017-06-20 | 合肥工业大学 | The excitation of gas ultrasonic flowmeter batch (-type) and signal processing method and system based on adjacent peak maximum difference |
| CN106643940A (en) * | 2017-02-20 | 2017-05-10 | 重庆川仪自动化股份有限公司 | Method for computing propagation time of ultrasonic flow meters on basis of echo energy |
| CN106768109A (en) * | 2017-02-21 | 2017-05-31 | 合肥工业大学 | Ultrasonic Wave Flowmeter signal processing method based on echo ascent stage peak fitting and based on backward energy point location |
Non-Patent Citations (4)
| Title |
|---|
| Design of Multi-channel Ultrasonic Flowmeter Based on ARM;Jie Yang;《International Conference on Electrical & Control Engineering》;20101231;全文 |
| 一种气体超声波流量计信号处理方法研究;汪伟;《电子测量与仪器学报》;20150930;全文 |
| 基于回波峰值拟合的气体超声流量计信号处理;田雷;《电子测量与仪器学报》;20170731;全文 |
| 基于能量变化率的气体超声波流量计信号处理方法;沈子文;《仪器仪表学报》;20150930;全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107655533A (en) | 2018-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107655533B (en) | A kind of Ultrasonic Wave Flowmeter signal processing method and system based on backward energy integral | |
| CN106768109B (en) | Based on echo ascent stage peak fitting and based on the Ultrasonic Wave Flowmeter signal processing method of backward energy point location | |
| CN106643939B (en) | The method for calculating ultrasonic propagation time for ultrasonic flowmeter | |
| CN103344288B (en) | A kind of transit-time ultrasonic flow meter measuring method analyzed based on zero point | |
| CN103776516B (en) | Ultrasonic flow metering system with upstream pressure transducer | |
| CN106768103B (en) | A kind of method of the automatic prover time deviation of ultrasonic flowmeter | |
| CN107843305B (en) | A kind of Ultrasonic Wave Flowmeter signal processing method based on echo signal envelope fitting | |
| CN107860430B (en) | Time difference measurement method of ultrasonic gas flowmeter based on time difference method | |
| MX2012009383A (en) | Flow meter validation. | |
| Tian et al. | Energy peak fitting of echo based signal processing method for ultrasonic gas flow meter | |
| CN107328447A (en) | Well head gas-liquid biphase flowmeter amount method and device | |
| Zhu et al. | Mathematical modeling of ultrasonic gas flow meter based on experimental data in three steps | |
| CN111044110A (en) | Gas ultrasonic flowmeter signal processing method based on similarity analysis | |
| Fang et al. | Development of a high-precision and wide-range ultrasonic water meter | |
| CN105606170A (en) | Ultrasonic gas metering device with self-learning template and metering method of device | |
| Chen et al. | Study on flow field and measurement characteristics of a small-bore ultrasonic gas flow meter | |
| US10416013B2 (en) | Ultrasonic flow measuring apparatus detecting propagation time period from a zero cross point when a peak waveform exceeds a trigger level | |
| CN112304376B (en) | Ultrasonic flowmeter flow measuring method based on data fusion | |
| CN113375737A (en) | Flow velocity metering method of time difference type ultrasonic gas flowmeter | |
| KR101764870B1 (en) | Signal processing system for ultrasonic floemeter | |
| CN110147567B (en) | Simulation method for obtaining internal sound field of ultrasonic flowmeter in non-ideal flow field | |
| Ma et al. | Signal processing method based on connection fitting of echo peak point with a large slope for ultrasonic gas flow meter | |
| US20140202258A1 (en) | Ultrasonic measurement device and a method for operating the same | |
| CN106643940B (en) | The method for calculating the ultrasonic flowmeter propagation time based on backward energy | |
| Li et al. | Research on transit-time ultrasonic flowmeter with signal characteristic analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |