Adaptive measurement method of ultrasonic flowmeter for unknown liquid flow

Abstract

The application discloses an ultrasonic flowmeter in the field of fluid measurement, and an adaptive measurement method for unknown liquid flow, which adopts a pi-shaped pipe, a temperature compensation table is preset in a PC, a temperature sensor measures the temperature of liquid in the pi-shaped pipe and transmits the temperature to a TDC-GP30 chip, the TDC-GP30 chip measures the upstream ultrasonic time t1 and the downstream ultrasonic time t2 of liquid passing through the pi-shaped pipe, the PC calculates the characteristic parameter c of liquid species in the pi-shaped pipe section, wherein c is the sound velocity, and p is the cross length of liquid flow line and ultrasonic wave when liquid flows through the pi-shaped pipe, temperature T and the distance l between two ultrasonic transducers are input into a trained support vector machine classifier to obtain the current liquid species, the sound velocity corresponding to the current liquid species is obtained according to the temperature compensation table, and the flow rate and the flow are calculated based on the sound velocity; the application associates the flight time measured by the TDC-GP30 with the characteristic parameters of liquid species, and excavates the relationship between the potential factors of the flight time and the characteristic parameters of liquid species, so that the measurement is accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fluid measurement, in particular to a method for measuring liquid flow by using an ultrasonic flowmeter. BACKGROUND

[0002] In the field of fluid measurement, ultrasonic flowmeters are widely used in various industrial and commercial scenarios to accurately measure the flow of fluids. Traditional ultrasonic flowmeters usually measure based on the time-of-flight principle, which uses the time difference of ultrasonic wave propagation in the liquid to calculate the flow rate. However, the speed of sound and acoustic impedance of different liquids vary greatly, so the precision and accuracy of the same ultrasonic flowmeter in different liquids will be affected. In practical applications, unknown liquids are often encountered, for example, in industrial production, the type of fluid may change due to changes in production processes. For this situation, if the same ultrasonic flowmeter and the usual time-of-flight principle are used to measure the flow of unknown liquids, the measurement result will certainly be inaccurate, so an adaptive method is needed to achieve accurate flow measurement.

[0003] The ultrasonic flowmeter disclosed in the document with the Chinese patent publication number CN109323730A and the name of a TDC-GP30-based double-channel gas ultrasonic flowmeter and a use method selects two groups of transducers through an electronic switch, respectively, a time chip transmits a pulse, the echo signal of the ultrasonic flow returns to the time chip, and the calculation result is transmitted to a single-chip microcomputer MCU for saving; the time chip measures the medium temperature of the pipeline in real time, the single-chip microcomputer MCU calculates the current temperature, the single-chip microcomputer MCU collects pressure from the pressure sensor and calculates the pipeline pressure value, the double-channel transducer is used to measure the instantaneous flow of the medium in the pipeline, and the measurement error caused by uneven distribution of the fluid in the pipeline is reduced. The method disclosed in the document with the Chinese patent publication number CN106996811A and the name of a high-accuracy intelligent liquid ultrasonic flowmeter metering method collects signals through a TDC-GP30 chip produced by the German ACAM company, the flow rate S1 of the ultrasonic flowmeter is K1*S, K1 is a correction coefficient used to correct the error caused by the inconsistency of each ultrasonic pipeline, and S is the original flow rate of the ultrasonic flowmeter; the K1 coefficient of 20 stages is divided according to the time-of-flight difference between the upstream and downstream, the time-of-flight is the propagation time of the upstream transducer and the downstream transducer of the ultrasonic instrument pipe section in the medium, the time-of-flight is determined to be in which stage, and the coefficient K1 of the stage value is obtained, K1 is obtained according to the position of the time-of-flight, and K2 is calculated according to the distance of the probe and the time-of-flight. The above two methods of measuring the flow by using the ultrasonic flowmeter are based on the measurement of the same known liquid, can only solve the measurement error caused by uneven distribution of the fluid or solve the error caused by the inconsistency of the ultrasonic pipeline, and still cannot solve the measurement of the unknown liquid flow, so the field urgently needs a self-adaptive method to realize accurate measurement of the unknown liquid flow. SUMMARY

[0004] The purpose of the present application is to solve the problem that the existing ultrasonic flowmeter cannot accurately measure the unknown liquid flow, and propose an ultrasonic flowmeter adaptive measurement method for unknown liquid flow, identify the characteristics of the liquid type, and realize accurate measurement of different liquids.

[0005] To achieve the above purpose, the technical solution adopted by the present application comprises the following steps:

[0006] Step 1): a π-shaped pipe is used, a first ultrasonic transducer is installed upstream of the π-shaped pipe, and a second ultrasonic transducer is installed downstream of the π-shaped pipe, and the two ultrasonic transducers and a TDC-GP30 chip and a master chip together constitute an ultrasonic flowmeter;

[0007] Step 2): presetting a temperature compensation table in the PC, wherein the temperature compensation table is a fitting formula of different liquid types and corresponding sound velocities, and the specific value of the sound velocity c of the liquid type at the temperature T is obtained;

[0008] Step 3): the temperature sensor measures the temperature T of the liquid in the π-shaped pipe and transmits to the TDC-GP30 chip, the TDC-GP30 chip measures the upstream flight time t1 and the downstream flight time t2 of the ultrasonic wave through the liquid in the π-shaped pipe, calculates the time difference Δt = t1-t2, and sends the temperature T, the time difference Δt, the flight time t1 and t2 to the master chip; if the liquid type is unknown, the master chip sends the liquid temperature T, the flight time t1 and t2 to the PC;

[0009] Step 4): the PC calculates the characteristic parameters of the liquid type in the π-shaped pipe segment according to the formula c is the sound velocity, l is the distance between the two ultrasonic transducers, and p is the length of the liquid flow line intersected by the ultrasonic wave when the liquid flows through the π-shaped pipe.

[0010] Step 5): inputting the temperature T and the distance l into the trained support vector machine classifier to obtain the current liquid type;

[0011] Step 6): obtaining the sound velocity corresponding to the current liquid type according to the temperature compensation table, and calculating the flow rate and the flow based on the sound velocity.

[0012] The beneficial effects of the technical scheme adopted by the present application are:

[0013] (1) The present application adopts a high-precision time measurement chip TDC-GP30 to read the time difference in the time conversion mode, measures the liquid flow by the time difference method, and simultaneously adopts a filtering algorithm for data preprocessing to ensure the measurement accuracy.

[0014] (2) The present application adopts the support vector classifier (SVC) method, associates the flight time measured by the TDC-GP30 with the characteristic parameters of the liquid type, and excavates the relationship between the potential factors of the flight time and the characteristic parameters of the liquid type, so that the measurement is accurate.

[0015] (3) The present application dynamically adjusts the parameters and the calibration model of the ultrasonic flowmeter according to the identified liquid type characteristics, so as to realize accurate measurement of different liquids. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a π-shaped pipe structure diagram;

[0017] Figure 2 is a structural principle diagram of an ultrasonic flowmeter;

[0018] ​​Figure 3 Flowchart of the measurement method of this invention. Detailed Implementation

[0019] See Figure 1 , Figure 2 and Figure 3 As shown, this invention first installs two ultrasonic transducers, T1 and T2, on a π-shaped tube. The two ultrasonic transducers T1 and T2, along with a TDC-GP30 chip and a main control chip, together form an ultrasonic flow meter. Then, the STM32F407VET6 main control chip in the ultrasonic flow meter is bidirectionally connected to a PC. A temperature compensation table is pre-set in the PC, providing information on the sound velocity c corresponding to different types of liquids at temperature T. Measurement is then initiated. The ultrasonic flow meter measures the ultrasonic flight time of the liquid passing through the π-shaped tube and preprocesses the flight time. Simultaneously, the liquid temperature T is transmitted to the STM32F407VET6 main control chip. For unknown liquid types, the liquid temperature T and flight time are sent to the PC, where the PC calculates the liquid type characteristic parameters. Temperature T and spacing l are input into a trained support vector machine (SVM) classifier. The trained SVM classifier is used to associate the liquid type. Finally, based on the temperature compensation table, a sound velocity fitting formula is selected to calculate the flow rate and volume at the current liquid temperature. Details are as follows:

[0020] Step 1: Install the ultrasonic flow meter

[0021] like Figure 1 and Figure 3 As shown, a π-shaped tube is used, through which the liquid flows. The inner diameter of the π-shaped tube is r = 20 mm. Two ultrasonic transducers, T1 and T2, are installed on opposite sides of the π-shaped tube. The selected ultrasonic transducers are 1 MHz transducers. The first ultrasonic transducer, T1, is installed upstream of the π-shaped tube, and the second ultrasonic transducer, T2, is installed downstream. The distance between the two ultrasonic transducers, T1 and T2, is l, and the length of intersection between the liquid streamline and the ultrasonic wave as the liquid flows through the π-shaped tube is p.

[0022] like Figure 2 and Figure 3As shown, both ultrasonic transducers T1 and T2 are bidirectionally connected with a high-precision time measurement chip, i.e., a TDC-GP30 chip, the TDC-GP30 chip is bidirectionally connected with an STM32F407VET6 master control chip, the STM32F407VET6 master control chip is used for data preprocessing and processing, and liquid flow is calculated. The TDC-GP30 chip is also connected with a temperature sensor, the temperature sensor is connected on the π-shaped pipe, the temperature sensor adopts a PT1000 sensor, is used for measuring the temperature T of the liquid in the π-shaped pipe, and transmits the temperature T to the TDC-GP30 chip. The ultrasonic flowmeter is composed of the two ultrasonic transducers T1 and T2, the TDC-GP30 chip and the master control chip, so that the ultrasonic flowmeter is installed on the π-shaped pipe.

[0023] Step two: preset temperature compensation table

[0024] The STM32F407VET6 master control chip is bidirectionally connected with a PC. The temperature compensation table is preset in the PC, as shown in Table 1 below:

[0025] Table 1 Temperature compensation table

[0026] Liquid species Sound velocity fitting formula (c = c0+ λ1T + λ2T 2 + λ3T 3 +...) Water [c = 1403.7 + 4.796T - 0.0462T 2 + 0.0001T 3 ]]> Ethylene glycol <![CDATA[c=1658.0+25.97T-0.9441T 2 +0.01135T 3 ]]> … …

[0027] Different liquid types and corresponding sound speed fitting formulas (c=c0+λ1T+λ2T 2 +λ3T 3 +...) are set in the temperature compensation table, wherein c0 represents the sound speed of the liquid at 0 degrees Celsius, λ1, λ2 and λ3 represent coefficients in the fitting formula, and the specific value of the sound speed c of the liquid at temperature T can be obtained from the known liquid type.

[0028] Step three: measure the time of flight and temperature

[0029] The TDC-GP30 chip measures the ultrasonic time of flight of the liquid passing through the π-shaped pipe by using the time difference method, and the calculation formula of the time of flight is obtained in the time conversion mode as follows:

[0030]

[0031] Wherein, t1 is the upstream time of flight, t2 is the downstream time of flight, v is the flow rate, c is the sound speed, and the sound speed c is unknown at this time. The purpose is to obtain the liquid type characteristic parameter in the following formula (2) through the principle of the calculation formula (1) of the time of flight. δ is a disturbance term, mainly generated by the l-p section of the liquid flow.

[0032] Then, the time difference At = t1-t2 is calculated, and the TDC-GP30 chip sends the time difference At, t1, t2 to the STM32F407VET6 master chip through SPI.

[0033] At the same time, the temperature sensor is used to measure the liquid temperature, and the liquid temperature T is obtained and sent to the STM32F407VET6 master chip through SPI.

[0034] Step four: preprocessing of time of flight

[0035] The STM32F407VET6 master chip preprocesses the time of flight t1, t2 data, and uses first-order Kalman filtering to preprocess the obtained time of flight t1, t2, to reduce the interference of Gaussian noise on the signal.

[0036] Step five: correlation of liquid type

[0037] The master chip judges whether the liquid type is known, if the liquid type is known, it can be directly searched, then the correlation is skipped, and the PC directly obtains the sound speed corresponding to the current liquid type according to the temperature compensation table 1, and calculates the flow rate and flow. If the liquid type is unknown, the liquid temperature T and the time of flight t1, t2 are sent to the PC, and the liquid type characteristic parameter is calculated in the PC, ignoring the disturbance term in equation (1), and the following equation can be obtained:

[0038]

[0039] Thus, the characteristic parameter of the liquid type in the π-shaped pipe section is obtained Because of the disturbance term, the sound speed obtained at this time is biased. Therefore, through the characteristic parameter The temperature T and the transducer spacing l are used to match the liquid type using the trained support vector machine classifier (SVC), so when the liquid type is obtained, the current liquid accurate sound speed c can be obtained by looking up table 1.

[0040] After the correlation is completed, return to step three to perform time of flight and temperature measurement.

[0041] Step six: calculate flow

[0042] After obtaining the current liquid type, start the ultrasonic flowmeter to measure the time of flight and temperature.

[0043] The PC selects the sound speed c fitting formula based on the pre-designed temperature compensation table 1, calculates the sound speed c under the current liquid temperature T, and then calculates the flow rate v according to the sound speed c:

[0044]

[0045] The time difference Δt=t1-t2, the sound speed c and the liquid flow line when the liquid flows through the π type pipe and the ultrasonic wave cross length are substituted into the formula (3) to obtain the liquid flow rate v.

[0046] Finally, the flow rate Q is calculated according to the flow rate v:

[0047] Q=KvA (4)

[0048] K is the correction coefficient K, which converts the linear flow rate into the surface flow rate, A is the cross-sectional area in the π type pipe, the coefficient K, the flow rate v, the cross-sectional area A in the pipe are substituted into the formula (4), and the measurement of the flow rate of the ultrasonic flowmeter in the unknown liquid is completed.

[0049] Step seven: return to step three, perform the time of flight and temperature measurement, and continue to perform step six, and measure in this way.

Claims

1. An adaptive measurement method of unknown liquid flow rate by an ultrasonic flow meter, characterized by The method comprises the following steps: Step 1): a first ultrasonic transducer is installed upstream of the π-shaped pipe, and a second ultrasonic transducer is installed downstream of the π-shaped pipe, and the two ultrasonic transducers, a TDC-GP30 chip and a master chip together constitute an ultrasonic flowmeter; Step 2): a temperature compensation table is preset in a PC, and the temperature compensation table is a fitting formula of the sound velocity corresponding to different liquid types; the specific value of the sound velocity c of a known liquid type at a temperature T is obtained; Step 3): a temperature sensor measures the temperature T of the liquid in the π-shaped pipe and transmits the temperature T to the TDC-GP30 chip, the TDC-GP30 chip measures the upstream flight time t1 and the downstream flight time t2 of the ultrasonic wave of the liquid passing through the π-shaped pipe, calculates the time difference Δt=t1-t2, and sends the temperature T, the time difference Δt, the flight time t1 and t2 to the master chip; if the liquid type is unknown, the master chip sends the liquid temperature T and the flight time t1 and t2 to the PC; Step 4): PC calculates the characteristic parameters of liquid species in the π-type pipe segment according to the formula The characteristic parameters of liquid species in the π-type pipe segment are calculated c is the sound velocity, l is the distance between the two ultrasonic transducers, and p is the length of the liquid flow line intersecting the ultrasonic waves when the liquid flows through the π-type pipe. Step 5): inputting the temperature T and the distance l into the trained support vector machine classifier to obtain the current liquid category; temperature T and distance l are input into the trained support vector machine classifier to obtain the current liquid category; Step 6): the sound velocity corresponding to the current liquid type is obtained according to the temperature compensation table, and the flow rate and the flow are calculated based on the sound velocity.

2. The ultrasonic flow meter adaptive measurement method for unknown liquid flow rate according to claim 1, wherein: The temperature compensation table is:

3. The ultrasonic flow meter adaptive measurement method for unknown liquid flow rate according to claim 1, wherein: The upstream time of flight The upstream time of flight v is the flow velocity, c is the sound velocity, and δ is the disturbance term resulting from the l-p section of liquid flow.

4. The ultrasonic flow meter adaptive measurement method for unknown liquid flow rate according to claim 1, wherein: The master chip pre-processes the flight time t1 and t2 by using a first-order Kalman filter.

5. The ultrasonic flow meter adaptive measurement method for unknown liquid flow rate according to claim 1, wherein: The master chip first judges whether the liquid type is known, and if so, the PC directly executes step 6).

6. The ultrasonic flow meter adaptive measurement method for unknown liquid flow rate according to claim 1, wherein: According to the formula The flow rate v is calculated.

7. The ultrasonic flow meter adaptive measurement method for unknown liquid flow rate according to claim 6, wherein: The flow Q is calculated according to the formula Q=KvA, K is a correction coefficient K, the linear flow rate is converted into the surface flow rate, and A is the cross-sectional area of the π-shaped pipe.

8. The ultrasonic flow meter adaptive measurement method for unknown liquid flow rate according to claim 1, wherein: The inner diameter r of the π-shaped pipe is 20 mm, a 1MHZ ultrasonic transducer is selected, and a PT1000 temperature sensor is used.

Images