A method for determining the pole of an echo signal of a gas ultrasonic flowmeter based on correlation blur distribution of ADC

Abstract

The application discloses a kind of based on the method for determining gas ultrasonic flowmeter echo signal pole of correlation fuzzy distribution of ADC, the echo signal is collected by the high-speed ADC of hardware circuit, and these data are screened and data processing, it is characterized by: the specific operation process of the method is as follows: S1.Record the complete data of ADC sampling: utilize data rule to filter out all data between two adjacent minimum values in this measurement data, and arrange the number of continuous data of each measurement.The present application uses the trend method of Gaussian fuzzy and substitution function in this range, and increases low-pass filter to suppress noise, which well avoids the random error of single-point sampling, better reflects the frequency and amplitude characteristics of the received signal, and greatly improves the accuracy of subsequent signal trend analysis and measurement precision of ultrasonic flowmeter.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of flow measurement, in particular to a method for determining echo signal pole points of a gas ultrasonic flowmeter based on correlation fuzzy distribution of ADC. BACKGROUND

[0002] The ultrasonic sensor used by the ultrasonic flowmeter is a transducer capable of emitting and receiving signals. In general, the sensor receives a pulse trigger signal, and then generates a set of corresponding signals according to the frequency and amplitude of the pulse signal for propagation in the medium. When the sensor receives the echo signal propagated in the medium, the signal is then collected by the ADC sampling circuit and data processing is performed to obtain relevant information and make correct judgments.

[0003] Due to the influence of pressure, temperature, humidity and regulating valves on the ultrasonic signal during propagation, the gas density changes, the flow field changes, the signal fluctuates and is even submerged by noise. In actual application, the data obtained by direct measurement of the ADC is inevitably affected by many external factors and is prone to misplacement. Therefore, the collection of the measurement points has randomness, which leads to objective uncertainty in accurately capturing the pole points. The application provides a method for determining echo signal pole points of a gas ultrasonic flowmeter based on correlation fuzzy distribution of ADC. SUMMARY

[0004] The application aims to provide a method for determining echo signal pole points of a gas ultrasonic flowmeter based on correlation fuzzy distribution of ADC, to solve the problem that the data obtained by direct measurement of the ADC is inevitably affected by many external factors and is prone to misplacement, resulting in randomness in the collection of the measurement points and leading to inaccurate capture of the pole points.

[0005] To achieve the above-mentioned purpose, the application provides the following technical scheme:

[0006] A method for determining echo signal pole points of a gas ultrasonic flowmeter based on correlation fuzzy distribution of ADC, which collects echo signals through a high-speed ADC of a hardware circuit, and screens and processes the data. The specific operation process of the method is as follows:

[0007] S1. Record the complete data of ADC sampling: use data rules to screen all data between two adjacent minimum values in the measurement data, and arrange the number of continuous data in each measurement;

[0008] S2. Process the collected data: establish a statistical model for the measurement data, and normalize all the data;

[0009] S3. Calculate the average value: in the model of data screening, the local peak value screening function can accurately find the local peak value, select a group of data on the single side, and calculate the arithmetic mean of the sampling data of the single side rising data (after normalization) by using the mean function

[0010] S4. Calculate the standard deviation sigma: calculate the standard deviation value of the data after normalization according to the standard deviation formula;

[0011] S5. Calculate the Gaussian function value: according to the function formula, the height of the data and the Gaussian blur one-dimensional function graph is highly related;

[0012] S6. Draw a fitting curve: the fitting curve is used to replace the graph in the local range;

[0013] S7. Calculate the extreme value: according to the extreme value of the fitting curve, the maximum value and the minimum value are obtained respectively through the curve formula or through the signal change trend;

[0014] S8. Select the signal pole: according to the change trend of the fitting curve interval, the corresponding maximum value is correctly selected as the pole of the replacement sampling signal;

[0015] S9. Pole verification: according to the local peak value, the pole is verified;

[0016] S10. Data comparison: compare and analyze the ADC sampling value and the pole amplitude data.

[0017] Preferably, in S1, the measurement of the number of continuous data adopts a periodic function ((T1, T2, T3…Tn)) for recording, and the number of measured continuous data is ≥8 groups.

[0018] Preferably, in S5, the calculated Gaussian function value is arranged in sequence, and the arrangement sequence corresponds to the ADC data acquisition sequence.

[0019] Preferably, in S7, the fitting curve includes an extreme value curve and a non-extreme value curve, the extreme value curve calculates the maximum value and the minimum value through the formula, and the non-extreme value curve obtains the maximum value and the minimum value by analyzing the specific change trend of the signal.

[0020] Compared with the prior art, the beneficial effects of the present application are:

[0021] The present application adopts the trend method of Gaussian blur and replacement function in this range, and increases low-pass filtering to suppress noise, which well avoids the random error of single-point sampling, better reflects the frequency and amplitude characteristics of the received signal, and greatly improves the signal trend analysis accuracy and measurement precision of the subsequent ultrasonic flowmeter. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 Echo signal diagram for ADC sampling data in the application;

[0023] Figure 2 Periodic pattern and Gaussian blur wide correlation diagram in the application;

[0024] Figure 3 Echo signal pole diagram in the application;

[0025] Figure 4 Echo signal T3 single side up diagram in the application;

[0026] Figure 5 Bar chart generated by fitting curve in the application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work are within the protection scope of the application.

[0028] EMBODIMENT

[0029] Please refer to Figures 1-5 The application provides a technical solution:

[0030] A method for determining the echo signal pole of a gas ultrasonic flowmeter based on the correlation blur distribution of ADC, which collects echo signals through high-speed ADC of hardware circuit, and screens and processes the data. The specific operation process of the method is as follows:

[0031] S1. Record the complete data of ADC sampling: use data rules to screen all data between two adjacent minimum values in the measurement data, and arrange the number of continuous data in each measurement, such as Figure 1 As shown in the figure, if the continuous data is ≥8 groups, it is the effective data of this measurement, otherwise it is considered as invalid data disturbed and does not enter the final data statistical sample space. This method realizes low-pass filtering to suppress noise, and can effectively eliminate some obviously abnormal data.

[0032] S2. Process the collected data: For the measured data, a statistical model is established, and the correctness and order of the data arrangement must be ensured. In the data screening model, the local peak value screening function can accurately find the local peak value, which is not necessarily the final calculated extreme point, but the measurement value closest to the extreme point, so it can be used as a verification and reference for subsequent calculation results. Then, the data is normalized;

[0033] S3. Calculate the average value: In S2, because the main problem to be solved is the uplink maximum value, only one set of data with unilateral uplink is needed. As shown in the box part of the T3 period echo signal, the data with unilateral downlink can be left for subsequent processing. The arithmetic mean value of the sample data in the unilateral uplink data is calculated using the mean function Figure 4

[0034] S4. Calculate the standard deviation σ: Using the data sample obtained after normalization, the standard deviation value of the data set is calculated according to the standard deviation formula:

[0035] S5. Calculate the Gaussian function value: Using the high correlation between data and Gaussian blur one-dimensional function graph, according to the function formula: Solve, since each calculation takes the current calculation point as the origin, μ is equal to 0, and the function is further simplified to: The Gaussian distribution value at this time is calculated, and the value is arranged in order according to the order of collection, and the position of any data cannot be changed to ensure that the arrangement order of the data corresponds to the ADC collection data order;

[0036] S6. Draw the fitting curve: Since the solution of the Gaussian function is relatively cumbersome, it is not convenient to use, so the fitting curve is used instead of the graph in the local range. According to the normal distribution result calculated in S5, a more intuitive column chart is generated, as shown in Figure 5 On this basis, the fitting line is drawn using the fitting function tool: f(x) = ax 3 +bx 2 +cx+d(a≠0);

[0037] S7. Calculate the extreme value: According to the extreme value of the fitting curve, the maximum and minimum values are obtained respectively through the curve formula or through the signal change trend,

[0038] ① In the case of extreme value of the fitting curve, the maximum and minimum values are calculated according to the curve formula in S6, and the specific method is as follows:

[0039] a>0,

[0040] Δ = 4b 2 ​​-12ac≤0, no extreme value, no need to calculate,

[0041] Δ=4b 2 -12ac>0, extreme value, need to calculate;

[0042] a<0,

[0043] Δ=4b 2 -12ac≤0, no extreme value, no need to calculate,

[0044] Δ=4b 2 -12ac>0, extreme value, need to calculate;

[0045] When the fitting curve has no extreme value, the specific change trend of the signal needs to be further analyzed, and a corresponding judgment is made. According to the change trend, the extreme value of the closest extreme point is analyzed.

[0046] S8. Selecting the signal extreme point: according to the maximum and minimum values of the fitting curve function, and combining the change trend of the curve in this interval, the corresponding maximum value is correctly selected as the extreme point of the replacement sampling signal, and the minimum value can be recorded for subsequent data validity verification;

[0047] S9. Extreme point verification: according to the local peak value in S2, the above extreme point is verified, and it is determined as the extreme point if it meets a certain error range;

[0048] S10. Data comparison: comparing and analyzing the ADC sampling value and the extreme point amplitude data.

[0049] According to the steps S1-S10, the following is a comparison and analysis of the ADC sampling values and the extreme point calculation values under the conditions of flow rates of 0 m / s, 5 m / s, and 10 m / s. The specific data are shown in the following table:

[0050] Unit: mV

[0051]

[0052]

[0053] According to the data in the above table, the extreme point amplitude fluctuation range determined and calculated by the method is significantly smaller than the direct ADC sampling value, which shows that compared with the existing technology of directly using the sampling value as the extreme point, the present application uses the trend method of Gaussian blur and replacement function in this range, and increases the low-pass filter to suppress noise, which well avoids the random error of single-point sampling, better reflects the frequency and amplitude characteristics of the received signal, and greatly improves the accuracy, stability and measurement precision of the subsequent signal trend analysis of the ultrasonic flowmeter.

[0054] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. A method for determining the pole of the echo signal of a gas ultrasonic flowmeter based on the correlation blur distribution of the ADC, which collects the echo signal through the high-speed ADC of the hardware circuit, and filters and processes the data, characterized in that: The specific operation process of the method is as follows: S1. Record the complete data of ADC sampling: use the data rule to screen out all the data between the two adjacent minimum values in the measurement data, and arrange the number of continuous data for each measurement; S2. Process the collected data: establish a statistical model for the measurement data, and normalize all the data; S3. Calculate the average value: in the model of data screening, with the local peak screening function, the local peak can be accurately found, and a set of single-side rising data is selected. The arithmetic mean of the single-side rising data (after normalization processing) is calculated by the mean function S4. Calculate the standard deviation σ: calculate the standard deviation value of the data according to the standard deviation formula after normalization processing; S5. Calculate the Gaussian function value: use the high correlation between data and Gaussian fuzzy one-dimensional function graph to solve according to the function formula; S6. Draw the fitting curve: replace the graph in the local range with the fitting curve; S7. Calculate the extreme value: according to the extreme value of the fitting curve, the maximum and minimum values are obtained respectively through the curve formula or the signal change trend; S8. Select the signal pole: combine the change trend of the fitting curve interval to correctly select the corresponding maximum value as the pole of the replaced sampling signal; S9. Pole verification: verify the pole according to the local peak value; S10. Data comparison: compare and analyze the ADC sampling value and the pole amplitude data.

2. A method of determining the pole of the echo signal of a gas ultrasonic flow meter based on the correlation blur distribution of the ADC according to claim 1, characterized in that: In S1, the number of continuous measurement data is recorded by using periodic function ((T1, T2, T3…Tn)), and the number of continuous measurement data is ≥8 groups.

3. The method of claim 1, wherein the method is based on the correlation of the ADC-based blur distribution to determine the pole of the echo signal of the gas ultrasonic flow meter. In S5, the calculated Gaussian function value is arranged in sequence, and the arrangement order is consistent with the ADC data collection order.

4. The method of claim 1, wherein the method is based on the correlation of the ADC-based blur distribution to determine the pole of the echo signal of the gas ultrasonic flow meter. In S7, the fitting curve includes the extreme value curve and the non-extreme value curve. The extreme value curve calculates the maximum and minimum values by formula, and the non-extreme value curve obtains the maximum and minimum values by analyzing the specific change trend of the signal.

Images