Ultrasonic water meter flow compensation coefficient calculation method

Abstract

This invention discloses a method for calculating the flow compensation coefficient of ultrasonic water meters, including: acquiring a raw dataset of multiple ultrasonic water meters, removing outliers, obtaining clusters of primary and non-primary water meters, obtaining flow compensation function curves, compensation, and flow calculation. In this scheme, the raw dataset is preprocessed to remove outliers to ensure subsequent clustering effectiveness. Water meters are divided into two categories based on the resonant and anti-resonant frequencies of the transducers. Primary water meters with high consistency are selected through clustering and directly use the default flow compensation coefficient, ensuring accuracy without experimental calibration and improving calibration efficiency. Non-primary water meters undergo experimental calibration using an adaptive flow compensation scheme. Different flow compensation functions are matched to the water meter error performance to ensure accuracy and reduce production defect rates.

Description

Technical Field

[0001] This invention relates to the technical field of flow calibration, and in particular to a method for calculating the flow compensation coefficient of an ultrasonic water meter. Background Technology

[0002] Currently, ultrasonic water meters are limited by the manufacturing process, resulting in inconsistencies in the performance of transducers, flow field structure, and hardware circuit characteristics among different water meters. In actual production, each water meter is calibrated through experiments. However, calibrating all water meters experimentally reduces the production efficiency of water meters. Furthermore, some water meters still fail to meet the accuracy requirements under the current calibration scheme. This is mainly because the characteristics of water meters vary greatly, and a single flow compensation function cannot meet the requirements. Summary of the Invention

[0003] In view of this, the present invention provides a method for calculating the flow compensation coefficient of ultrasonic water meters. This method involves preprocessing the original dataset to remove outliers, ensuring effective subsequent clustering. Water meters are divided into two categories based on the transducer's resonant and anti-resonant frequencies. The main category of water meters, characterized by high consistency, is selected through clustering and directly uses the default flow compensation coefficient, ensuring accuracy without experimental calibration and improving calibration efficiency. Non-main category water meters undergo experimental calibration using an adaptive flow compensation scheme. Different flow compensation functions are matched to the water meter's error performance to ensure accuracy and reduce production defect rates.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A method for calculating the flow compensation coefficient of an ultrasonic water meter includes the following steps: S100: Obtain the raw data set from multiple ultrasonic water meters; S200, Outlier Removal: Data preprocessing of the original dataset; S300. Obtain the main class water meter cluster and non-main class water meter cluster: Apply clustering algorithm to perform feature clustering on the resonant frequency and anti-resonant frequency in the original dataset; S400. Obtain the flow compensation function curve: The flow compensation coefficient is generated by summing the propagation time difference of the upstream and downstream received signals in the original data of the main type water meter and the measured flow rate. The formula is as follows: , Where i is the water meter index representing different water meters, and j is the flow index representing different flow rates; The propagation time difference of the received signals upstream and downstream. This is the actual measured flow rate; Obtain a scatter plot of the flow compensation coefficients and fit it to the default temperature compensation function. The flow compensation function has the following form: , Where a, b, and c are the formula fitting coefficients; S500, Compensation: When the target water meter belongs to a non-main category water meter cluster, the propagation time difference of the upstream and downstream received signals and the measured flow are collected under three flow states: Q2, Q3, and 0.1*(Q2+Q3). ​​Following step S400, a flow compensation function is fitted, and flow verification is performed. The verification error calculation results are as follows. , Where j is the traffic index; When the verification error exceeds three-quarters of the maximum permissible error, a piecewise linear fitting method is adopted. The flow rate interval is divided into several sub-flow rate intervals by selecting flow rate segment points, and then linear fitting is performed segment by segment. The fitting formula for each segment is as follows: ; S600, Flow Calculation: The formula is as follows , In the formula, It is the instantaneous flow rate, and Δt is the propagation time difference of the upstream and downstream received signals under the current flow rate.

[0005] Preferably, in step S100, the original data set includes the transducer resonant frequency and anti-resonant frequency, the propagation time difference of the upstream and downstream received signals, and the measured flow rate.

[0006] Preferably, in step S200, the data preprocessing of the original dataset includes sorting the resonant frequencies and anti-resonant frequencies in the dataset from smallest to largest, extracting the middle 50% of the data to calculate a normal distribution, and identifying the remaining data points that fall outside three times the standard deviation of the mean as outliers.

[0007] Preferably, in step S400, j = (1, 2, 3, 4, 5, 6, 7, 8, 9, 10), representing flow rates of 5, 10, 16, 35, 65, 110, 160, 360, 560, and 2500 L / h, respectively.

[0008] Preferably, in step S400, obtaining the scatter plot of the flow compensation coefficient through the propagation time difference of the upstream and downstream received signals includes taking the logarithm of the propagation time difference of the upstream and downstream received signals, as shown in the formula: .

[0009] Preferably, in step S500, when the target water meter belongs to the main category water meter cluster, the default flow compensation function is directly called. Perform traffic compensation.

[0010] Preferably, in step S500, when the verification error is within three-quarters of the maximum permissible error, the current flow compensation function is considered usable.

[0011] Preferably, in step S500, the selection principle for the flow segmentation point is as follows: Set initial flow segment points, including Q2, Q3, and 0.1*(Q2+Q3).

[0012] Optimize the selection of flow segment points based on relative error RD: If the RD after linear fitting of the j-th sub-flow interval j If the error exceeds three-quarters of the maximum permissible error, an additional flow point is added at the middle of the sub-flow interval, dividing it into two sub-flow intervals. If the RD after linear fitting of the j-th flow interval j If the error is less than or equal to three-quarters of the maximum permissible error, then the selected flow range is correct. Continue optimizing other sub-flow ranges until the RD of all sub-flow ranges is within the range. j Continue until all requirements are met.

[0013] As can be seen from the above technical solution, the ultrasonic water meter flow compensation coefficient calculation method provided by the present invention preprocesses the original dataset to remove outliers to ensure the subsequent clustering effect; it divides the water meters into two categories based on the resonant frequency and anti-resonant frequency of the transducer, selects the main category water meters with high consistency through clustering, and directly uses the default flow compensation coefficient, which can ensure the accuracy requirements without experimental calibration, thus improving calibration efficiency; the non-main category water meters are experimentally calibrated through a calibration scheme, which adopts an adaptive flow compensation scheme, matching different flow compensation functions based on the water meter error performance, ensuring the accuracy of the water meter and reducing the production defect rate. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating a method for calculating the flow compensation coefficient of an ultrasonic water meter according to an exemplary embodiment; Figure 2 This is a scatter plot of the flow compensation coefficients in step S400 according to an exemplary embodiment; Figure 3 The flow compensation function curve in step S400 is shown according to an exemplary embodiment. Detailed Implementation

[0016] This invention discloses a method for calculating the flow compensation coefficient of ultrasonic water meters. It preprocesses the original dataset to remove outliers, ensuring effective subsequent clustering. Water meters are divided into two categories based on the transducer's resonant and anti-resonant frequencies. The main category of water meters, characterized by high consistency, is selected through clustering and uses the default flow compensation coefficient directly, guaranteeing accuracy without experimental calibration and improving calibration efficiency. Non-main category water meters undergo experimental calibration using an adaptive flow compensation scheme. Different flow compensation functions are matched to the water meter's error performance to ensure accuracy and reduce production defect rates.

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] This disclosure provides an exemplary embodiment of a method for calculating the flow compensation coefficient of an ultrasonic water meter, such as... Figure 1 As shown, Figure 1 This is a flowchart illustrating a method for calculating the flow compensation coefficient of an ultrasonic water meter according to an exemplary embodiment; Figure 2 This is a scatter plot of the flow compensation coefficients in step S400 according to an exemplary embodiment; Figure 3 This is the flow compensation function curve in step S400 as illustrated in an exemplary embodiment. The following is in conjunction with... Figures 1 to 3 To explain.

[0019] The specific embodiments described below are intended to help those skilled in the art understand this embodiment, but this embodiment is not limited to the specific embodiments described below.

[0020] Reference Figure 1 This disclosure provides an exemplary embodiment of a method for calculating the flow compensation coefficient of an ultrasonic water meter, which includes the following steps: Step S100: Obtain the raw data set of multiple ultrasonic water meters; Step S200: Remove outliers: Perform data preprocessing on the original dataset; Step S300: Obtain the main water meter cluster and non-main water meter cluster: Apply a clustering algorithm to perform feature clustering on the resonant frequency and anti-resonant frequency in the original dataset; Step S400: Obtain the flow compensation function curve: Generate the flow compensation coefficient by summing the propagation time difference of the received signals from upstream and downstream in the original data of the main type water meter and the measured flow rate. The formula is as follows: , Where i is the water meter index representing different water meters, and j is the flow index representing different flow rates; The propagation time difference of the received signals upstream and downstream. This is the actual measured flow rate; Obtain a scatter plot of the flow compensation coefficients and fit it to the default temperature compensation function. The flow compensation function has the following form: , Where a, b, and c are the formula fitting coefficients; Step S500, Compensation: When the target water meter belongs to a non-main category water meter cluster, the propagation time difference of the upstream and downstream received signals and the measured flow rate are collected under three flow conditions: Q2, Q3, and 0.1*(Q2+Q3). ​​Following step S400, a flow compensation function is fitted, and flow rate verification is performed. The verification error calculation results are as follows. , Where j is the traffic index; When the verification error exceeds three-quarters of the maximum permissible error, a piecewise linear fitting method is adopted. The flow interval is divided into several sub-flow intervals by selecting flow interval points, and then linear fitting is performed piecewise. The fitting formula for each segment is as follows: ; Step S600, Flow Calculation: The formula is as follows , In the formula, It is the instantaneous flow rate, and Δt is the propagation time difference of the upstream and downstream received signals under the current flow rate.

[0021] For example, refer to Figures 1 to 3 Step S100, obtaining the raw data set of multiple ultrasonic water meters, specifically includes: obtaining the raw data set of multiple ultrasonic water meters, including the transducer resonant frequency and anti-resonant frequency, the propagation time difference of the upstream and downstream received signals, and the measured flow rate.

[0022] Step S200: Remove outliers: Data preprocessing of the original dataset specifically includes: preprocessing the original dataset to remove outliers to ensure the subsequent clustering effect.

[0023] The data preprocessing method specifically includes: The resonant frequencies of the water meter transducers conform to a normal distribution. The resonant frequencies in the dataset are sorted from smallest to largest. The middle 50% of the data are extracted and their normal distribution is calculated. The remaining data points falling outside three standard deviations of the mean are considered outliers. The anti-resonant frequencies undergo the same preprocessing operation.

[0024] Step S300: Apply a clustering algorithm to perform feature clustering on the resonant frequencies and anti-resonant frequencies in the original dataset, and divide them into main class water meter clusters and non-main class water meter clusters.

[0025] Step S400, obtaining the flow compensation function curve specifically includes: calculating the propagation time difference of upstream and downstream received signals from the original data set of the main type of water meter. Actual flow rate Generate the traffic compensation coefficient. The formula for calculating the traffic compensation coefficient is as follows. , i represents the water meter index, and the time difference of signal propagation between upstream and downstream of different water meters is not consistent; j represents the flow index, j=(1,2,3,4,5,6,7,8,9,10), which represent flow rates of 5, 10, 16, 35, 65, 110, 160, 360, 560, and 2500 L / h, respectively. The actual measured flow rates of different water meters are consistent.

[0026] Then, take the logarithm of the propagation time difference between the upstream and downstream received signals: , Obtain a scatter plot of the flow compensation coefficients, such as Figure 2 As shown.

[0027] Then, the default temperature compensation function is fitted, and the flow compensation function takes the following form. , in, It is the result of taking the logarithm of the propagation time difference of the received signals upstream and downstream, and the fitting coefficients are a, b, and c.

[0028] The flow compensation function curve is as follows Figure 3 As shown.

[0029] Step S500, the compensation specifically includes: when the target water meter belongs to the main type water meter cluster, the default flow compensation function is directly called to perform flow compensation.

[0030] When the target water meter belongs to a non-primary water meter cluster, the flow compensation function is generated under actual flow conditions. For the three flow rates Q2, Q3, and 0.1*(Q2+Q3), the propagation time difference of the upstream and downstream received signals and the measured flow rate are collected, and the flow compensation function is fitted according to the above step S400.

[0031] The flow rate was then verified, and the verification error was calculated as follows. , Where j represents the traffic index.

[0032] If the error is within three-quarters of the maximum permissible error, the current flow compensation function is considered usable.

[0033] If the error exceeds three-quarters of the maximum permissible error, the current flow compensation function is considered unusable. In this case, a piecewise linear fitting method can be used. The entire flow interval is divided into several sub-flow intervals by selecting flow segmentation points, and then linear fitting is performed piecewise. The fitting formula for each segment is as follows: , The selection principles for traffic segmentation points are as follows: Set initial flow segment points, including Q2, Q3, and 0.1*(Q2+Q3).

[0034] Optimize the selection of flow segment points based on relative error RD: If the RD after linear fitting of the j-th sub-flow interval j If the error exceeds three-quarters of the maximum permissible error, an additional flow point is added at the middle of the sub-flow interval, dividing it into two sub-flow intervals. If the RD after linear fitting of the j-th flow interval j If the error is less than or equal to three-quarters of the maximum permissible error, then the selected flow range is correct. Continue optimizing other sub-flow ranges until the RD of all sub-flow ranges is within the range. j Continue until all requirements are met.

[0035] Step S600, the formula for flow calculation is as follows: , In the formula, It is the instantaneous flow rate, and Δt is the propagation time difference of the upstream and downstream received signals under the current flow rate.

[0036] In this embodiment, the original dataset is preprocessed to remove outliers to ensure the subsequent clustering effect. Water meters are divided into two categories based on the resonant and anti-resonant frequencies of the transducers. The main category water meters with high consistency are selected by clustering and the default flow compensation coefficient is used directly to ensure accuracy without experimental calibration, thus improving calibration efficiency. Non-main category water meters are calibrated experimentally using a calibration scheme with an adaptive flow compensation scheme. Different flow compensation functions are matched based on the water meter error performance to ensure water meter accuracy and reduce production defect rate.

[0037] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calculating the flow compensation coefficient of an ultrasonic water meter, characterized in that, Includes the following steps: S100: Obtain the raw data set from multiple ultrasonic water meters; S200, Outlier Removal: Data preprocessing of the original dataset; S300. Obtain the main class water meter cluster and non-main class water meter cluster: Apply clustering algorithm to perform feature clustering on the resonant frequency and anti-resonant frequency in the original dataset; S400. Obtain the flow compensation function curve: The flow compensation coefficient is generated by summing the propagation time difference of the upstream and downstream received signals in the original data of the main type water meter and the measured flow rate. The formula is: , Where i is the water meter index representing different water meters, and j is the flow index representing different flow rates; The propagation time difference of the received signals upstream and downstream. This is the actual measured flow rate; Obtain a scatter plot of the flow compensation coefficients and fit it to the default temperature compensation function. The flow compensation function has the following form: , Where a, b, and c are the formula fitting coefficients; S500, Compensation: When the target water meter belongs to a non-main category water meter cluster, the propagation time difference of the upstream and downstream received signals and the measured flow are collected under three flow states: Q2, Q3, and 0.1*(Q2+Q3). ​​Following step S400, a flow compensation function is fitted, and flow verification is performed. The verification error calculation results are as follows. , Where j is the traffic index; When the verification error exceeds three-quarters of the maximum permissible error, a piecewise linear fitting method is adopted. The flow rate interval is divided into several sub-flow rate intervals by selecting flow rate segment points, and then linear fitting is performed segment by segment. The fitting formula for each segment is as follows: ； S600, Flow Calculation: The formula is as follows , In the formula, It is the instantaneous flow rate, and Δt is the propagation time difference of the upstream and downstream received signals under the current flow rate.

2. The method for calculating the flow compensation coefficient of an ultrasonic water meter according to claim 1, characterized in that, In step S100, the original data set includes the transducer resonant frequency and anti-resonant frequency, the propagation time difference of the upstream and downstream received signals, and the measured flow rate.

3. The method for calculating the flow compensation coefficient of an ultrasonic water meter according to claim 2, characterized in that, In step S200, the data preprocessing of the original dataset includes sorting the resonant frequencies and anti-resonant frequencies in the dataset from smallest to largest, extracting the middle 50% of the data to calculate a normal distribution, and identifying the remaining data points as outliers if they fall outside three times the standard deviation of the mean.

4. The method for calculating the flow compensation coefficient of an ultrasonic water meter according to claim 3, characterized in that, In step S400, j = (1, 2, 3, 4, 5, 6, 7, 8, 9, 10), representing flow rates of 5, 10, 16, 35, 65, 110, 160, 360, 560, and 2500 L / h, respectively.

5. The method for calculating the flow compensation coefficient of an ultrasonic water meter according to claim 4, characterized in that, In step S400, obtaining the scatter plot of the flow compensation coefficient through the propagation time difference of the upstream and downstream received signals includes taking the logarithm of the propagation time difference of the upstream and downstream received signals, as shown in the formula: .

6. The method for calculating the flow compensation coefficient of an ultrasonic water meter according to claim 5, characterized in that, In step S500, when the target water meter belongs to the main type of water meter cluster, the default flow compensation function is directly called. Perform traffic compensation.

7. The method for calculating the flow compensation coefficient of an ultrasonic water meter according to claim 6, characterized in that, In step S500, when the verification error is within three-quarters of the maximum permissible error, the current flow compensation function is considered usable.

8. The method for calculating the flow compensation coefficient of an ultrasonic water meter according to claim 7, characterized in that, In step S500, the selection principle for the flow segmentation point is as follows: Set initial flow segment points, including Q2, Q3, and 0.1*(Q2+Q3); Optimize the selection of flow segment points based on relative error RD: If the RD after linear fitting of the j-th sub-flow interval j If the error exceeds three-quarters of the maximum permissible error, an additional flow point is added at the middle of the sub-flow interval, dividing it into two sub-flow intervals. If the RD after linear fitting of the j-th flow interval j If the error is less than or equal to three-quarters of the maximum permissible error, then the selected flow range is correct. Continue optimizing other sub-flow ranges until the RD of all sub-flow ranges is within the range. j Continue until all requirements are met.

Images