A method and system for monitoring pipe network pressure distribution

Abstract

The present application discloses a method and system for monitoring the pressure distribution of a pipeline network, the method comprising the following steps: constructing an algorithm model library, the algorithm model library comprising at least two interpolation algorithm models; calculating the first pressure distribution data of the pipeline network respectively based on at least two interpolation algorithm models of the algorithm model library according to at least part of the pressure data; verifying the calculation result of the first pressure distribution data according to at least part of the pressure data, and obtaining the interpolation error of the corresponding interpolation algorithm model in the algorithm model library; assigning weights to the corresponding interpolation algorithm model in the algorithm model library according to the interpolation error, the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the interpolation error corresponding to the interpolation algorithm model; obtaining the second pressure distribution data of the pipeline network by comprehensive calculation according to the first pressure distribution data calculated by the interpolation algorithm model and the weight coefficient corresponding to the interpolation algorithm. The present application can improve the accuracy and reliability of monitoring the pressure distribution of the pipeline network.

Description

Technical Field

[0001] The present application relates to the field of pipeline network safety, and in particular to a pipeline network pressure distribution monitoring method and system. Background Art

[0002] In the current field of pipeline pressure distribution monitoring, related technologies mainly focus on using a single algorithm model to calculate pressure distribution data, ignoring the complexity and diversity of pipeline network structures. In a pipeline network, different pipeline network structures have different pressure distributions, and the pressure distribution of the pipeline network changes dynamically. The single algorithm model used by related technologies lacks adaptability to pipeline network structure and pressure changes, and cannot intelligently adjust the calculation method to cope with different situations, limiting the reliability and accuracy of pressure distribution calculations. The existence of these problems seriously affects the accuracy and reliability of the system's monitoring of pipeline pressure, and urgently needs to be broken through through technological innovation. Summary of the invention

[0003] The purpose of this application is to provide a technical solution to solve the problem that a single algorithm model existing in the related art is prone to errors and is difficult to accurately reflect the pressure distribution status of the pipeline network.

[0004] In a first aspect, the present application provides a method for monitoring pressure distribution in a pipe network, wherein the pipe network is provided with a plurality of monitoring points, and a collection unit for collecting pressure data is provided at the monitoring points, wherein the method comprises the following steps:

[0005] Building an algorithm model library, which includes at least two interpolation algorithm models;

[0006] According to at least part of the pressure data, first pressure distribution data of the pipe network is calculated based on at least two interpolation algorithm models in the algorithm model library;

[0007] Verifying the calculation result of the first pressure distribution data according to at least part of the pressure data to obtain an interpolation error of a corresponding interpolation algorithm model in the algorithm model library;

[0008] According to the interpolation error, a weight is assigned to the corresponding interpolation algorithm model in the algorithm model library, and the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the interpolation error corresponding to the interpolation algorithm model;

[0009] According to the first pressure distribution data calculated by the interpolation algorithm model and the weight coefficient corresponding to the interpolation algorithm, the second pressure distribution data of the pipeline network is obtained by comprehensive calculation.

[0010] Furthermore, the weight coefficient is calculated by the following formula:

[0011] ;

[0012] Where Ei represents the interpolation error corresponding to the i-th interpolation algorithm model, and k represents the number of interpolation algorithm models.

[0013] Furthermore, the second pressure distribution data is calculated by the following formula:

[0014] ;

[0015] In the formula, represents the first pressure distribution data of the pipe network calculated according to the i-th interpolation algorithm model, represents the weight coefficient corresponding to the i-th interpolation algorithm model, k represents the number of interpolation algorithm models, .

[0016] In a second aspect, the present application provides a pipeline network pressure distribution monitoring system, the system comprising:

[0017] A number of collection units, which are arranged at different monitoring points of the pipe network and are used to collect pressure data of corresponding monitoring points of the pipe network;

[0018] An algorithm model library, the algorithm model library includes at least two interpolation algorithm models;

[0019] A first pressure calculation unit, which calculates first pressure distribution data of the pipe network based on at least part of the pressure data and at least two interpolation algorithm models in the algorithm model library;

[0020] An error acquisition unit, wherein the error acquisition unit verifies the first pressure distribution data according to at least part of the pressure data to obtain an interpolation error of a corresponding interpolation algorithm model in an algorithm model library;

[0021] The weight calculation unit assigns weights to the corresponding interpolation algorithm models in the algorithm model library according to the interpolation error, and the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the interpolation error corresponding to the interpolation algorithm model;

[0022] The second pressure calculation unit obtains the second pressure distribution data of the pipe network by comprehensive calculation according to the first pressure distribution data calculated by the interpolation algorithm model and the weight coefficient corresponding to the interpolation algorithm.

[0023] Furthermore, the error acquisition unit acquires the interpolation error through the following formula:

[0024] E i =α*MAE+(1-α)*RMSE;

[0025] E i represents the interpolation error, and RMSE represents the value calculated using the root mean square error algorithm.i The error of the interpolation algorithm model, MAE represents the error of the first interpolation algorithm calculated using the mean absolute error algorithm. i The error of the interpolation algorithm model, α represents the weight coefficient of the root mean square error algorithm, and (1-α) represents the weight coefficient of the mean absolute error algorithm.

[0026] Furthermore, the weight coefficient is calculated by the following formula:

[0027] ;

[0028] Where Ei represents the interpolation error corresponding to the i-th interpolation algorithm model, and k represents the number of interpolation algorithm models.

[0029] Furthermore, the second pressure distribution data is calculated by the following formula:

[0030] ;

[0031] In the formula, represents the first pressure distribution data of the pipe network calculated according to the i-th interpolation algorithm model, represents the weight coefficient corresponding to the i-th interpolation algorithm model, k represents the number of interpolation algorithm models, .

[0032] Furthermore, the algorithm model library includes at least two of the following interpolation algorithm models: inverse distance weighted interpolation, Kriging interpolation model, and polynomial interpolation model.

[0033] Furthermore, the pipe network pressure distribution monitoring system also includes:

[0034] A constraint condition setting unit, which is used to set monitoring point distribution density constraint conditions according to the topological structure of the pipe network and the location of the monitoring points;

[0035] The error acquisition unit selects the pressure data of the corresponding monitoring point according to the monitoring point distribution density constraint condition to verify the first pressure distribution data, and obtains the interpolation error of the corresponding interpolation algorithm model.

[0036] Furthermore, the pipeline network pressure distribution monitoring system also includes a rendering unit, which displays the second pressure distribution data of the pipeline network on the pipeline network topology model in the form of a gradient color heat map based on the second pressure distribution data of the pipeline network.

[0037] Furthermore, the pipeline network pressure distribution monitoring system also includes a pressure anomaly analysis unit, which analyzes the degree to which the pressure exceeds the limit in each abnormal area based on the second pressure distribution data, and determines the abnormal level of the abnormal area.

[0038] According to the above description, the present application provides a pipeline pressure distribution monitoring system and method, which is based on the intelligent combination of multiple interpolation algorithm models. By verifying the calculation error of each interpolation algorithm model, the weight of the interpolation algorithm model in calculating the second pressure distribution data is dynamically adjusted to calculate more accurate pipeline pressure distribution data, thereby improving the accuracy and reliability of the pipeline pressure distribution monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A partial structural diagram of a pipe network pressure distribution monitoring system provided in an embodiment of the present application;

[0040] Figure 2 A schematic diagram of an algorithm model library in a pipe network pressure distribution monitoring system provided in an embodiment of the present application;

[0041] Figure 3 Another partial structural diagram of a pipe network pressure distribution monitoring system provided in an embodiment of the present application;

[0042] Figure 4 A schematic diagram of the division of regional pressure levels in a pipe network pressure distribution monitoring system provided in an embodiment of the present application;

[0043] Figure 5 A flow chart of a method for monitoring pressure distribution in a pipeline network provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] The present application will be described in detail below in conjunction with the specific implementation modes shown in the accompanying drawings, but these implementation modes do not limit the present application. Structural, methodological, or functional changes made by ordinary technicians in the field based on these implementation modes are included in the protection scope of the present application.

[0045] The purpose of this application is to provide a technical solution to solve the problem that a single algorithm model in the related art is prone to errors and is difficult to accurately reflect the pressure distribution status of the pipeline network. Based on this purpose, this application provides a pipeline network pressure distribution monitoring system, which can intelligently combine multiple algorithm models to calculate pipeline network pressure distribution data, and dynamically adjust the weight of the interpolation algorithm model in calculating the second pressure distribution data by verifying the calculation error of each interpolation algorithm model, so as to obtain the pressure distribution data of the entire pipeline network with high precision, and effectively improve the monitoring ability of pipeline network pressure risks.

[0046] like Figure 1 As shown, a pipeline network pressure distribution monitoring system provided by an embodiment of the present application may include:

[0047] A plurality of collection units 11, which are arranged at different monitoring points of the pipe network and are used to collect pressure data of corresponding monitoring points of the pipe network;

[0048] An algorithm model library 12, wherein the algorithm model library 12 includes at least two interpolation algorithm models;

[0049] A first pressure calculation unit 13, which calculates first pressure distribution data of the pipe network based on at least part of the pressure data and at least two interpolation algorithm models in the algorithm model library 12;

[0050] An error acquisition unit 14, the error acquisition unit 14 verifies the first pressure distribution data according to at least part of the pressure data, and obtains an interpolation error of the interpolation algorithm model corresponding to the algorithm model library;

[0051] The weight calculation unit 15 assigns weights to the corresponding interpolation algorithm models in the algorithm model library according to the interpolation error, and the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the interpolation error corresponding to the interpolation algorithm model;

[0052] The second pressure calculation unit 16 obtains the second pressure distribution data of the pipe network through comprehensive calculation according to the first pressure distribution data calculated by the interpolation algorithm model and the weight coefficient corresponding to the interpolation algorithm.

[0053] Specifically, in the pipeline pressure distribution monitoring system of the embodiment of the present application, several acquisition units 11 are deployed at different monitoring points of the pipeline network. The deployment positions of the monitoring points are usually selected based on the layout structure of the pipeline network. Preferably, they can be selected to be deployed in the cross-node area in the pipeline network and the pressure fluctuation area that may exist in the pipeline network. Through the method of multi-monitoring point acquisition, the acquisition range covers the entire pipeline network to ensure the comprehensiveness of data acquisition. The acquisition unit 11 can collect pressure data of each monitoring point in the pipeline network in real time or at a preset frequency to ensure the accuracy of data acquisition. In this embodiment, the collected pressure data may include pressure monitoring data and monitoring point location data. The acquisition unit 11 may include a pressure acquisition sensor, a pressure data recorder, and a communication transmission device to ensure the storage and timely transmission of the collected data.

[0054] In one embodiment, the collection unit may include a separately arranged pressure collection sensor. In another embodiment, the collection unit may also include different types of devices with pressure collection capabilities that already exist in the pipeline network. For example, pressure data may be collected through devices such as pressure regulators, terminal flow meters, valve control devices, etc. that exist in the pipeline network.

[0055] In one embodiment, the interpolation algorithm models stored in the algorithm model library 12 may include at least two interpolation algorithm models including an inverse distance weighted interpolation model, a Kriging interpolation model, and a polynomial interpolation model. It should be noted that the algorithm model library 12 can be expanded according to actual needs and can include more types of algorithm models for interpolation calculation of pipe network pressure distribution.

[0056] For ease of explanation, in the embodiment of the present application, the algorithm model library 12 includes three interpolation algorithm models, namely, the inverse distance weighted interpolation model, the Kriging interpolation model, and the polynomial interpolation model. In the face of the complex and changeable pipe network environment, the pressure data collected by the acquisition unit 11 is also complex and diverse. The interpolation algorithm models stored in the algorithm model library 12 have their unique advantages. For example, the Kriging interpolation model performs well in processing spatial autocorrelation data, while the inverse distance weighted interpolation model is more applicable when the data points are relatively evenly distributed. The algorithm model library 12 provides a variety of tools for subsequent data processing and data analysis to ensure the accuracy and reliability of the calculation results.

[0057] The first pressure calculation unit 13 selects at least part of the collected pressure data, and calculates the first pressure distribution data respectively according to the selected pressure data based on at least two interpolation algorithm models in the algorithm model library 12. In the above processing, the original collected data can be preprocessed first to remove interference data to ensure the quality of data used for input calculation. Then, the first pressure calculation unit 13 uses different interpolation algorithm models to calculate the pipeline network area based on at least two interpolation algorithm models included in the algorithm model library 12, and combines the monitoring point location data of the pressure data, thereby obtaining the first pressure distribution data of the pipeline network area under different interpolation algorithm models.

[0058] The error acquisition unit 14 compares the actual pressure value of the monitoring point collected by the collection unit 11 with the first pressure distribution data of the corresponding pipe network area calculated by the interpolation algorithm model, calculates the error between the first pressure distribution data of the corresponding pipe network area calculated by each interpolation algorithm model and the actual pressure data, and obtains the interpolation error of each interpolation algorithm model in the algorithm model library (the method of obtaining the interpolation error is explained later).

[0059] The weight calculation unit 15 assigns weights to the corresponding interpolation algorithm models in the algorithm model library based on the interpolation error provided by the error acquisition unit 14. The above assignment process follows a basic principle: the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the corresponding interpolation error, that is, the smaller the interpolation error, the higher the weight obtained by the interpolation algorithm model. It should be noted that in this embodiment, the sum of all weights is always 1.

[0060] The second pressure calculation unit 16 is the part of the pipeline network pressure distribution monitoring system responsible for calculating the second pressure distribution data. The second pressure calculation unit 16 calculates the first pressure distribution data of the pipeline network area based on the interpolation algorithm models obtained by the first pressure calculation unit 13, and combines the weight coefficients assigned to each interpolation algorithm model by the weight calculation unit 15. After comprehensive calculation (the specific calculation method is described later), the second pressure distribution data of the pipeline network is obtained. The second pressure calculation unit 16 performs comprehensive calculations by combining the weight proportions of each interpolation algorithm model and the first pressure distribution data. The output second pressure distribution data represents the optimal calculation result of the pipeline network pressure distribution monitoring system for the pipeline network pressure distribution, which provides an important basis for the subsequent management and decision-making of the pipeline network by the pipeline network pressure distribution monitoring system.

[0061] As an optional implementation, the second pressure distribution data is calculated by the following formula:

[0062] ;

[0063] In the formula, represents the first pressure distribution data of the pipe network calculated according to the i-th interpolation algorithm model, represents the weight coefficient corresponding to the i-th interpolation algorithm model, k represents the number of interpolation algorithm models, .

[0064] Specifically, in the pipeline network pressure distribution monitoring system of the embodiment of the present application, the pipeline network pressure distribution monitoring system adopts a second pressure distribution data calculation formula based on the first pressure distribution data and the weight coefficient. In the above second pressure distribution data calculation formula, Indicates the second pressure distribution data of the pipeline network, which represents the optimal calculation result of the pipeline network pressure distribution monitoring system for the pipeline network pressure distribution; represents the first pressure distribution data of the pipe network calculated according to the i-th interpolation algorithm model, that is, the pipe network pressure distribution value initially calculated by each interpolation algorithm model involved in the calculation; represents the weight coefficient corresponding to the i-th interpolation algorithm model, which determines the proportion of the interpolation algorithm model in calculating the second pressure distribution data; k represents the total number of interpolation algorithm models involved in calculating the first pressure distribution data; Indicates that the sum of the weight coefficients of the various interpolation algorithm models involved in the calculation is 1.

[0065] In one embodiment, it is assumed that there are N interpolation algorithm models involved in the calculation of the first pressure distribution data, and the initial weight of each interpolation algorithm model is 1 / N. The weight calculation unit 15 selects the optimal algorithm combination for the pipe network topology structure based on the weight allocation strategy of the interpolation error and the adaptability of each interpolation algorithm model to the pipe network topology structure, and adjusts the weight proportion of different interpolation algorithm models in the algorithm combination, ensuring that in subsequent calculations, a more accurate interpolation algorithm model can have a greater impact on the calculation results, thereby improving the accuracy of subsequent calculation results.

[0066] Through the above description, the pipeline pressure distribution monitoring system provided in the embodiment of the present application realizes accurate monitoring and analysis of the pipeline pressure distribution through a combination of multiple interpolation algorithms, ensuring the accuracy and reliability of the pipeline pressure distribution monitoring system, and providing data support for efficient management of the pipeline network and subsequent maintenance decisions.

[0067] As an optional implementation method, the weight coefficient is calculated by the following formula:

[0068] ;

[0069] In the formula, E i represents the interpolation error corresponding to the i-th interpolation algorithm model, and k represents the number of interpolation algorithm models.

[0070] In one embodiment, the interpolation error of the i-th interpolation algorithm model may be calculated using a root mean square error (RMSE) or a mean absolute error (MAE).

[0071] For example, the first pressure distribution data is calculated using the i-th interpolation algorithm model, and the error between the first pressure distribution data and the actual pressure monitoring data is calculated using, for example, the root mean square error (RMSE), and the error is used as the interpolation error of the i-th interpolation algorithm model.

[0072] In another embodiment, the interpolation error may be obtained by comprehensive calculation using the root mean square error (RMSE) and the mean absolute error (MAE).

[0073] Specifically, the error acquisition unit acquires the interpolation error through the following formula:

[0074] E i =α*MAE+(1-α)*RMSE;

[0075] E i represents the interpolation error, and RMSE represents the value calculated using the root mean square error algorithm. iThe error of the interpolation algorithm model, MAE represents the error of the first interpolation algorithm calculated using the mean absolute error algorithm. i The error of the interpolation algorithm model, α represents the weight coefficient of the root mean square error algorithm, and (1-α) represents the weight coefficient of the mean absolute error algorithm.

[0076] Through the above calculation formula, the pipeline network pressure distribution monitoring system can adjust the weight of the interpolation algorithm model in the subsequent pressure distribution data calculation according to the interpolation error of each interpolation algorithm model in different pipeline network topology environments. The interpolation algorithm model with a smaller interpolation error obtains a higher weight coefficient and occupies a larger proportion in the comprehensive calculation. The calculation and allocation of the weight coefficient ensures the effective combination of the interpolation algorithm model and improves the accuracy of calculating the second pressure distribution data. The pipeline network pressure distribution monitoring system responds to the interpolation error of the interpolation algorithm model and effectively improves the intelligence level of the pipeline network pressure distribution monitoring system in calculating the second pressure score data by adjusting the weight coefficient.

[0077] As an optional implementation, the algorithm model library 12 includes at least two of the following interpolation algorithm models: inverse distance weighted interpolation, Kriging interpolation model, and polynomial interpolation model.

[0078] Specifically, in the pipe network pressure distribution monitoring system of the embodiment of the present application, Figure 2 As shown, the algorithm model library 12 includes at least two interpolation algorithm models, such as the inverse distance weighted interpolation model, the Kriging interpolation model, and the polynomial interpolation model. The selection and application of the above interpolation algorithm models are directly related to the accuracy of the pipe network pressure distribution data. The characteristics of these interpolation algorithm models and their application in the pipe network pressure distribution monitoring system will be specifically described below:

[0079] The inverse distance weighted interpolation model is a weighted average interpolation method that assumes that the value of an unknown point can be obtained by weighted average of the values ​​of surrounding known points. The weight is inversely proportional to the distance. The closer the known point data is to the unknown point, the greater the impact on the interpolation result. In the pipe network pressure distribution monitoring system, when calculating the second pressure distribution data of the pipe network area where the monitoring points are more evenly distributed, the inverse distance weighted interpolation model occupies a higher weight in the algorithm combination;

[0080] The Kriging interpolation model is an interpolation method based on probability statistics theory. It interpolates by considering the spatial autocorrelation between monitoring points, predicting the value of a point as the weighted average of surrounding known points, with the weight calculated based on the distance and similarity between data points. In the pipe network pressure distribution monitoring system, when calculating the second pressure distribution data of a pipe network area with a complex pipe network layout, uneven distribution of monitoring points, and spatial correlation, the Kriging interpolation model has a higher weight in the algorithm combination;

[0081] The polynomial interpolation model fits known data points by constructing a polynomial function to infer the value of unknown points. In the pipe network pressure distribution monitoring system, when calculating the second pressure distribution data of the pipe network area with strong distribution regularity of monitoring points and can be expressed by mathematical models, the polynomial interpolation model has a higher weight in the algorithm combination;

[0082] The pipeline network pressure distribution monitoring system can adapt to complex and diverse pipeline network layouts and data characteristics by adopting a variety of interpolation algorithm models. At the same time, based on the characteristics of the interpolation algorithm model and the weight distribution of errors, the pipeline network pressure distribution monitoring system can select a more reasonable combination of interpolation algorithm models to avoid the impact of errors caused by a single interpolation algorithm model, thereby improving the flexibility and accuracy of the pipeline network pressure distribution monitoring system.

[0083] As an optional implementation, Figure 3 As shown, the pipeline network pressure distribution monitoring system also includes:

[0084] The constraint condition setting unit 17 is used to set the monitoring point distribution density constraint condition according to the topological structure of the pipe network and the location of the monitoring point;

[0085] The error acquisition unit 14 selects the pressure data of the corresponding monitoring point according to the monitoring point distribution density constraint condition to verify the first pressure distribution data, and obtains the interpolation error of the corresponding interpolation algorithm model in the algorithm model library.

[0086] Specifically, in the pipe network pressure distribution monitoring system of the embodiment of the present application, the constraint condition setting unit 17 sets the constraint condition of the monitoring point distribution density according to the topological structure of the pipe network and the positions of the monitoring points deployed by the acquisition unit 11 .

[0087] The error acquisition unit 14 selects the corresponding monitoring point data of the pipeline area where the pressure data used by the first pressure calculation unit 13 to calculate the first pressure distribution data is located according to the monitoring point distribution density constraint condition set by the constraint condition setting unit 17, compares and verifies the data with the first pressure distribution data, and obtains the interpolation error of the interpolation algorithm model involved in calculating the first pressure distribution data.

[0088] The error acquisition unit 14 selects the pressure data of the corresponding monitoring point based on the monitoring point distribution density constraint condition. By comparing the actual pressure data with the calculation results of the interpolation algorithm model, the error acquisition unit 14 can quantify the interpolation error of each interpolation algorithm model at the corresponding monitoring point, and provide a basis for the subsequent weight allocation of each interpolation algorithm model.

[0089] Exemplarily, the monitoring point distribution density constraint condition may be to randomly select pressure data of 10 known monitoring points in the pipe network to verify the first pressure distribution data.

[0090] As shown in Table 1, it shows the first pressure distribution value obtained by the interpolation calculation of each interpolation algorithm model at each monitoring point. As shown in Table 2, it shows the error calculated by each interpolation algorithm model using the data of the 10 monitoring points as the true value. It is easy to understand that under different pipe network topologies, according to the different settings of the monitoring point distribution density constraint conditions (the number and location of optional monitoring points), the interpolation error of each interpolation algorithm model will also change accordingly, thereby affecting the weight coefficient of the interpolation algorithm model in the adaptive hybrid interpolation result provided in this application.

[0091]

[0092] Table 1

[0093]

[0094] Table 2

[0095] It can be seen from Table 2 that according to the adaptive hybrid interpolation algorithm model provided in the embodiment of the present application, the error of the calculated pipeline pressure distribution (ie, the second pressure distribution data) is significantly lower than that of the pipeline pressure distribution calculated using a single interpolation algorithm.

[0096] Through the above processing mechanism, the pipeline network pressure distribution monitoring system can select the optimal algorithm combination in the algorithm database and construct a hybrid interpolation model that adapts to the characteristics of the pipeline network topology structure based on various constraints including the distribution density of monitoring points and data quality, combined with the interpolation errors of each interpolation algorithm model obtained by the error acquisition unit 14, and comprehensively consider the adaptability of various interpolation algorithm models to the pipeline network topology structure. The pipeline network pressure distribution monitoring system dynamically optimizes the weight ratio of each interpolation algorithm model when the hybrid interpolation model calculates the second pressure distribution data based on the interpolation errors of each interpolation algorithm model, thereby enhancing the adaptability of the pipeline network pressure distribution monitoring system to different pipeline network topologies and the intelligent level of dynamically adjusting the calculation method.

[0097] As an optional implementation, Figure 3 As shown, the pipeline network pressure distribution monitoring system further includes a rendering unit 18, which displays the second pressure distribution data of the pipeline network on the pipeline network topology model in the form of a gradient color heat map according to the second pressure distribution data of the pipeline network.

[0098] Specifically, in the pipe network pressure distribution monitoring system of the embodiment of the present application, the rendering unit 18 converts the second pressure distribution data into color information, and different colors represent different pressure distribution data. For example, the color can gradually change from a cold color (such as blue) to a warm color (such as red), indicating the change of the second pressure distribution data from low to high. The rendering unit 18 maps the second pressure distribution data converted into color information to the pipe network topology model according to the pipe network area, and presents the size of the second pressure distribution data of the pipe network area in an intuitive and easy-to-read form through the gradual change of color, thereby realizing the fusion display of the pressure distribution data and the spatial structure of the pipe network, improving the visualization effect of the pipe network pressure distribution data, and providing intuitive data support for the monitoring and maintenance of the pipe network by the pipe network pressure distribution monitoring system. The pipe network pressure distribution monitoring system can quickly identify the pipe network area with abnormal pressure by observing the color changes of the pipe network area, and make reasonable response measures based on the visualization data, effectively preventing potential safety risks, and ensuring the safe and stable operation of the pipe network.

[0099] As an optional implementation, Figure 3 As shown, the pipe network pressure distribution monitoring system further includes a pressure anomaly analysis unit 19. The pressure anomaly analysis unit 19 analyzes the degree to which the pressure in each abnormal area exceeds the limit value based on the second pressure distribution data, and determines the abnormal level of the abnormal area.

[0100] Specifically, in the pipe network pressure distribution monitoring system of the embodiment of the present application, the pressure anomaly analysis unit 19 accurately identifies and analyzes the area in the pipe network that exceeds the normal pressure range based on the second pressure distribution data. This process may include multiple steps of anomaly detection, degree quantification and grade classification of the pipe network area. First, the pressure anomaly analysis unit 19 identifies the pipe network area where the second pressure distribution data exceeds the preset pressure limit; then, the pressure anomaly analysis unit 19 quantifies the degree of excess of the second pressure distribution data of this part of the pipe network area, that is, the difference between the pressure value and the preset pressure limit; finally, Figure 4 As shown, the pressure anomaly analysis unit 19 divides the pipeline network area into different abnormal levels according to the degree of excess of the second pressure distribution data, such as a mild abnormality area, a moderate abnormality area, and a severe abnormality area. The pipeline network area where the second pressure distribution data does not exceed the preset pressure limit can be divided into a normal pressure area.

[0101] The pressure anomaly analysis unit 19 divides the abnormal area of ​​the pipeline network into different abnormal levels based on the quantitative structure of the degree to which the second pressure distribution data exceeds the preset pressure limit value through the preset division rules. The pipeline network pressure distribution monitoring system executes emergency measures corresponding to the abnormal level based on the abnormal level of the abnormal area determined by the pressure anomaly analysis unit 19. For example, when a mild abnormal area appears, the system issues a prompt alarm; when a moderate abnormal area appears, the system issues a text message alarm; when a severe abnormal area appears, the system issues a voice telephone alarm call. Through the mechanism of determining the level of abnormal areas in the pipeline network, the pipeline network pressure distribution monitoring system can promptly identify and respond to pressure anomalies, implement effective countermeasures, and ensure the safe and stable operation of the pipeline network.

[0102] like Figure 5 As shown, the embodiment of the present application also provides a method for monitoring the pressure distribution of a pipe network. The pipe network is provided with a plurality of monitoring points, and a collection unit 11 for collecting pressure data is provided at the monitoring points. The method is applicable to the above-mentioned pipe network pressure distribution monitoring system, and the method comprises the following steps:

[0103] Step S101, constructing an algorithm model library 12, wherein the algorithm model library 12 includes at least two interpolation algorithm models;

[0104] Step S102, calculating first pressure distribution data of the pipe network respectively based on at least two interpolation algorithm models in the algorithm model library 12 according to at least part of the pressure data;

[0105] Step S103, verifying the calculation result of the first pressure distribution data according to at least part of the pressure data, and obtaining the interpolation error of the corresponding interpolation algorithm model in the algorithm model library;

[0106] Step S104, assigning weights to the corresponding interpolation algorithm models in the algorithm model library according to the interpolation error, wherein the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the interpolation error corresponding to the interpolation algorithm model;

[0107] Step S105, obtaining second pressure distribution data of the pipe network through comprehensive calculation based on the first pressure distribution data calculated by the interpolation algorithm model and the weight coefficient corresponding to the interpolation algorithm.

[0108] As an optional implementation, in the pipe network pressure distribution monitoring method of the embodiment of the present application, the weight coefficient is calculated by the following formula:

[0109] ;

[0110] In the formula, represents the weight coefficient of the i-th interpolation algorithm model, E irepresents the interpolation error corresponding to the i-th interpolation algorithm model, and k represents the number of interpolation algorithm models.

[0111] As an optional implementation, in the pipeline network pressure distribution monitoring method of the embodiment of the present application, the second pressure distribution data is calculated by the following formula:

[0112] ;

[0113] In the formula, represents the second pressure distribution data, represents the first pressure distribution data of the pipe network calculated according to the i-th interpolation algorithm model, represents the weight coefficient corresponding to the i-th interpolation algorithm model, k represents the number of interpolation algorithm models, and the sum of the weight coefficients of the multiple interpolation algorithm models involved in the calculation is 1, which is expressed as .

[0114] It can be understood that this method is applicable to the above-mentioned pipeline pressure distribution monitoring system. The specific implementation of this method has been reflected in the above description and will not be repeated here.

[0115] In summary, the pipeline pressure distribution monitoring system and method provided in the present application can intelligently combine multiple interpolation algorithm models, dynamically adjust the weight of the algorithm in calculating the second pressure score data based on the difference error of the interpolation algorithm model, and obtain more accurate pipeline pressure distribution data, thereby improving the accuracy and reliability of the pipeline pressure distribution monitoring system.

[0116] It will be appreciated that the word "exemplary" as used herein means "serving as an example, instance, or illustration". Any embodiment described as "exemplary" is not necessarily preferred or superior to other embodiments and / or does not exclude the combination of features of other embodiments. It will be appreciated that certain features of the present application described in the context of separate embodiments for the sake of clarity may also be provided in a single embodiment by combination. Conversely, various features of the present application described in the context of a single embodiment for the sake of clarity may also be provided individually or in any suitable combination or as any other described embodiment of the present application.

[0117] The above disclosure is only the preferred embodiment of the present application, but it is not intended to limit the scope of rights of the present application. A person of ordinary skill in the art can understand that without departing from the spirit and scope of the present application and the appended claims, changes, modifications, substitutions, combinations, and simplifications should all be equivalent replacement methods and still fall within the scope of the invention.

[0118] The application fields of the technology of the present invention include but are not limited to pipeline safety. As long as the essence of the technical solution does not deviate from the scope of the technical solution of each embodiment of the present invention, it falls within the protection scope of the present invention.

Claims

1. A method for monitoring pressure distribution in a pipe network, wherein the pipe network is provided with a plurality of monitoring points, and each monitoring point is provided with a collection unit for collecting pressure data, characterized in that: The method comprises the following steps: Building an algorithm model library, wherein the algorithm model library includes at least two interpolation algorithm models; According to at least part of the pressure data, first pressure distribution data of the pipe network is calculated based on at least two interpolation algorithm models in the algorithm model library; Verifying the calculation result of the first pressure distribution data according to at least a remaining part of the pressure data to obtain an interpolation error of the corresponding interpolation algorithm model in the algorithm model library; According to the interpolation error, weights are assigned to the corresponding interpolation algorithm models in the algorithm model library, wherein the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the interpolation error corresponding to the interpolation algorithm model; According to the first pressure distribution data calculated by the interpolation algorithm model and the weight coefficient corresponding to the interpolation algorithm, the second pressure distribution data of the pipe network is obtained by comprehensive calculation; The weight coefficient is calculated by the following formula: ； In the formula, E i Indicates i The interpolation error corresponding to the interpolation algorithm model is k Indicates the number of interpolation algorithm models; The interpolation error is obtained by the following formula: E i =α*MAE+(1-α)*RMSE; E i represents the interpolation error, and RMSE represents the value calculated using the root mean square error algorithm. i The error of the interpolation algorithm model, MAE represents the error of the first interpolation algorithm calculated using the mean absolute error algorithm. i The error of the interpolation algorithm model, α represents the weight coefficient of the root mean square error algorithm, and (1-α) represents the weight coefficient of the mean absolute error algorithm.

2. The method for monitoring the pressure distribution of a pipe network according to claim 1, characterized in that: The second pressure distribution data is calculated by the following formula: ； In the formula, Indicates that according to i The first pressure distribution data of the pipe network calculated by the interpolation algorithm model, Indicates i The weight coefficients corresponding to the interpolation algorithm models are: k Indicates the number of interpolation algorithm models, .

3. A pipe network pressure distribution monitoring system, characterized in that: The pipeline network pressure distribution monitoring system includes: A plurality of collection units, each of which is arranged at different monitoring points of the pipe network and is used to collect pressure data of corresponding monitoring points of the pipe network; An algorithm model library, wherein the algorithm model library includes at least two interpolation algorithm models; a first pressure calculation unit, which calculates first pressure distribution data of the pipe network based on at least part of the pressure data and at least two interpolation algorithm models in the algorithm model library; an error acquisition unit, wherein the error acquisition unit verifies the first pressure distribution data according to at least a remaining portion of the pressure data to obtain an interpolation error of the corresponding interpolation algorithm model in the algorithm model library; A weight calculation unit, which assigns a weight to a corresponding interpolation algorithm model in the algorithm model library according to the interpolation error, wherein the weight coefficient of the interpolation algorithm model is positively correlated with the inverse of the interpolation error corresponding to the interpolation algorithm model; A second pressure calculation unit, wherein the second pressure calculation unit obtains second pressure distribution data of the pipe network by comprehensive calculation according to the first pressure distribution data calculated by the interpolation algorithm model and a weight coefficient corresponding to the interpolation algorithm; The weight coefficient is calculated by the following formula: ； In the formula, E i Indicates i The interpolation error corresponding to the interpolation algorithm model is k Indicates the number of interpolation algorithm models; The error acquisition unit acquires the interpolation error by the following formula: E i =α*MAE+(1-α)*RMSE; E i represents the interpolation error, and RMSE represents the value calculated using the root mean square error algorithm. i The error of the interpolation algorithm model, MAE represents the error of the first interpolation algorithm calculated using the mean absolute error algorithm. i The error of the interpolation algorithm model, α represents the weight coefficient of the root mean square error algorithm, and (1-α) represents the weight coefficient of the mean absolute error algorithm.

4. The pipe network pressure distribution monitoring system according to claim 3, characterized in that: The second pressure distribution data is calculated by the following formula: ； In the formula, Indicates that according to i The first pressure distribution data of the pipe network calculated by the interpolation algorithm model, Indicates i The weight coefficients corresponding to the interpolation algorithm models are: k Indicates the number of interpolation algorithm models, .

5. The pipe network pressure distribution monitoring system according to claim 3, characterized in that: The algorithm model library includes at least two of the following interpolation algorithm models: inverse distance weighted interpolation, Kriging interpolation model, and polynomial interpolation model.

6. The pipe network pressure distribution monitoring system according to claim 3, characterized in that: The system further comprises: A constraint condition setting unit, the constraint condition setting unit is used to set a monitoring point distribution density constraint condition according to the topological structure of the pipe network and the location of the monitoring point; The error acquisition unit selects pressure data of corresponding monitoring points according to the monitoring point distribution density constraint condition to verify the first pressure distribution data, and obtains an interpolation error corresponding to the interpolation algorithm model.

7. The pipe network pressure distribution monitoring system according to claim 3, characterized in that: The system further includes a pressure anomaly analysis unit, which analyzes the degree to which the pressure of each abnormal area exceeds the limit value based on the second pressure distribution data, and determines the abnormal level of the abnormal area.

Images