Intelligent analysis and diagnosis system for caliber matching of gas measuring instrument

Abstract

The invention provides an intelligent analysis and diagnosis system for caliber matching of a gas measuring instrument, which relates to the technical field of data processing and comprises an extraction module, an analysis and diagnosis module and an analysis and diagnosis module, selecting a main pipe intersection P1, a user access point P2 and a pressure regulating station output point P3 as three dynamic topology reference points; constructing a dynamic topological unit according to the three points, performing regional division on the unit, and extracting pressure-flow coupling feature values of sub-regions; the correction module is used for performing weighted correction on the user gas consumption demand sudden change probability feature vector through the pressure-flow coupling feature value to generate a corrected sudden change probability feature vector; and the processing module is used for performing tensor splicing on the corrected mutation probability feature vector and the pipeline equivalent inner diameter dynamic feature vector, and inputting a full-connection diagnosis network to generate a caliber matching deviation index. The intelligent operation efficiency of gas supply can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to an intelligent analysis and diagnosis system for gas meter caliber matching. Background Art

[0002] In a large commercial complex, which covers various gas - using scenarios such as shopping malls, hotels, and restaurants, the gas demand is complex and fluctuates greatly. During the initial construction, the caliber of gas meters was configured based on conventional estimates and experience. As the operation progresses, the number of catering merchants in the shopping mall increases, and their business hours vary, resulting in insufficient gas supply during peak gas - using periods. The gas - using equipment of some merchants cannot operate at full load normally, affecting the normal operation of the merchants.

[0003] During off - peak hours, the overly large caliber of the meter reduces the measurement accuracy, causing inaccurate gas measurement. Disputes often arise between gas companies and merchants regarding gas volume settlement. In addition, due to long - term unreasonable gas supply, some gas - using equipment frequently malfunctions due to unstable pressure, increasing the equipment maintenance cost. Therefore, this highlights the defects of some traditional gas meter caliber matching methods in dealing with complex gas - using scenarios. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an intelligent analysis and diagnosis system for gas meter caliber matching, which can improve the intelligent operation efficiency of the gas supply system.

[0005] To solve the above - mentioned technical problem, the technical solution of the present invention is as follows: In a first aspect, an intelligent analysis and diagnosis system for gas meter caliber matching includes: A collection module, configured to collect pipeline pressure gradient data, flow phase data, and user gas - using time - series fluctuation data in real time through pressure sensors deployed at the three - dimensional space coordinate positions of the main nodes of the gas pipeline network, user access points, and pressure regulating stations; An extraction module, configured to select the main pipeline intersection point P1, user access point P2, and pressure regulating station output point P3 as three dynamic topological reference points based on the pipeline pressure gradient data, flow phase data, and user gas - using time - series fluctuation data; construct a dynamic topological unit according to the three points, perform regional division on the unit based on the Delaunay triangulation algorithm, and extract the pressure - flow coupling eigenvalue of the sub - region; A correction module, configured to perform weighted correction on the user gas - using demand mutation probability eigenvector through the pressure - flow coupling eigenvalue to generate a corrected mutation probability eigenvector; The processing module is used to perform tensor splicing on the corrected mutation probability feature vector and the pipeline equivalent inner diameter dynamic feature vector, and input the fully connected diagnostic network to generate a caliber matching deviation index; when the deviation exceeds a set threshold, the linkage timing prediction network outputs a safe redundant caliber replacement plan.

[0006] In a second aspect, a method for intelligent analysis and diagnosis of gas metering instrument caliber matching is provided, the method comprising: Step 1: By deploying pressure sensors at the three-dimensional spatial coordinate positions of the gas pipeline network trunk nodes, user access points and pressure regulating stations, pipeline pressure gradient data, flow phase data and user gas consumption time series fluctuation data are collected in real time; Step 2: Based on the pipeline pressure gradient data, flow phase data and user gas consumption time series fluctuation data, the main pipeline intersection point P1, the user access point P2 and the pressure regulating station output point P3 are selected as three vertices, and a dynamic triangle is formed according to the three vertices; the dynamic triangle is meshed based on the Delaunay triangulation algorithm to extract the pressure-flow coupling eigenvalue of the mesh; Step 3, weighted correction is performed on the user's gas demand mutation probability characteristic vector by using the pressure-flow coupling characteristic value to obtain a corrected mutation probability characteristic vector; Step 4: Perform tensor splicing on the corrected mutation probability feature vector and the pipeline equivalent inner diameter dynamic feature vector, and input them into the fully connected diagnostic network to generate a caliber matching deviation index; when the deviation exceeds the threshold, the linkage timing prediction network outputs a safe redundant caliber replacement plan.

[0007] According to a third aspect, a computing device includes: one or more processors; The storage device is used to store one or more computer programs. When the one or more computer programs are executed by the one or more processors, the one or more processors implement the system.

[0008] In a fourth aspect, a computer-readable storage medium stores a computer program, and the computer program implements the system when executed by a processor.

[0009] The above scheme of the present invention includes at least the following beneficial effects.

[0010] The acquisition module deploys pressure sensors at the main nodes of the gas pipeline network, user access points and pressure regulating stations to obtain real-time pipeline pressure gradient data, flow phase data and user gas consumption timing fluctuation data. Compared with the traditional method that relies on manual or static monitoring, the system can comprehensively capture dynamic information during the operation of the pipeline network and avoid caliber matching deviations caused by data missing or lags.

[0011] Based on the collected multi-dimensional data, the extraction module innovatively selects three dynamic topological reference points to construct a dynamic topological unit, and uses the Delaunay segmentation algorithm to divide the region and extract the pressure-flow coupling characteristic values ​​of the sub-region. This method can effectively integrate the spatial topological structure of the pipeline network with the fluid mechanics characteristics, and convert the complex pipeline network operation status into quantifiable characteristic parameters, which greatly improves the ability to analyze the operation laws of the pipeline network.

[0012] The correction module uses the pressure-flow coupling eigenvalue to perform weighted correction on the characteristic vector of the probability of sudden change in user gas demand, fully considering the impact of the pipeline network operation status on the user's gas consumption behavior. This dynamic correction mechanism can accurately capture sudden changes in user gas demand. Compared with the traditional fixed parameter prediction model, it can be more in line with the actual gas consumption scenario and avoid caliber matching errors caused by demand estimation deviations.

[0013] The processing module performs tensor splicing on the corrected mutation probability feature vector and the dynamic feature vector of the pipeline equivalent inner diameter, generates a caliber matching deviation index through a fully connected diagnostic network, and links the timing prediction network to output a safe redundant caliber replacement plan when the deviation exceeds the threshold. This process realizes fully automated intelligent diagnosis from data input to decision output, which not only greatly improves the efficiency of caliber matching diagnosis, but also provides reasonable caliber replacement plans based on scientific algorithms, effectively reducing pressure loss in the gas transportation process, improving metering accuracy, and reducing equipment wear and energy waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic diagram of an intelligent analysis and diagnosis system for caliber matching of gas metering instruments provided by an embodiment of the present invention.

[0015] Figure 2 It is a flow chart of a method for intelligent analysis and diagnosis of caliber matching of gas metering instruments provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0017] like Figure 1 As shown, an embodiment of the present invention provides an intelligent analysis and diagnosis system for gas metering instrument caliber matching, comprising: The acquisition module is used to collect pipeline pressure gradient data, flow phase data, and user gas consumption time series fluctuation data in real time through pressure sensors deployed at the three-dimensional spatial coordinate positions of the main nodes, user access points, and pressure regulating stations of the gas pipeline network; The extraction module is used to select the main pipeline intersection point P1, user access point P2, and pressure regulating station output point P3 as three dynamic topology reference points based on the pipeline pressure gradient data, flow phase data, and user gas consumption time series fluctuation data; construct a dynamic topology unit according to the three points, divide the area of the unit based on the Delaunay triangulation algorithm, and extract the pressure-flow coupling eigenvalue of the sub-region; The correction module is used to perform weighted correction on the user gas consumption demand mutation probability feature vector through the pressure-flow coupling eigenvalue to generate a corrected mutation probability feature vector; The processing module is used to perform tensor splicing on the corrected mutation probability feature vector and the pipeline equivalent inner diameter dynamic feature vector, and input it into the fully connected diagnosis network to generate a caliber matching deviation index; when the deviation exceeds the set threshold, link the time series prediction network to output a safe redundant caliber replacement plan.

[0018] In the embodiment of the present invention, the acquisition module can obtain pipeline pressure gradient, flow phase, and user gas consumption time series fluctuation data in real time by deploying pressure sensors at key positions of the gas pipeline network; compared with the traditional manual monitoring or intermittent data acquisition method, this system can realize the all-round and dynamic monitoring of the operation state of the pipeline network, effectively avoiding the caliber matching error caused by data lag or missing.

[0019] The extraction module constructs a topology unit with dynamic topology reference points, divides the area by using the Delaunay triangulation algorithm, and accurately extracts the pressure-flow coupling eigenvalue, fully considering the dynamic correlation between the pipeline network spatial topology and fluid mechanics, and can deeply explore the operation law of the pipeline network, converting the complex pipeline network working conditions into quantifiable and analyzable characteristic parameters. The correction module uses the pressure-flow coupling eigenvalue to perform weighted correction on the user gas consumption demand mutation probability feature vector, fully considering the influence of the pipeline network operation state on the user gas consumption behavior. This mechanism can dynamically adjust the gas consumption demand prediction in real time, accurately capture the sudden changes in the user gas consumption demand, and can be more in line with the actual gas consumption scenario compared with the traditional fixed parameter prediction model, effectively avoiding the caliber matching error caused by the demand estimation deviation. The processing module realizes the automatic generation of the caliber matching deviation index through tensor splicing and the fully connected diagnosis network, and links the time series prediction network to output a safe redundant caliber replacement plan when the deviation exceeds the threshold. This process is completely based on intelligent algorithms and models, realizing full-process automation from data processing to decision output, improving the diagnosis efficiency, effectively reducing the gas transmission pressure loss, improving the measurement accuracy, reducing equipment wear and energy waste, and ensuring the safe and stable operation of the gas supply system.

[0020] In a preferred embodiment of the present invention, pressure sensors deployed at the three-dimensional spatial coordinate positions of the main nodes, user access points, and pressure regulating stations of the gas pipeline network are used to collect pipeline pressure gradient data, flow phase data, and user gas consumption time series fluctuation data in real time, including: Establish a pipeline network topology mapping relationship based on the three-dimensional spatial coordinates of each sensor; Synchronously collect the real-time pressure values along the pipeline direction between the main nodes, calculate the pressure difference correlation spatial distance between adjacent nodes, and generate pressure gradient data; Monitor the output of the flow sensors at the user access points, extract the phase offset and amplitude characteristics of the flow waveform within a period, and generate flow phase data; Record the instantaneous fluctuation signal of the gas consumption downstream of the pressure regulating station, calculate the fluctuation frequency and the absolute difference between the peak and valley within a unit time, and form the user gas consumption time series fluctuation data.

[0021] In the embodiment of the present invention, when the above steps are specifically applied, they can be specifically implemented through the following steps. For example: Obtain the three-dimensional spatial coordinates of the pressure sensors deployed at the main nodes, user access points, and pressure regulating stations of the gas pipeline network. Each sensor has corresponding X, Y, and Z coordinate values, which reflect the spatial position of the sensor in the actual pipeline network. Then, according to the actual laying situation and connection relationship of the gas pipeline network, associate the sensors with connection relationships. For example, if two main nodes are connected by a pipeline, then mark the sensors corresponding to these two nodes as connected in space; for the user access points and pressure regulating stations, also determine their connection relationships with the sensors of the main nodes according to the actual positions where they are connected to the pipeline network, and finally construct the topology mapping relationship of the entire gas pipeline network, just like drawing a detailed pipeline network map with the connection conditions of each node marked.

[0022] By establishing the pipeline network topology mapping relationship, the spatial structure and node connection conditions of the gas pipeline network can be clearly presented. When calculating and analyzing data such as pressure and flow, the topological relationship can be combined to more accurately judge the data transfer path and mutual influence, improve the accuracy and reliability of data processing, and help comprehensively understand the overall operation status of the pipeline network.

[0023] Let the pressure sensors distributed at the backbone nodes start collecting data simultaneously to obtain the real-time pressure values of each backbone node at the same moment. Then, for every two adjacent backbone nodes, subtract the pressure of the node with a smaller pressure value from the pressure of the node with a larger pressure value to obtain the pressure difference between these two adjacent nodes. Next, measure the actual distance between two adjacent nodes in three-dimensional space, and correlate the pressure difference with the corresponding spatial distance. For example, if node A is adjacent to node B, the pressure difference between them is 5 units and the spatial distance is 10 meters, then this combination of the pressure difference and the spatial distance is taken as a set of data. Repeat this process to calculate the pressure differences and spatial distances between all adjacent backbone nodes, and integrate these data to generate the pressure gradient data reflecting the pressure change trend of the pipeline network.

[0024] The pressure gradient data of the present invention can intuitively reflect the pressure change situation when gas flows in the pipeline. By analyzing the pressure gradient data, it is possible to promptly detect whether there are pressure abnormal areas in the pipeline. For example, a sudden drop in pressure may mean problems such as pipeline leakage or blockage. This helps gas enterprises quickly locate the fault points and take corresponding maintenance measures to ensure the safety and stability of gas transmission.

[0025] Continuously monitor the flow data output by the flow sensors at the user access points. Over time, these data will form a continuously changing flow waveform. Select a fixed time period and observe the change situation of the flow waveform within this period. For the phase offset, compare it with the standard flow waveform under normal conditions to see whether the current flow waveform has shifted left or right on the time axis, and measure the length of the shifted time; for the amplitude characteristics, find the maximum and minimum values of the flow waveform within this period and calculate the difference between them. This difference is the change range of the flow amplitude. Record these two sets of data, namely the phase offset and the amplitude characteristics. For each user access point, repeat the above operations multiple times to obtain data for multiple periods, and summarize these data to generate the flow phase data.

[0026] The flow phase data can accurately reflect the change rules and characteristics of users' gas consumption. The phase offset can reflect the change in the gas consumption time of users, such as whether the gas consumption peak is advanced or postponed; the amplitude characteristics reflect the fluctuation range of the gas consumption volume of users. By analyzing these data, gas enterprises can better understand the gas consumption habits of users, rationally allocate gas resources, improve the efficiency and stability of gas supply, and at the same time help predict the future gas consumption demands of users.

[0027] Real-time record the gas consumption data downstream of the pressure regulating station. These data will generate instantaneous fluctuations over time, forming a series of fluctuation signals. Select a fixed time unit, such as 1 minute, and count the number of fluctuations in gas consumption within this 1 minute. This number is the fluctuation frequency per unit time. Then, find the maximum value (peak) and minimum value (trough) of the gas consumption within this time unit, and calculate the absolute difference between them. For example, within 1 minute, the maximum gas consumption is 20 cubic meters and the minimum is 10 cubic meters, then the peak-trough absolute difference is 10 cubic meters. Record the fluctuation frequency and peak-trough absolute difference within each time unit. Continuously count multiple time units and integrate these data together, which constitutes the user gas consumption time-series fluctuation data, and it can clearly show the fluctuation of user gas consumption in the time series.

[0028] The user gas consumption time-series fluctuation data of the present invention can intuitively present the instantaneous change situation of user gas consumption downstream of the pressure regulating station. The fluctuation frequency reflects the frequency of gas consumption changes, and the peak-trough absolute difference reflects the severity of gas consumption changes. By analyzing these data, gas enterprises can timely discover abnormal fluctuations in user gas consumption, take countermeasures in advance, ensure the stability and safety of gas supply, and at the same time help optimize the operating parameters of the pressure regulating station and improve the gas transmission efficiency.

[0029] In a preferred embodiment of the present invention, based on the pipeline pressure gradient data, flow phase data, and user gas consumption time-series fluctuation data, select the main pipeline intersection point P1, user access point P2, and pressure regulating station output point P3 as three dynamic topology reference points, including: According to the network topology mapping relationship, select the main pipeline intersection point with the largest pressure change rate in the pressure gradient data as P1, and use its three-dimensional space coordinates as the initial reference coordinates; Taking the initial reference coordinates of the main pipeline intersection point P1 as a reference, combining the phase offset and amplitude characteristics in the flow phase data, calculate the spatial influence weight of the user access point P2; perform offset compensation on the original three-dimensional coordinates of the user access point P2 according to the spatial influence weight to obtain the real-time monitoring coordinates of the weighted user access point P2; Based on the real-time monitoring coordinates of the weighted user access point P2, perform time-series displacement compensation on the original coordinates of the pressure regulating station output point P3 through the fluctuation frequency and peak-trough difference in the user gas consumption time-series fluctuation data to obtain the real-time regulation coordinates of the pressure regulating station output point P3; Synchronize the initial reference coordinates of the main pipeline intersection point P1, the real-time monitoring coordinates of the user access point P2, and the real-time regulation coordinates of the pressure regulating station output point P3 in space, and convert them to a two-dimensional plane through dimensionality reduction mapping; connect the three points P1, P2, and P3 in the two-dimensional plane to form a dynamic topology unit.

[0030] In the embodiment of the present invention, the above steps, when applied specifically, can be implemented by the following steps, for example: According to the established pipe network topology mapping relationship, all main pipe intersections are determined; then, the performance of each main pipe intersection in the pressure gradient data is checked, focusing on the change of pressure with spatial distance, that is, the pressure change rate. For example, the relationship between the pressure difference and distance between adjacent nodes at different intersections is compared. A large pressure change rate means that the pressure changes significantly within a short distance; find the main pipe intersection with the largest pressure change rate, determine it as P1, and record the three-dimensional spatial coordinates of the point as the initial reference coordinates for subsequent operations, which is equivalent to selecting a key reference "anchor point" in the pipe network.

[0031] The present invention selects the main pipe intersection point with the largest pressure change rate as P1 because this point is often located at a location where pressure fluctuations are more drastic in the pipe network, and may be a key node or potential problem area for gas transportation. Using it as the initial reference coordinate can provide a highly representative reference point for subsequent analysis, helping to quickly locate abnormal pressure changes in the pipe network.

[0032] Taking the three-dimensional coordinates of P1 as a reference, for each user access point P2, analyze the phase offset and amplitude characteristics in its flow phase data; the phase offset reflects the difference between the user's gas usage time and the normal situation, and the amplitude characteristics reflect the fluctuation of gas usage; taking these two factors into consideration, for example, the user access point with early gas usage time and large fluctuation may have a greater impact on the pipeline network, and give it a higher spatial impact weight; conversely, the weight is low for a small impact; after determining the spatial impact weight of each P2 point, adjust its original three-dimensional coordinates according to this weight. If the weight is large, the coordinates will be relatively offset in space according to certain rules; if the weight is small, the offset amplitude is small, and finally the weighted real-time monitoring coordinates of the user access point P2 are obtained, so that the coordinates can better reflect the degree of influence of this point on the actual operation of the pipeline network.

[0033] For example, the spatial influence weight determination process and value range of user access point P2 are as follows: Setting evaluation indicators and grading: Phase offset assessment divides the phase offset of gas consumption into five levels. Based on the normal gas consumption time, an offset time within ±15 minutes is considered "minor offset"; 15-30 minutes is "minor offset"; 30-60 minutes is "medium offset"; 60-120 minutes is "large offset"; and more than 120 minutes is "large offset". The longer the offset time, the greater the difference between the user's gas consumption time and the normal situation, and the more significant the impact on pipeline scheduling and pressure balance may be.

[0034] The amplitude characteristic assessment also divides the amplitude characteristics of gas consumption into five levels. The gas consumption fluctuation amplitude (the difference between the peak and the trough) within 5% of the rated gas consumption is "extremely small fluctuation"; 5%-15% is "small fluctuation"; 15%-30% is "medium fluctuation"; 30%-50% is "large fluctuation"; and more than 50% is "extreme fluctuation". The larger the amplitude, the more drastic the change in the user's gas consumption, and the greater the impact on the pipeline network.

[0035] Quantitative scoring and weight calculation: Assign a corresponding score to each level: "extremely small offset or fluctuation" corresponds to 1 point, "small offset / fluctuation" corresponds to 2 points, "medium offset / fluctuation" corresponds to 3 points, "large offset / fluctuation" corresponds to 4 points, and "large offset or fluctuation" corresponds to 5 points. For each user access point P2, obtain the level score corresponding to its phase offset and amplitude characteristics, and then add the two scores to obtain the comprehensive impact score of the point. For example, the phase offset of a user access point P2 is "large offset" (4 points), and the amplitude characteristic is "medium fluctuation" (3 points), and its comprehensive impact score is 7 points.

[0036] The comprehensive influence scores of all user access points P2 are normalized and mapped to the value range of [0, 1] to obtain the spatial influence weight of each P2 point. Assuming that the sum of the comprehensive influence scores of all user access points P2 is S, and the comprehensive influence score of a certain P2 point is s, then the spatial influence weight of the point w = s / S.

[0037] Weight value range: The weight is close to 0. When the phase offset of the user access point P2 is "extremely small offset" and the amplitude characteristic is "extremely small fluctuation", its spatial impact weight is close to 0. This type of user has stable gas consumption time and small fluctuation in gas consumption, and the impact on the operation of the pipeline network can be ignored.

[0038] 0<weight≤0.2. If the phase offset and amplitude characteristics are in the combination of "small offset", "small fluctuation" or lower levels, the weight is in this range. It means that the user's gas consumption behavior has little impact on the pipeline network, which is a normal and stable gas consumption situation.

[0039] 0.2<weight≤0.5, when there is a combination of "medium deviation" and "medium fluctuation" or below, or a combination of "large deviation or fluctuation" and "extremely small or small deviation or fluctuation", the weight is in this range. This indicates that the user's gas consumption behavior has a certain impact on the pipeline network, and it needs to be paid proper attention in pipeline network analysis and scheduling.

[0040] 0.5<weight≤0.8. If there is a combination of "large deviation" and "medium / large fluctuation", or a combination of "large deviation / fluctuation" and "small deviation or fluctuation", the weight is in this range, which means that the user's gas usage behavior has a greater impact on the pipeline network, which may cause local pressure fluctuations or changes in flow distribution in the pipeline network.

[0041] The weight is close to 1. When the phase offset of the user access point P2 is "extremely large offset" and the amplitude characteristic is "extremely large fluctuation", the weight is close to 1. This type of user gas consumption behavior has a great impact on the operation of the pipeline network, and it may be necessary to make targeted adjustments to the pipeline network or strengthen monitoring.

[0042] The present invention calculates the spatial impact weight and compensates the coordinates in combination with the flow phase data, fully considering the impact of the user's gas usage behavior on the pipeline network. The real-time monitoring coordinates of the user access point P2 obtained in this way are no longer simple geographical location coordinates, but incorporate the user's gas usage characteristics, which can more accurately reflect the actual role of the user access point in the operation of the pipeline network, and help gas companies to more accurately grasp the impact of user gas usage on the pipeline network and optimize gas distribution and scheduling strategies.

[0043] The number of fluctuations within a unit time (such as 1 hour) is divided into five levels: less than 5 times is "very low frequency", 5-15 times is "low frequency", 15-30 times is "medium frequency", 30-50 times is "high frequency", and more than 50 times is "very high frequency". The higher the frequency, the more frequent the user's gas consumption changes, and the pressure regulating station needs to adjust the output status more frequently.

[0044] The peak-to-valley difference classification is based on the rated output of the pressure regulating station, and the ratio of the peak-to-valley difference to the rated output is divided into five levels. The difference ratio is within 5% for "very small fluctuation", 5%-15% for "small fluctuation", 15%-30% for "medium fluctuation", 30%-50% for "large fluctuation", and more than 50% for "extremely large fluctuation". The larger the difference, the more drastic the change in user gas consumption, and the greater the impact on the output stability of the pressure regulating station.

[0045] Comprehensive impact assessment: An evaluation matrix is ​​constructed by combining the five levels of fluctuation frequency with the five levels of peak-to-trough difference to form a 5×5 evaluation matrix. For example, when the fluctuation frequency is "high frequency" and the peak-to-trough difference is "large fluctuation", a specific combination in the matrix is ​​corresponding.

[0046] Determine the impact level. According to different combination situations, the degree of influence of the output state of the pressure regulating station by users is divided into five levels. For example, combinations such as "high frequency + large fluctuation", "extremely high frequency + medium fluctuation" and above correspond to the "extremely large impact" level; combinations such as "high frequency + medium fluctuation", "medium frequency + large fluctuation" correspond to the "relatively large impact" level, and so on, until "extremely low frequency + extremely small fluctuation" corresponds to the "extremely small impact" level.

[0047] Determine the compensation direction: Positive compensation (time axis shifted backward). When there are situations such as the peak of user gas consumption advancing and the fluctuation frequency increasing, it means that the pressure regulating station needs to respond to these changes earlier. At this time, a positive displacement compensation is performed on the original coordinates of P3 on the time axis, and the coordinate point is moved backward.

[0048] Negative compensation (time axis shifted forward). If the peak of user gas consumption is delayed and the fluctuation frequency decreases, it indicates that the response of the pressure regulating station can be appropriately lagged. At this time, a negative displacement compensation is performed on the original coordinates of P3 on the time axis, and the coordinate point is moved forward.

[0049] Calculate the compensation amplitude: Set the basic compensation unit. Set a basic compensation time unit, such as 5 minutes. This unit represents the time displacement amount of the coordinates of P3 under the minimum impact level (extremely small impact).

[0050] Adjust the grading multiple. According to the five levels of the comprehensive impact degree, set the corresponding compensation multiples respectively. "Extremely small impact" corresponds to 1 times the basic unit, "small impact" corresponds to 2 times, "medium impact" corresponds to 4 times, "relatively large impact" corresponds to 8 times, and "extremely large impact" corresponds to 16 times. For example, when the comprehensive impact is "extremely large impact", the compensation amplitude is 16×5 = 80 minutes.

[0051] Determine the final compensation amplitude. On the basis of the grading multiple, make fine-tuning in combination with the specific value of the peak-valley difference. If the difference is in the upper half of the grade interval, take the upper limit value of the compensation multiple of this grade; if it is in the lower half, take the lower limit value. For example, under the "extremely large impact" level, when the difference is in the upper half, compensate 80 minutes, and when it is in the lower half, compensate 70 minutes.

[0052] Coordinate adjustment and generation of real-time control coordinates: Displacement in the time dimension. According to the determined compensation direction and amplitude, displace the original coordinates of P3 in the time dimension. For example, if the compensation direction is positive and the amplitude is 60 minutes, increase the time coordinate value of P3 by 60 minutes; if it is negative, reduce the corresponding time value.

[0053] Spatial coordinate synchronous adjustment. While displacing in the time dimension, consider the impact of the user's gas consumption fluctuations on the spatial state of the pressure regulating station. If the fluctuations cause significant changes in the output pressure or flow rate of the pressure regulating station, make minor adjustments to the coordinates of P3 in the spatial dimension, and the adjustment amplitude is proportional to the degree of influence of the fluctuations.

[0054] Real-time regulation coordinate generation. After adjustments in the time and spatial dimensions, obtain the real-time regulation coordinates of the output point P3 of the pressure regulating station. These coordinates not only reflect the time correlation between the output state of the pressure regulating station and the user's gas consumption fluctuations but also embody the impact of the fluctuations on the spatial state of the pressure regulating station.

[0055] The present invention uses the time-series fluctuation data of the user's gas consumption to perform time-series displacement compensation on the coordinates of the output point P3 of the pressure regulating station, closely linking the dynamic changes of the user's gas consumption with the regulation of the pressure regulating station, enabling the real-time regulation coordinates of P3 to reflect the response of the pressure regulating station to the user's gas consumption fluctuations, which helps gas enterprises better understand the operating state of the pressure regulating station, ensure the stability and safety of gas supply, and achieve refined regulation of the pipe network.

[0056] The process of constructing the dynamic topology unit specifically includes: Spatial unified calibration: Coordinate alignment. First, determine a unified spatial origin. For example, select the geometric center of the entire gas pipe network coverage area as the origin. Then, for the initial reference coordinates of P1, the real-time monitoring coordinates of P2, and the real-time regulation coordinates of P3, calculate their offsets relative to this origin respectively. For example, if the original coordinates of P1 are (X1, Y1, Z1), calculate the distances it needs to translate in the X, Y, and Z directions with the new origin as the reference, and adjust its coordinates to the coordinate system based on the new origin. In the same way, perform translation operations on the coordinates of P2 and P3 to make the coordinates of the three points based on the same origin and achieve preliminary alignment.

[0057] Scale unification. Check whether the scales of the three points in the X, Y, and Z axis directions are the same. Since there may be differences in measurement units or precisions in different regions of the actual pipe network, it is necessary to unify the scales. For example, if it is found that the measurement unit of the Z-axis coordinate of P1 is meters, while the measurement units of the Z-axis coordinates of P2 and P3 are centimeters, divide the Z-axis coordinate values of P2 and P3 by 100 to convert them to meters, ensure that the scales of the three points in the three axis directions are the same, complete the spatial calibration, and make them in the same reference system.

[0058] Dimensionality reduction mapping to a two-dimensional plane: Determine the projection direction. Select a suitable projection direction. Commonly, the direction perpendicular to the main laying plane of the pipe network can be selected as the projection direction. For example, if the gas pipe network is mainly laid on the underground plane, the vertically upward direction can be selected as the projection direction.

[0059] Projection operation: Project the three-dimensional coordinates of the calibrated points P1, P2, and P3 onto a two-dimensional plane along the selected projection direction. Imagine a beam of parallel light shining on the three points from the projection direction, and their shadows on the two-dimensional plane are the projected coordinate positions. Calculate the corresponding X and Y coordinate values of the three points on the two-dimensional plane respectively, and discard the Z coordinate value to obtain the new coordinates of the three points on the two-dimensional plane.

[0060] Coordinate adjustment and optimization: Check and adjust the two-dimensional coordinates after projection to ensure that the relative position relationship between points conforms to the spatial relationship in the actual pipe network. If it is found that the positions of some points after projection show unreasonable overlap or offset, fine-tune the coordinate values so that their layout on the two-dimensional plane can more accurately reflect the relative positions of the three points in the actual pipe network for subsequent analysis.

[0061] Construct a dynamic topological unit: Connection operation: On the two-dimensional plane, use the line segment tool to connect the three points P1, P2, and P3 in sequence. Start from point P1 and draw a line segment to connect to point P2, then draw a line segment from point P2 to connect to point P3, and finally draw a line segment from point P3 to connect back to point P1 to form a closed triangular region.

[0062] Information integration and annotation: Label the data information related to the three points collected previously, such as the pressure change rate of point P1, the spatial influence weight of point P2, and the basis for time series displacement compensation of point P3, in the corresponding points or triangular regions. You can also add notes to explain the relationship between these data and the points, as well as their significance to the entire dynamic topological unit, so that this triangular region not only becomes a geometric figure but also a visual analysis unit integrating important information of key nodes in the pipe network, which can intuitively present the spatial relationship and mutual influence between nodes.

[0063] The present invention integrates and simplifies the complex spatial relationships and operation data of key nodes in the pipe network by constructing a dynamic topological unit. Through this unit, the associations between the intersection points of the main pipes, user access points, and the output points of the pressure regulating stations can be intuitively analyzed, and the operation characteristics of local areas of the pipe network can be quickly grasped. It helps to analyze the gas pipe network from the perspective of combining the whole and the part, provides a clear analysis framework for subsequent operations such as regional division based on the Delaunay triangulation algorithm and extraction of pressure-flow coupling characteristic values, and improves the analysis efficiency and accuracy of the operation state of the pipe network.

[0064] In a preferred embodiment of the present invention, a dynamic topological unit is constructed according to three points, and the unit is divided into regions based on the Delaunay triangulation algorithm, and the pressure-flow coupling characteristic values of the sub-regions are extracted, including: Perform the Delaunay triangulation algorithm on the dynamic topology unit to generate a minimum angle division structure containing at least three sub-regions, including: using the real-time two-dimensional plane coordinates of points P1, P2, and P3 in the dynamic topology unit as the triangulation input point set to construct the initial convex hull boundary; based on the real-time change amplitude of the user gas consumption time series fluctuation data, when the fluctuation amplitude exceeds the set threshold, dynamically insert an auxiliary vertex Q1 at the midpoint of the line connecting P2 and P3; according to the phase offset of the flow phase data, insert an auxiliary vertex Q2 at the midpoint of the line connecting P1 and P2; add Q1 and Q2 to the triangulation point set and update the convex hull to obtain the updated triangulation point set; traverse the updated triangulation point set and perform diagonal exchange detection on any quadrilateral formed by four adjacent points; if the minimum interior angle increases after the exchange, update the topological connection relationship; iterate and optimize until all quadrilaterals satisfy the Delaunay empty circle criterion to generate the final triangulation structure; define each minimum topological unit in the final triangulation structure as a sub-region to obtain the sub-region vertex coordinate set and the adjacency relationship table; For each divided sub-region, extract the pipeline pressure gradient data associated with the vertex coordinates of the sub-region, and synchronously obtain the amplitude characteristics of the flow phase data in the same spatial domain; Calculate the arithmetic mean of the product of the pressure gradient data and the flow phase amplitude characteristics, and output the basic spatial coupling factor of the sub-region; Based on the basic spatial coupling factor, locate the gas consumption time series fluctuation data of the user access points under the jurisdiction of the sub-region, calculate the statistical variance value of the fluctuation data of the users under the jurisdiction within a continuous time window, and perform linear normalization processing on the variance value in the interval [0, 1] to generate a time series correction coefficient; Add the basic spatial coupling factor and the time series correction coefficient to generate an intermediate coupling feature; Multiply the intermediate coupling feature by the spatial weight coefficient of the sub-region in the pipe network topology, and finally output the pressure-flow coupling feature value.

[0065] In the embodiments of the present invention, when the above steps are specifically applied, they can be specifically implemented through the following steps, for example: Based on the two-dimensional coordinates of points P1, P2, and P3, judge the relative position relationship of these three points. Since there are only three points, they are either collinear or can form a triangle. In actual pipe network analysis, these three points are usually not collinear, so directly connect these three points to form a triangle, which is the smallest convex polygon that can enclose these three points, that is, the initial convex hull.

[0066] Mark the boundary direction, and mark the three sides of the triangle in clockwise or counterclockwise order to determine the connection order of each vertex.

[0067] Dynamically insert auxiliary vertices: Detect the fluctuation conditions, monitor the change amplitude of the real-time user gas consumption timing fluctuation data, and compare it with a pre-set threshold. For example, set the threshold to 1.5 times the average fluctuation amplitude. When the actual fluctuation amplitude exceeds this threshold, it indicates that the gas consumption situation in this area is unstable.

[0068] Insert vertex Q1. When the fluctuation amplitude exceeds the threshold, calculate the midpoint coordinates of the line connecting P2 and P3. The specific method is to take the average of the coordinate values of P2 and P3 on the X-axis and Y-axis respectively, and the new coordinate point obtained is Q1. Insert Q1 into the point set. At this time, the point set contains four points: P1, P2, Q1, P3.

[0069] Detect the phase shift, analyze the phase shift amount of the flow phase data, and determine whether it reaches a certain degree. For example, when the phase shift amount exceeds 20% of the normal fluctuation range, it is considered that a more detailed division of this area is required.

[0070] Insert vertex Q2. If the phase shift amount reaches the set degree, calculate the midpoint coordinates of the line connecting P1 and P2, and the method is similar to that of calculating Q1. The new coordinate point obtained is Q2. Insert Q2 into the point set as well, forming a new triangulation point set containing P1, P2, P3, Q1, and Q2.

[0071] Update the convex hull boundary: Re-determine the peripheral points. After adding Q1 and Q2, re-check the positional relationship of these five points, and find the points located on the outermost periphery. These points form the new convex hull boundary. The specific method is to start from any point and check each point in a certain direction (such as clockwise) in turn, and determine whether there are other points outside the line connecting this point and its adjacent point. If so, adjust the convex hull boundary until all peripheral points are determined.

[0072] Connect the new boundary. Connect the determined peripheral points in turn to form a new convex polygon, and this convex polygon is the smallest convex hull boundary containing five points. For example, if the order of the new peripheral points is P1, Q2, P2, Q1, P3, then connect these five points in turn to form a pentagonal convex hull boundary.

[0073] Diagonal exchange detection and optimization: Traverse the quadrilaterals. In the updated triangulation point set, find any four adjacent points that form a quadrilateral. For example, there may be a quadrilateral formed by P1, Q2, P2, Q1, or a quadrilateral formed by Q2, P2, Q1, P3, etc.

[0074] Detect the effect of diagonal exchange. For each quadrilateral, consider its two diagonals separately. For example, for the quadrilateral P1-Q2-P2-Q1, the two diagonals are P1-P2 and Q2-Q1. Calculate the minimum interior angle of the new triangle formed after exchanging the diagonals, and compare the sizes of the minimum interior angles before and after the exchange.

[0075] Update the topological connection. If the minimum interior angle increases after the diagonal exchange, it means that the new connection method better meets the requirements of Delaunay triangulation. Replace the original diagonal with the new diagonal and update the topological connection relationship of the quadrilateral. For example, if the minimum interior angle increases after exchanging P1-P2 and Q2-Q1, split the quadrilateral P1-Q2-P2-Q1 into two triangles: P1-Q2-Q1 and Q2-P2-Q1.

[0076] Iterative optimization. Repeat the above diagonal exchange detection process, check and optimize all possible quadrilaterals until all quadrilaterals meet the Delaunay empty circle criterion, that is, there are no other points inside the circumcircle of each triangle. At this time, the triangulation structure reaches the optimal state and the final triangulation structure is generated.

[0077] Define sub-regions and adjacency relationships: Divide the triangular sub-regions. Define each triangle in the final triangulation structure as a sub-region and record the coordinates of the three vertices of each sub-region. For example, a sub-region may be composed of vertices P1, Q2, Q1, and another sub-region may be composed of vertices Q2, P2, Q1, etc.

[0078] Establish an adjacency relationship table. Check the shared edge situation of each sub-region with other sub-regions and record the numbers of adjacent sub-regions. For example, if sub-region A and sub-region B share an edge, record in the adjacency relationship table that A is adjacent to B. In this way, a complete adjacency relationship table is formed to describe the spatial connection relationship between each sub-region.

[0079] Calculation process for extracting the pressure-flow coupling eigenvalue of sub-regions: Search for pressure values. For each sub-region, sequentially view the coordinates of its three vertices. In the originally collected and stored raw pipeline pressure gradient data, find the pressure values corresponding to these three vertex coordinates. For example, if the three vertex coordinates of the sub-region are (X1, Y1), (X2, Y2), (X3, Y3) respectively, search for the pressure values recorded at these three coordinate points in the pressure gradient data and extract them one by one.

[0080] Flow phase amplitude acquisition: In the same spatial range, that is, the pipe network area covered by the sub-area, find the flow phase data, and extract the amplitude characteristics of the flow waveform from these data, that is, the difference between the maximum and minimum values ​​of the flow waveform. For example, if the maximum value of the flow waveform in a sub-area is 100 cubic meters / hour and the minimum value is 20 cubic meters / hour, then the flow phase amplitude characteristic of the area is 80 cubic meters / hour.

[0081] Calculate the basic spatial coupling factor: Data multiplication: The extracted pressure gradient data in the sub-region is multiplied with the flow phase amplitude characteristics according to the corresponding relationship. For example, the pressure values ​​corresponding to the three vertices of the sub-region are P1, P2, and P3, and the flow phase amplitude characteristic is A. Then P1×A, P2×A, and P3×A are calculated respectively to obtain three sets of product data.

[0082] To find the arithmetic mean, add the three sets of product data and divide them by the number of data (3 here) to calculate their arithmetic mean. This average is the basic spatial coupling factor of the sub-area, which reflects the degree of correlation between pressure and flow in the sub-area in the spatial dimension.

[0083] Generate timing correction factors: Locate the fluctuation data and find the user access points under its jurisdiction according to the scope of the sub-area. Among the gas usage timing fluctuation data of these user access points, filter out the fluctuation data belonging to the sub-area. For example, if sub-area A covers 3 user access points, all the gas usage timing fluctuation data of these 3 user access points will be extracted.

[0084] Calculate the variance value. For the extracted fluctuation data, calculate the statistical variance value in the continuous time window. The calculation of the variance value is to measure the degree of dispersion of these fluctuation data. The larger the variance value, the more drastic the fluctuation of user gas consumption. For example, by calculating the gas consumption data over a period of time, the variance value of the gas consumption fluctuation data of users in this sub-area is 25.

[0085] Linear normalization is to transform the calculated variance value linearly and map it to the interval [0, 1]. The specific operation is to scale the variance value according to a certain ratio based on its relative size in all sub-region variance values. For example, if the maximum variance value in all sub-regions is 100 and the minimum is 0, and the variance value of the current sub-region is 25, after calculation and conversion, the value obtained is in the interval [0, 1], and this value is the timing correction coefficient.

[0086] Create an intermediate coupling feature: Add the data. Add the previously calculated basic spatial coupling factor and the timing correction coefficient to obtain a new value, which is the intermediate coupling eigenvalue. In this way, the pressure-flow correlation characteristics in the spatial dimension and the user gas consumption fluctuation characteristics in the time dimension are preliminarily integrated.

[0087] Calculate the final eigenvalue: Allocate spatial weight coefficients. According to the importance of each sub-region in the entire pipe network topology, assign a spatial weight coefficient to it. The judgment of importance can refer to factors such as whether the sub-region is close to the main pipe and whether it contains key user access points. For example, a sub-region close to the pressure regulating station and with a dense user population may be assigned a higher weight coefficient of 0.8; while a sub-region at the end of the pipe network with fewer users may be assigned a lower weight coefficient of 0.2.

[0088] Calculate the final value. Multiply the intermediate coupling eigenvalue by the spatial weight coefficient assigned to this sub-region. The result obtained is the final pressure-flow coupling eigenvalue. This eigenvalue comprehensively considers the spatial characteristics of the sub-region, the time characteristics of user gas consumption, and the importance of this sub-region in the pipe network, and can comprehensively reflect the coupling relationship between pressure and flow in the sub-region.

[0089] By combining the pipeline pressure gradient data, the flow phase amplitude characteristics, and the user gas consumption timing fluctuation data, comprehensively evaluate the operating state of the pipe network from both spatial and time dimensions, changing the limitations of the previous single-dimensional analysis, and more comprehensively reflecting the impact of gas flow characteristics and user gas consumption behavior on the pipe network; the basic spatial coupling factor reflects the spatial correlation between pressure and flow in the sub-region, and the timing correction coefficient reflects the fluctuation characteristics of user gas consumption in time. The intermediate coupling characteristics generated by the combination of the two can accurately depict the unique operating characteristics of each sub-region. Combined with the spatial weight coefficient, it highlights the importance of key regions in the pipe network, making the analysis results more in line with the actual operating conditions of the pipe network, and helping to discover weak links and potential problem areas in the pipe network.

[0090] By dynamically inserting auxiliary vertices at the midpoints of the lines connecting P2 and P3, and P1 and P2, the subdivision structure can be adaptively adjusted according to the temporal fluctuations of user gas consumption and the flow phase offset, more precisely capturing the pressure and flow change-sensitive areas in the pipe network and improving the analysis accuracy; the triangulation structure generated by the Delaunay triangulation algorithm has good geometric properties (maximizing the minimum interior angle), avoiding the appearance of long and narrow triangles, making the sub-region division more uniform and reasonable, and facilitating subsequent feature extraction and analysis; by combining pressure gradient data, flow phase amplitude characteristics, and temporal fluctuations of user gas consumption data, the operating state of the pipe network is comprehensively evaluated from both spatial and temporal dimensions, more comprehensively reflecting the gas flow characteristics and the impact of user gas consumption behavior on the pipe network; by assigning spatial weight coefficients to sub-regions, the influence of sub-regions where key nodes in the pipe network (such as near pressure regulating stations and user-dense areas) are located can be highlighted.

[0091] In a preferred embodiment of the present invention, the user gas demand mutation probability eigenvector is weighted and corrected by the pressure-flow coupling eigenvalue to generate a corrected mutation probability eigenvector, including: Input the pressure-flow coupling eigenvalues of each sub-region into a matrix converter, sort them according to the spatial weight coefficients of the sub-regions in the pipe network topology, and scale the eigenvalues with the pipe length under the jurisdiction of each sub-region as the normalization factor to generate a diagonal correction coefficient matrix with the same dimension as the mutation probability eigenvector; According to the preset user gas demand mutation probability eigenvector, perform a Hadamard product operation on the correction coefficient matrix and the mutation probability eigenvector to obtain a weighted eigenvector; Perform L2 norm normalization processing on the weighted eigenvector to obtain a corrected mutation probability eigenvector.

[0092] In the embodiment of the present invention, when the above steps are specifically applied, they can be specifically implemented through the following steps, for example: Sort the pressure-flow coupling eigenvalues calculated for each sub-region according to the magnitude of the spatial weight coefficients of the sub-regions in the pipe network topology. The sub-region with a higher spatial weight coefficient will have its eigenvalue ranked higher in the sorting; for example, if the spatial weight coefficient of sub-region A is 0.8, sub-region B is 0.6, and sub-region C is 0.4, then the eigenvalue sorting is A, B, C.

[0093] Normalize and scale, using the pipe length under the jurisdiction of each sub-region as the normalization factor, scale the sorted eigenvalues. The specific method is to divide the eigenvalue of each sub-region by the pipe length under the jurisdiction of that sub-region to obtain the scaled eigenvalue. For example, if the eigenvalue of sub-region A is 10 and the pipe length under its jurisdiction is 5 kilometers, then the scaled value is 10÷5 = 2.

[0094] Generate a diagonal matrix, arrange the scaled eigenvalues in sequence on the diagonal of the matrix, and fill the other positions with 0 to form a diagonal matrix, whose dimension is the same as that of the user gas demand mutation probability eigenvector.

[0095] Hadamard product operation, specifically including: Prepare the original vector, obtain the preset user gas demand mutation probability eigenvector, which contains multiple dimensions, and each dimension represents the mutation probability of different types of gas demand.

[0096] Element-wise multiplication, perform Hadamard product operation on the diagonal correction coefficient matrix and the mutation probability eigenvector, that is, multiply the elements at the corresponding positions. For example, multiply the first diagonal element of the diagonal matrix by the first element of the eigenvector, the second diagonal element by the second element, and so on, to obtain a new vector, which is the weighted eigenvector, and each element in it has been corrected by the pressure-flow coupling eigenvalue.

[0097] L2 norm normalization: Calculate the vector norm. For the weighted eigenvector, calculate its L2 norm, that is, the square root of the sum of the squares of each element of the vector; divide each element of the weighted eigenvector by its L2 norm to obtain the corrected mutation probability eigenvector. After normalization, the norm of the vector is 1, which is convenient for subsequent analysis and comparison, and at the same time maintains the relative proportional relationship between the elements of the vector.

[0098] In the embodiment of the present invention, by integrating the pressure-flow coupling eigenvalue into the correction of the mutation probability eigenvector, the influence of the actual operation state of the pipe network on the user gas demand is fully considered. The changes in the pipe network pressure and flow are often precursors to the changes in user demand. This data-driven correction method can more accurately capture the mutation trend of user gas demand and improve the accuracy of the prediction model. The application of the spatial weight coefficient makes the influence of the pressure-flow characteristics in the key areas of the pipe network (such as near the pressure regulating station and the user-dense area) on the mutation probability greater. This helps to highlight the potential risks in these areas, enabling the gas enterprise to monitor and regulate more targeted, and prevent the possible large fluctuations in gas demand in advance; the normalization process eliminates the dimensional influence caused by the differences in the pipe lengths of different sub-regions, making the eigenvalues of each sub-region comparable. At the same time, the L2 norm normalization ensures that the corrected eigenvector is within a reasonable range, which is convenient for subsequent fusion analysis with other models or indicators, and improves the overall stability and reliability of the system.

[0099] In a preferred embodiment of the present invention, tensor splicing is performed on the corrected mutation probability eigenvector and the dynamic eigenvector of the equivalent inner diameter of the pipeline, and the full connection diagnosis network is input to generate the caliber matching deviation index, including: According to the corrected mutation probability eigenvector and the real-time updated dynamic eigenvector of the pipeline equivalent inner diameter, expand them into a three-dimensional tensor along the feature dimension. Among them, the first dimension is the time step, the second dimension is the spatial partition, and the third dimension is the feature channel; Input the three-dimensional tensor into a three-layer fully connected diagnostic network, where: The first layer, the linear transformation layer reduces the input feature dimension to 1 / 2 of the original dimension, through the ReLU activation function; The second layer, the linear transformation layer further reduces it to 1 / 4 of the original dimension, through the Sigmoid activation function; The third layer, a single-neuron linear layer outputs the original value of the deviation; Perform dynamic threshold calibration on the original value output by the third layer. Through the reference distribution curve fitted by the preset historical normal operating condition data, calculate the Z-score standardized offset of the current value relative to the reference distribution.

[0100] In the embodiments of the present invention, when the above steps are specifically applied, they can be specifically implemented through the following steps. For example: Obtain the corrected mutation probability eigenvector and the real-time updated dynamic eigenvector of the pipeline equivalent inner diameter. The former reflects the possibility of sudden changes in the user's gas consumption demand, and the latter reflects the change of the pipeline inner diameter with the operation state of the pipe network.

[0101] Expand these two eigenvectors into a three-dimensional tensor along the feature dimension. First, determine the time step as the first dimension. For example, take 1 hour as a time step to record data at different times; the second dimension is the spatial partition, corresponding to the previously divided pipe network sub-regions, and each sub-region has corresponding data records; the third dimension is the feature channel, which stores the mutation probability feature and the dynamic feature of the pipeline equivalent inner diameter respectively; data filling, in the order of time step, spatial partition, and feature channel, fill the corresponding data into each position of the three-dimensional tensor in turn to form a complete three-dimensional data structure.

[0102] Fully connected diagnostic network operation: The first layer operation, input the constructed three-dimensional tensor into the first layer of the fully connected diagnostic network. This layer is a linear transformation layer, which will process the input feature dimension and reduce it to 1 / 2 of the original dimension. After the processing is completed, through the ReLU activation function, this function will directly change all negative eigenvalue features to 0 and retain the positive eigenvalue features, so as to screen out more valuable features and enhance the non-linear expression ability of the network.

[0103] The second - layer operation: The data processed in the first layer enters the second layer, which is also a linear transformation layer. This layer further reduces the feature dimension to 1 / 4 of the original dimension, performing a deeper compression and feature extraction on the data. Then, through the Sigmoid activation function, which maps the feature values to the range between 0 and 1, the features are transformed into a probability form.

[0104] The third - layer operation: The data output from the second layer enters the third single - neuron linear layer. This layer performs the final processing on the input data and outputs an original deviation value, which reflects the preliminary evaluation result of the gas meter caliber matching situation under the current pipeline network state.

[0105] Dynamic threshold calibration: Select all records labeled as "normal operating conditions" from the historical data, exclude data in abnormal states such as pipeline network leakage, equipment failures, and extreme gas - usage peaks. Clean the selected data to remove missing values and obvious outliers, ensuring the integrity and reliability of the data.

[0106] Statistical feature calculation: Calculate the statistics of the caliber - matching - related indicators in the normal - operating - condition data, including the mean (reflecting the average level in the normal state), standard deviation (reflecting the degree of data dispersion), quantiles (such as the 25th percentile and 75th percentile, used to describe the range of data distribution), etc.

[0107] Normal distribution fitting calculation process: Extract the original data set of caliber - matching - related indicators (such as the original deviation value of historical data) from the historical normal - operating - condition data, ensuring that the data only contains records of stable pipeline network operation without abnormal events; sort the data set in ascending order of numerical values to facilitate subsequent observation of the data distribution trend. For example, arrange 1000 deviation data under normal operating conditions in ascending order and observe their central tendency and dispersion range.

[0108] Draw a histogram: Divide the sorted data into several intervals (such as with a bin width of 0.1), count the frequency of data in each interval, and draw a histogram. Initially judge whether the data presents a bell - shaped distribution feature of "high in the middle and low on both sides" through the histogram, that is, whether it conforms to the visual feature of the normal distribution.

[0109] Calculate the key parameters of the normal distribution: Calculate the mean μ: Add all the data and divide by the total number of data to obtain the mean. The mean represents the central position of the normal distribution. For example, if the mean of 1000 data is 1.2, it means that the average level of the deviation index under normal operating conditions is 1.2.

[0110] Calculate the standard deviation σ. First, calculate the square of the difference between each data point and the mean. Then sum all the squared values and divide by the total number of data points to obtain the variance. Next, take the square root of the variance to get the standard deviation. The standard deviation reflects the degree of dispersion of the data. For example, if the standard deviation is 0.3, it means that under normal operating conditions, the deviation index usually fluctuates within the range of mean ± 0.3.

[0111] Generate a reference distribution curve: Determine the curve shape. Based on the calculated mean and standard deviation, draw a bell-shaped curve centered at the mean with the standard deviation as the width. The highest point of the curve corresponds to the mean position, and the curve gradually decreases on both sides. Theoretically, it extends to positive and negative infinity, but in practice, about 95% of the data falls within the range of mean ± 2 standard deviations, and about 99.7% of the data falls within the range of mean ± 3 standard deviations.

[0112] Mark key positions. Mark key positions such as the mean μ, mean ± 1 standard deviation (μ ± σ), mean ± 2 standard deviations (μ ± 2σ), etc. on the curve to clarify the distribution range of the data under normal operating conditions. For example, if the mean is 1.2 and the standard deviation is 0.3, then the curve contains approximately 68% of the data between 0.9 (μ - σ) and 1.5 (μ + σ), and approximately 95% of the data between 0.6 (μ - 2σ) and 1.8 (μ + 2σ).

[0113] In the present invention, the probability density function curve of the normal distribution is superimposed on the histogram to observe whether the shape of the curve basically coincides with that of the histogram. For example, if the peak of the histogram coincides with the peak position of the curve, and the frequency attenuation trend on both sides is the same as that of the curve, it indicates a good fitting effect. By visual observation or simple statistical methods (such as calculating whether the actual proportion of data within μ ± σ, μ ± 2σ is close to the theoretical proportions of 68%, 95%), verify whether the fitted normal distribution can reasonably describe the data distribution characteristics. If the actual proportion is close to the theoretical proportion, the fitting is considered valid. By fitting the normal distribution, the index distribution under normal operating conditions is transformed into two key parameters, the mean and the standard deviation, clearly defining the numerical range of the "normal" state (such as μ ± 2σ). Operators can quickly judge whether the current data falls within the normal range. For example, when the deviation index at a certain moment is 2.0, and μ = 1.2, σ = 0.3, 2.0 > μ + 2σ × 1.8, it can be immediately identified as abnormal.

[0114] Unify the abnormal judgment standard. The standardization characteristics of the normal distribution (such as Z-score) enable indicators with different dimensions and value ranges to be uniformly measured by the "number of standard deviation multiples from the mean" to represent the degree of deviation. For example, whether it is a pressure difference indicator (unit: kPa) or a flow rate fluctuation indicator (unit: m 3 / h), they can be converted into comparable abnormal levels through Z-score, avoiding judgment confusion caused by differences in indicator units.

[0115] The reference distribution curve is not fixed. The mean and standard deviation can be recalculated regularly based on newly collected normal operating condition data to update the curve shape, enabling the system to adapt to long-term changes in the pipeline network operation (such as user structure adjustment and equipment aging), and continuously maintaining the timeliness of the abnormal judgment criteria. For example, during the winter heating period, the increase in user gas consumption may cause the overall upward shift of the mean and standard deviation of the deviation index. After refitting, it can avoid misjudging normal fluctuations as abnormalities.

[0116] Based on the "3σ principle" of the normal distribution (i.e., 99.7% of the data falls within the range of μ±3σ), the situation where the absolute value of the Z-score is greater than 3 can be defined as "extremely high risk", automatically triggering an emergency warning. This warning logic based on statistical laws does not require manual experience intervention, reduces subjective judgment errors, and at the same time ensures a rapid response to extreme abnormal states.

[0117] Visual verification: Compare the fitted distribution curve with the histogram or probability density map of the actual data to check whether the fitting effect meets the expectations and ensure that the curve can accurately reflect the distribution law of the indicators under normal operating conditions.

[0118] Calculate the standardized offset (Z-score): Locate the position of the original value. Substitute the original deviation value output by the fully connected network into the mathematical expression of the reference distribution curve to determine its relative position in the reference distribution. For example, if the reference distribution is a normal distribution and the original value is on the right side of the mean, it indicates that the current state deviates from the normal level in the "positive" direction.

[0119] Calculate the degree of deviation. According to the mean μ and standard deviation σ of the reference distribution, calculate the difference between the original value x and the mean (x - μ), and then divide the difference by the standard deviation to obtain the standardized offset (Z-score). This value represents the multiple of the standard deviation that the original value is away from the mean. For example, Z = 1 means the original value is 1 standard deviation higher than the mean, and Z = -1 means it is 1 standard deviation lower than the mean.

[0120] Map it to the deviation index. Map the standardized offset to a specific deviation index according to business requirements. For example, set the correspondence between the absolute value range of the Z-score and the deviation level: |Z|≤1 is "normal", 1<|Z|≤2 is "slight deviation", and |Z|>2 is "severe deviation", which is convenient for operators to intuitively understand the current state of the pipeline network.

[0121] The present invention realizes the data fusion of time, space, and different feature dimensions by constructing a three-dimensional tensor from the corrected mutation probability feature vector and the dynamic feature vector of the equivalent inner diameter of the pipeline. This fusion method can comprehensively capture the dynamic information of the changes in user demands and pipeline states during the operation of the pipe network. The fully connected diagnostic network gradually extracts features and reduces the dimension of the input data through multi-layer linear transformation and activation function processing. It not only effectively reduces the data dimension and computational complexity but also retains key features, highlights the factors that have an important impact on diameter matching, and improves the accuracy and efficiency of analysis. The dynamic threshold calibration based on the reference distribution curve fitted from historical normal operating condition data can adaptively adjust the evaluation criteria of the deviation index according to the actual situation of the pipe network operation. Compared with the method of fixed threshold, this method can better adapt to the changes in the operation state of the pipe network, accurately identify abnormal conditions, avoid false judgments and missed judgments, and improve the reliability of the diameter matching deviation index.

[0122] As Figure 2 shown, an embodiment of the present invention further provides an intelligent analysis and diagnosis method for the diameter matching of gas metering appliances, including: Step 1, real-time collect pipeline pressure gradient data, flow phase data, and user gas consumption time series fluctuation data through pressure sensors deployed at the three-dimensional spatial coordinate positions of the main nodes, user access points, and pressure regulating stations of the gas pipeline network. Step 2, based on the pipeline pressure gradient data, flow phase data, and user gas consumption time series fluctuation data, select the main pipe intersection point P1, user access point P2, and pressure regulating station output point P3 as three vertices, and form a dynamic triangle according to the three vertices; perform mesh division on the dynamic triangle based on the Delaunay triangulation algorithm, and extract the pressure-flow coupling eigenvalue of the mesh. Step 3, weight and correct the mutation probability feature vector of user gas consumption demands through the pressure-flow coupling eigenvalue to obtain the corrected mutation probability feature vector. Step 4, perform tensor splicing on the corrected mutation probability feature vector and the dynamic feature vector of the equivalent inner diameter of the pipeline, and input it into the fully connected diagnostic network to generate a diameter matching deviation index; when the deviation exceeds the threshold, link the time series prediction network to output a safe redundant diameter replacement plan.

[0123] An embodiment of the present invention further provides a computing device, including: a processor and a memory storing a computer program. When the computer program is run by the processor, it executes the method as described above. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0124] An embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when run on a computer, cause the computer to execute the method described above. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0125] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent analysis and diagnosis system for gas metering appliance caliber matching, characterized in that, Including: A collection module, which is used to collect pipeline pressure gradient data, flow phase data and user gas consumption time series fluctuation data in real time through pressure sensors deployed at the three-dimensional spatial coordinate positions of the main nodes, user access points and pressure regulating stations of the gas pipeline network; An extraction module, which is used to select the main pipeline intersection point P1, user access point P2, and pressure regulating station output point P3 as three dynamic topology reference points based on the pipeline pressure gradient data, flow phase data and user gas consumption time series fluctuation data; construct a dynamic topology unit according to the three points, and perform regional division on the unit based on the Delaunay triangulation algorithm to extract the pressure-flow coupling eigenvalue of the sub-region; A correction module, which is used to perform weighted correction on the user gas consumption demand mutation probability eigenvector through the pressure-flow coupling eigenvalue to generate a corrected mutation probability eigenvector; A processing module, which is used to perform tensor splicing on the corrected mutation probability eigenvector and the pipeline equivalent inner diameter dynamic eigenvector, and input it into a fully connected diagnostic network to generate a diameter matching deviation index; when the deviation exceeds the set threshold, link the time series prediction network to output a safe redundant diameter replacement plan.

2. The intelligent analysis and diagnosis system for gas meter caliber matching according to claim 1, characterized in that Collecting pipeline pressure gradient data, flow phase data and user gas consumption time series fluctuation data in real time through pressure sensors deployed at the three-dimensional spatial coordinate positions of the main nodes, user access points and pressure regulating stations of the gas pipeline network, including: Establishing a pipeline network topology mapping relationship based on the three-dimensional spatial coordinates of each sensor; Synchronously collecting the real-time pressure values along the pipeline direction between the main nodes, calculating the pressure difference correlation space distance between adjacent nodes, and generating pressure gradient data; Monitoring the output of the flow sensor at the user access point, extracting the phase offset and amplitude characteristics of the flow waveform within the period, and generating flow phase data; Recording the instantaneous fluctuation signal of the gas consumption downstream of the pressure regulating station, calculating the fluctuation frequency and the absolute difference between the peak and valley within the unit time, and constituting the user gas consumption time series fluctuation data.

3. The intelligent analysis and diagnosis system for gas meter caliber matching according to claim 2, characterized in that Selecting the main pipeline intersection point P1, user access point P2, and pressure regulating station output point P3 as three dynamic topology reference points based on the pipeline pressure gradient data, flow phase data and user gas consumption time series fluctuation data, including: According to the pipeline network topology mapping relationship, selecting the main pipeline intersection point with the largest pressure change rate in the pressure gradient data as P1, and using its three-dimensional spatial coordinates as the initial reference coordinates; Taking the initial reference coordinates of the main pipeline intersection point P1 as a reference, combining the phase offset and amplitude characteristics in the flow phase data, and calculating the spatial influence weight of the user access point P2; compensating the offset of the original three-dimensional coordinates of the user access point P2 according to the spatial influence weight to obtain the real-time monitoring coordinates of the weighted user access point P2; Based on the real-time monitoring coordinates of the weighted user access point P2, performing time series displacement compensation on the original coordinates of the pressure regulating station output point P3 through the fluctuation frequency and peak-valley difference in the user gas consumption time series fluctuation data to obtain the real-time regulation coordinates of the pressure regulating station output point P3; Spatially synchronize the initial reference coordinates of the main pipeline intersection point P1, the real-time monitoring coordinates of the user access point P2, and the real-time regulation coordinates of the pressure regulating station output point P3, and convert them to a two-dimensional plane through dimensionality reduction mapping; connect the three points P1, P2, and P3 in the two-dimensional plane to form a dynamic topological unit.

4. The intelligent analysis and diagnosis system for caliber matching of gas metering appliances according to claim 3, characterized in that, Construct a dynamic topological unit based on the three points, perform regional division on this unit based on the Delaunay triangulation algorithm, and extract the pressure-flow coupling eigenvalue of the sub-region, including: Execute the Delaunay triangulation algorithm on the dynamic topological unit to generate a minimum angle division structure containing at least three sub-regions; For each divided sub-region, extract the pipeline pressure gradient data associated with the vertex coordinates of this sub-region, and synchronously obtain the amplitude characteristics of the flow phase data within the same spatial domain; Calculate the product arithmetic mean of the pressure gradient data and the flow phase amplitude characteristics, and output the basic spatial coupling factor of this sub-region; Based on the basic spatial coupling factor, locate the gas usage time series fluctuation data of the user access points under the jurisdiction of this sub-region, calculate the statistical variance value of the user fluctuation data within a continuous time window, and perform linear normalization processing on the variance value in the interval [0, 1] to generate a time series correction coefficient; Add the basic spatial coupling factor and the time series correction coefficient to generate an intermediate coupling feature; Multiply the intermediate coupling feature by the spatial weight coefficient of this sub-region in the pipe network topology, and finally output the pressure-flow coupling eigenvalue.

5. The intelligent analysis and diagnosis system for gas meter caliber matching according to claim 4, wherein Execute the Delaunay triangulation algorithm on the dynamic topological unit to generate a minimum angle division structure containing at least three sub-regions, including: Use the real-time two-dimensional plane coordinates of the three points P1, P2, and P3 in the dynamic topological unit as the input point set for triangulation, and construct an initial convex hull boundary; Based on the real-time change amplitude of the user gas usage time series fluctuation data, when the fluctuation amplitude exceeds the set threshold, dynamically insert an auxiliary vertex Q1 at the midpoint of the line connecting P2 and P3; according to the phase offset of the flow phase data, insert an auxiliary vertex Q2 at the midpoint of the line connecting P1 and P2; add Q1 and Q2 to the triangulation point set and update the convex hull to obtain the updated triangulation point set; Traverse the updated triangulation point set, and perform diagonal exchange detection on any quadrilateral formed by four adjacent points; if the minimum interior angle increases after the exchange, update the topological connection relationship; iterate and optimize until all quadrilaterals meet the Delaunay empty circle criterion to generate the final triangulation structure; Define each minimum topological unit in the final triangulation structure as a sub-region to obtain the sub-region vertex coordinate set and the adjacency relationship table.

6. The intelligent analysis and diagnosis system for caliber matching of gas metering appliances according to claim 5, characterized in that, Perform weighted correction on the user gas usage demand mutation probability feature vector through the pressure-flow coupling eigenvalue, and generate a corrected mutation probability feature vector, including: Input the pressure-flow coupling eigenvalues of each sub-region into a matrix converter, sort them according to the spatial weight coefficient of the sub-region in the pipe network topology, and scale the eigenvalues with the pipeline length under the jurisdiction of each sub-region as the normalization factor to generate a diagonal correction coefficient matrix with the same dimension as the mutation probability feature vector; According to the preset user gas usage demand mutation probability feature vector, perform Hadamard product operation on the correction coefficient matrix and the mutation probability feature vector to obtain the weighted feature vector; Perform L2 norm normalization on the weighted feature vector to obtain a corrected mutation probability feature vector.

7. The intelligent analysis and diagnosis system for gas meter caliber matching according to claim 6, characterized in that Perform tensor splicing on the corrected mutation probability feature vector and the pipeline equivalent inner diameter dynamic feature vector, and input it into a fully connected diagnostic network to generate a caliber matching deviation index, including: According to the corrected mutation probability feature vector and the real-time updated pipeline equivalent inner diameter dynamic feature vector, expand it into a three-dimensional tensor along the feature dimension, where the first dimension is the time step, the second dimension is the spatial partition, and the third dimension is the feature channel; Input the three-dimensional tensor into a three-layer fully connected diagnostic network, where: The first layer, the linear transformation layer reduces the input feature dimension to 1 / 2 of the original dimension, and passes through the ReLU activation function; The second layer, the linear transformation layer further reduces it to 1 / 4 of the original dimension, and passes through the Sigmoid activation function; The third layer, the single neuron linear layer outputs the original value of the deviation; Perform dynamic threshold calibration on the original value output by the third layer, and calculate the Z-score standardized offset of the current value relative to the reference distribution curve fitted by the preset historical normal operating condition data.

8. An intelligent analysis and diagnosis method for the caliber matching of gas metering instruments, characterized in that, The method is used to execute the system described in any one of claims 1 to 7, and the method includes: Step 1, through the pressure sensors deployed at the three-dimensional spatial coordinate positions of the main nodes, user access points, and pressure regulating stations of the gas pipeline network, collect pipeline pressure gradient data, flow phase data, and user gas consumption time series fluctuation data in real time; Step 2, based on the pipeline pressure gradient data, flow phase data, and user gas consumption time series fluctuation data, select the main pipe intersection point P1, user access point P2, and pressure regulating station output point P3 as the three vertices, and form a dynamic triangle according to the three vertices; based on the Delaunay triangulation algorithm, perform grid division on the dynamic triangle, and extract the pressure-flow coupling eigenvalue of the grid; Step 3, weight and correct the user gas consumption demand mutation probability feature vector through the pressure-flow coupling eigenvalue to obtain a corrected mutation probability feature vector; Step 4, perform tensor splicing on the corrected mutation probability feature vector and the pipeline equivalent inner diameter dynamic feature vector, and input it into a fully connected diagnostic network to generate a caliber matching deviation index; when the deviation exceeds the threshold, link the time series prediction network to output a safe redundant caliber replacement plan.

9. A computing device, characterized in that, Including: One or more processors; A storage device for storing one or more computer programs, and when the one or more computer programs are executed by the one or more processors, the one or more processors implement the system described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the system described in any one of claims 1 to 7 is implemented.

Images