A valve opening and closing state real-time tracking and abnormal diagnosis method

Abstract

This application provides a method for real-time tracking and anomaly diagnosis of valve opening and closing status, including: obtaining the flow contribution variation of each branch by comparing the impedance growth rate of each branch with increased roughness with a preset flow distribution rating; determining the interference level of each branch based on the flow contribution variation; analyzing the anomaly type of the abnormal valve and the flow of its corresponding branch; evaluating the influence weight between the anomaly type and the flow distribution based on the interference level; determining the precise coordinates of the anomaly location; obtaining the actual flow at the precise coordinates of the anomaly location; determining the fluctuation range between the expected flow and the actual flow; determining the tracking and judgment boundary value of the valve opening and closing status based on the fluctuation range; and processing the real-time roughness change rate and opening reading through the tracking and judgment boundary value of the valve opening and closing status to evaluate whether the valve opening and closing status is normal.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for real-time tracking and anomaly diagnosis of valve opening and closing status. Background Technology

[0002] In the field of industrial pipeline network operation and management, real-time tracking and anomaly diagnosis of valve opening and closing status are crucial, directly impacting the safety and efficiency of system operation. Whether in water treatment, heating, or chemical production, valves, as core components of flow control, play a decisive role in ensuring the stable operation of the entire pipeline network through accurate status assessment. Any misjudgment or anomaly in valve status can lead to flow runaway, even system failure, affecting production safety. However, current valve status monitoring relies too heavily on the opening data of individual valves to infer their flow contribution, neglecting changes in external conditions and resulting in inaccurate judgments. Especially when the resistance characteristics of certain pipes in the network change, even if the opening of a valve remains constant, the actual flow rate may fluctuate significantly, rendering traditional judgment methods unreliable. A deeper technical challenge lies in the core factor of the dynamic changes in pipeline network impedance distribution. Impedance distribution refers to the resistance characteristics of each pipe in the network to flow, which is constantly changing due to factors such as pipe aging and sediment accumulation. When impedance distribution changes, the logic of estimating flow rate based on a fixed valve opening no longer applies. This is because, at the same opening, the actual flow rate controlled by the valve is affected by the redistribution of resistance across the entire pipeline network. This change is not only difficult to predict but also leads to deviations in valve status assessment. For example, in a multi-branch parallel heating network, if a branch pipe becomes roughened due to long-term use, increasing resistance, the flow rate in other branches may increase significantly even if the valve opening remains unchanged, rendering the original flow rate estimation benchmark completely invalid. Therefore, accurately determining the true flow contribution and opening / closing status of each valve in the complex environment of constantly changing pipeline impedance distribution is a key problem that this research urgently needs to solve. Summary of the Invention

[0003] This invention provides a method for real-time tracking and anomaly diagnosis of valve opening and closing status, including:

[0004] The system acquires real-time valve opening values, branch flow distribution, and pipe roughness changes, performs timestamp alignment and correlation processing, and obtains the pipeline impedance distribution matrix and the mapping function between valve opening and flow rate.

[0005] Based on the pipeline impedance distribution matrix, the impedance growth rate of each branch is identified, and the correlation coefficient between the impedance growth rate and the historical roughness data sequence is calculated to identify branches with increased roughness.

[0006] The flow contribution variation is calculated by increasing the impedance growth rate of the branch by increasing the roughness and the preset flow distribution rating, and the interference level of each branch is determined based on the flow contribution variation.

[0007] The expected flow rate at the target valve opening is calculated based on the mapping function between valve opening and flow rate. The flow rate difference is then compared with the actual flow rate data to identify valves with abnormal flow rates.

[0008] Obtain the abnormal valve's abnormality type identifier, its associated branch flow data, and interference level; construct an influence weight matrix; and combine this with the set of valve-related nodes to determine the precise coordinates of the abnormal location.

[0009] Obtain the actual flow rate data and expected flow rate value at the precise coordinates of the abnormal location, calculate the deviation rate, and determine the boundary value for tracking and judging the valve opening and closing status.

[0010] The valve opening and closing status is evaluated by comparing and correcting the real-time roughness change information with the opening reading through the boundary value of the valve opening and closing status tracking judgment.

[0011] Furthermore, the process of acquiring real-time valve opening values, branch flow distribution, and pipe roughness variation information, performing timestamp alignment and correlation processing, and obtaining the mapping function between the pipeline impedance distribution matrix and valve opening and flow rate includes:

[0012] Valve opening values, branch flow monitoring data, changes in pipe roughness coefficient, and node pressure values ​​are acquired by sensors, and the data are aligned according to timestamps to obtain a synchronous pipeline network operation status dataset.

[0013] Based on the synchronous pipeline network operation status dataset, the pressure difference between adjacent nodes is calculated, and the Darcy-Wiesbach formula is used to calculate the impedance coefficient of each pipe segment using the pressure difference, flow rate and roughness coefficient variation to obtain the pipeline network impedance distribution matrix.

[0014] By fitting the pipeline impedance distribution matrix and the real-time valve opening value using the least squares method, a mapping function between valve opening and flow rate is established.

[0015] Furthermore, the step of identifying the impedance growth rate of each branch based on the pipeline impedance distribution matrix, calculating the correlation coefficient between the impedance growth rate and the historical roughness data sequence, and identifying branches with increasing roughness includes:

[0016] Obtain the impedance value sequence of each branch at continuous time points in the pipeline impedance distribution matrix, and perform linear regression fitting with time as the independent variable to obtain the impedance growth rate.

[0017] Extract the historical roughness data sequence of the corresponding branch and calculate the roughness change rate at adjacent time points;

[0018] The Pearson correlation coefficient is calculated by comparing the roughness change rate with the impedance growth rate to determine the branch with increasing roughness.

[0019] Furthermore, the step of calculating the flow contribution variation by increasing the impedance growth rate of the branch through roughness and the preset flow distribution rating, and determining the interference level of each branch based on the flow contribution variation, includes:

[0020] The flow deviation coefficient is calculated by obtaining the rate of increase in impedance of the branch with increased roughness and the preset flow distribution rating. The flow contribution variation is determined based on the flow deviation coefficient and the current actual flow.

[0021] The influence weight value is calculated by using the change in flow contribution and the total flow of the pipeline network, and the low, medium and high interference levels are determined based on the influence weight value.

[0022] Furthermore, it also includes obtaining the current impedance growth rate of the roughness-increasing branch and calculating the flow attenuation ratio based on the preset flow distribution rating, collecting data on the thickness of the roughness layer on the inner wall of the pipe and the degree of pressure loss, and determining the transport difference of the roughness-increasing branch, specifically including:

[0023] The current impedance growth rate of the branch with increased roughness is obtained and the preset flow distribution rating is used to calculate the flow attenuation ratio. Data on the roughness layer thickness and pressure loss are collected.

[0024] The influence coefficient is calculated using the Hayzen-Williams formula based on the flow attenuation ratio, roughness layer thickness data and pressure loss degree. The influence coefficient is then multiplied by the original flow capacity of the pipeline to obtain the actual flow capacity.

[0025] The downstream flow shortage value is obtained by comparing the actual flow capacity with the rated flow demand of the downstream water supply area. The transport difference of the roughness-increased branch is determined based on the downstream flow shortage value and the flow attenuation ratio.

[0026] Furthermore, the step of calculating the expected flow rate at the target valve opening based on the mapping function between the valve opening and the flow rate, comparing it with the actual flow rate data to obtain the flow rate difference value, and identifying valves with abnormal flow rates includes:

[0027] The expected flow rate is obtained by linearly interpolating the current opening value of the target valve according to the mapping function between valve opening and flow rate, and then compared with the actual flow rate data read by the flow sensor to obtain the flow rate difference value.

[0028] The abnormal valves can be identified as either clogged or leaking based on the positive or negative value of the flow drop.

[0029] Furthermore, the process of obtaining the abnormal valve's abnormality type identifier, its associated branch flow data, and interference level, constructing an influence weight matrix, and determining the precise coordinates of the abnormal location by combining the valve's associated node set includes:

[0030] Obtain the abnormal valve's abnormal type identifier and the real-time flow data of its branch; extract the upstream and downstream node numbers and their spatial coordinates based on the pipeline network topology; and construct a set of valve-related nodes.

[0031] A judgment matrix is ​​constructed by using anomaly type identifiers, branch flow data and interference levels, and the influence weight matrix is ​​obtained by calculating the weights of each factor.

[0032] Based on the influence weight matrix, the decreasing influence of anomalies on adjacent nodes is analyzed, and nodes with weights greater than a preset threshold are identified to obtain the anomaly influence distribution vector.

[0033] The anomaly center location is calculated using the weighted average method based on the spatial coordinates of the three nodes with the largest weights in the anomaly influence distribution vector, thus determining the precise coordinates of the anomaly location.

[0034] Furthermore, the step of obtaining the actual flow rate data and expected flow rate value at the precise coordinates of the abnormal location, calculating the deviation rate, and determining the boundary value for valve opening and closing status tracking includes:

[0035] Obtain the actual flow data and expected flow value of the pipeline node corresponding to the precise coordinates of the abnormal location, and calculate the deviation rate;

[0036] Based on the statistical distribution of historical operating data, the deviation rate range corresponding to the normal fluctuation range is determined, and the boundary value for tracking and judging the valve opening and closing status is obtained.

[0037] Furthermore, the step of comparing and correcting the real-time roughness change information with the opening degree reading by tracking and judging the boundary value of the valve opening and closing state to evaluate the valve opening and closing state includes:

[0038] Obtain real-time roughness change information, calculate the roughness influence factor, and multiply it by the opening value to obtain the corrected opening value;

[0039] The corrected opening value is compared with the valve opening and closing status tracking and judgment boundary value to determine whether the valve opening and closing status is normal or abnormal.

[0040] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0041] This invention discloses a method for real-time tracking and anomaly diagnosis of valve opening and closing status. Addressing the comprehensive problems of impedance growth, abnormal flow distribution, and valve malfunctions caused by increased pipe roughness in pipeline systems, this method acquires real-time information on valve opening, branch flow, and pipe roughness changes. It constructs an initial impedance distribution-flow mapping relationship, identifies branches with excessive impedance growth rates, and assesses the interference level and flow capacity attenuation by combining historical roughness data and flow contribution variations. This allows for accurate calculation of downstream water supply shortages and delivery differences. Furthermore, by analyzing valve flow drop and anomaly types, and combining branch interference levels, this invention determines the coordinates of anomaly locations and tracks valve opening and closing status boundary values, ensuring the accuracy of status monitoring. Ultimately, this invention achieves comprehensive monitoring and dynamic control of pipeline impedance, flow distribution, and valve status, effectively improving pipeline network operational stability and water supply reliability. Attached Figure Description

[0042] Figure 1 This is a flowchart of a valve opening and closing status real-time tracking and anomaly diagnosis method according to the present invention.

[0043] Figure 2 This is a schematic diagram of a valve opening and closing status real-time tracking and anomaly diagnosis method according to the present invention.

[0044] Figure 3 This is another schematic diagram of a valve opening and closing status real-time tracking and anomaly diagnosis method according to the present invention. Detailed Implementation

[0045] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0046] like Figures 1-3 This embodiment of a method for real-time tracking and anomaly diagnosis of valve opening and closing status may specifically include:

[0047] S101. Obtain the real-time opening degree of each valve and the branch flow distribution, and at the same time obtain the pipe roughness change rate. Integrate and correlate them to obtain the initial pipeline impedance distribution and the mapping relationship between valve opening degree and flow.

[0048] Real-time valve opening values ​​and corresponding branch flow monitoring data are acquired through sensors. Simultaneously, changes in the pipe wall roughness coefficient and pressure values ​​at each node are collected. Data is aligned based on the timestamps of the opening values ​​and flow data to obtain a synchronized pipeline network operation status dataset. Based on the branch flow and node pressure values ​​in the pipeline network operation status dataset, the pressure difference between adjacent nodes is calculated. Using the Darcy-Wiesbach formula, with the pressure difference, flow rate, and roughness coefficient change as input parameters, the impedance coefficient of each pipe segment is calculated. If the roughness coefficient exceeds a preset threshold, the impedance coefficient is corrected to obtain the pipeline network impedance distribution matrix. Using the pipeline network impedance distribution matrix and the real-time valve opening values, the least squares method is used to fit the functional relationship between the valve opening and the flow rate of its controlled branch, establishing a mapping function between valve opening and flow rate, and determining the flow contribution value of each valve at different opening degrees.

[0049] Specifically, in one implementation, the pipeline monitoring system collects valve rotation angle data in real time using opening sensors deployed at each valve location, converting the angle values ​​into percentage opening values. Simultaneously, ultrasonic flow meters are installed in each branch pipeline, recording instantaneous flow rates every five seconds. The pipeline inner wall roughness coefficient is measured indirectly using the differential pressure method, i.e., pressure sensors are installed at both ends of the pipe section, and the roughness is inferred from the relationship between pressure loss and flow velocity. The timestamp alignment employs an interpolation algorithm to unify data from different sampling frequencies onto the same time reference.

[0050] Specifically, the application of the Darcy-Wiesbach formula in pipeline impedance calculation is as follows: First, obtain the pressure values ​​at both ends of the pipe segment and calculate the pressure difference as the driving head; then, divide the pressure difference by the measured flow rate of the pipe segment to obtain the initial impedance coefficient. This impedance coefficient reflects the resistance characteristics of the pipeline to the fluid and is closely related to the pipe length, diameter, and inner wall roughness. When the inner wall of the pipe develops deposits or corrosion due to long-term use, the roughness coefficient will increase, requiring the introduction of a correction factor. The correction process involves calculating the ratio of the measured roughness coefficient to the standard roughness coefficient. If the ratio exceeds a preset threshold of 1.2, the initial impedance coefficient is multiplied by the square root of this ratio to obtain the corrected impedance coefficient. By performing the above calculations on all pipe segments in the pipeline network, an impedance matrix reflecting the resistance distribution of the entire pipeline network is formed.

[0051] In one possible implementation, the process of establishing a mapping function between valve opening and flow rate using the least squares method includes: collecting the opening value of the same valve at different times and the flow rate value of its corresponding branch to form a data point set; selecting a quadratic polynomial as the fitting function form, and determining the polynomial coefficients by minimizing the sum of squared errors between the measured flow rate and the fitted flow rate; the mapping function can predict the corresponding flow contribution value based on any opening value.

[0052] It should be noted that the flow contribution value represents the degree of influence of a single valve on the flow of its controlled branch. This value is dynamically adjusted as the impedance of other parts of the pipeline changes, thereby achieving accurate tracking of the valve status.

[0053] S102. Identify the impedance growth rate of each branch based on the initial pipeline impedance distribution, and assess whether the impedance growth rate exceeds the set impedance growth limit. If it does, compare the impedance growth rate with the historical roughness data to identify branches with increased roughness.

[0054] The impedance value sequence of each branch in the pipeline impedance distribution matrix at continuous time points is obtained. A linear regression is performed with time as the independent variable and impedance value as the dependent variable. The slope of the fitted line is calculated as the impedance growth rate of each branch. The impedance growth rate is compared with a preset impedance growth limit. If the limit is exceeded, the historical roughness data sequence of the corresponding branch is extracted from the pipeline historical monitoring database. The roughness change rate is calculated by dividing the difference in roughness between adjacent time points by the time interval. A Pearson correlation coefficient is calculated between the roughness change rate and the impedance growth rate. When the correlation coefficient is greater than a preset threshold, it is determined that the branch has a roughness increase problem.

[0055] Specifically, in one implementation, the pipeline monitoring system records the impedance values ​​of each branch hourly, forming a time-series dataset. For each branch, the impedance value sequence for the most recent seven days is extracted to construct a time-impedance coordinate system, where time is in hours. A least squares method is used for linear regression fitting, and the slope of the fitted line is calculated as the impedance growth rate, which reflects the changing trend of the pipeline resistance characteristics.

[0056] It should be noted that the preset impedance growth limit is determined based on the pipeline network's design life and material characteristics. For cast iron pipelines, the limit is set at no more than 2% of the initial value per month; for cement pipelines, the limit is appropriately relaxed to 3%. When the impedance growth rate of a branch exceeds the limit, the system automatically retrieves the roughness monitoring records of that branch from the historical database for the past three months.

[0057] Specifically, the calculation of the Pearson correlation coefficient involves a statistical correlation analysis of the roughness change rate sequence and the impedance growth rate sequence. First, the two sequences are standardized to eliminate the influence of dimensions. Then, the covariance of the two sequences is calculated and divided by the product of their respective standard deviations to obtain the correlation coefficient. This coefficient ranges from -1 to +1. When the coefficient is greater than 0.7, it indicates that increased roughness is the main cause of impedance growth. In heating pipe networks, scale deposition on the inner wall of pipes can simultaneously cause an increase in both roughness and impedance, showing a strong positive correlation. In chemical pipe networks, the accumulation of corrosion products produces a similar effect. By judging the correlation, it is possible to distinguish whether the impedance growth is caused by changes in roughness or by other factors such as pipe diameter deformation or valve malfunction.

[0058] Preferably, for cases where the correlation coefficient is between 0.5 and 0.7, the system will further analyze the time distribution characteristics of the roughness change. If the change shows a continuous growth trend rather than random fluctuation, it will still be determined that the branch has a roughness increase problem.

[0059] In one possible implementation, the preset threshold is adjusted according to different pipe network types, with the threshold set to 0.75 for domestic water supply pipe networks and 0.65 for industrial circulating water pipe networks, to adapt to the monitoring needs of different application scenarios.

[0060] S103. By comparing the impedance growth rate of each branch with increased roughness with the preset flow distribution rating, the flow contribution variation of each branch is obtained, and the interference level of each branch is determined based on the flow contribution variation.

[0061] The impedance growth rate of each branch with increased roughness is obtained, and the ratio of this rate to the preset flow distribution rating of each branch is used to calculate the flow deviation coefficient. The product of this flow deviation coefficient and the current actual flow rate of the branch is used to determine the flow contribution variation of each branch. The ratio of this flow contribution variation to the total flow rate of the pipeline network is used to calculate the influence weight of each branch on the pipeline network operation, obtaining the influence weight value for each branch. This influence weight value is compared with a preset interference level threshold. If the influence weight is less than the lower threshold, it is determined to be a low interference level; if it is between the lower and upper thresholds, it is determined to be a medium interference level; and if it exceeds the upper threshold, it is determined to be a high interference level.

[0062] Specifically, in one implementation, the flow deviation coefficient characterizes the degree to which the actual operating state of a branch deviates from the design conditions by measuring the ratio of the impedance growth rate to a preset flow distribution rating. The preset flow distribution rating is determined during the pipeline network design phase based on the water demand of each area and the pipeline's carrying capacity, serving as a benchmark for ideal operating conditions. When impedance growth causes a change in flow distribution, the deviation coefficient reflects the magnitude of this change.

[0063] Specifically, the calculation of the flow contribution variation involves a comprehensive consideration of the deviation coefficient and the actual flow rate. The deviation coefficient reflects the degree of deviation of the impedance growth rate from the preset flow allocation rating. The flow contribution variation is calculated using the formula ΔQ = (deviation coefficient - 1) × Q. actual Calculate, where Q actual This represents the current actual flow rate of the branch. If the deviation coefficient of a branch is 1.2 and the actual flow rate is 50 liters per second, then the change in flow contribution is (1.2-1)×50=10 liters per second, indicating the change in flow rate from the rated value due to increased roughness. A deviation coefficient greater than 1 indicates that the flow distribution deviates from the design conditions due to increased impedance. The larger the absolute value of the change, the more significant the impact on the flow distribution of the pipeline network.

[0064] It should be noted that the influence weight is used to quantify the impact of each branch on the overall operation by measuring the proportion of the change in flow contribution to the total flow of the pipeline network. In heating pipeline networks, the influence weight of the main pipeline is usually larger, while the influence weight of the branch pipeline is relatively smaller.

[0065] Preferably, the interference level classification adopts a three-level threshold mechanism, with a lower threshold set at 0.05 and an upper threshold set at 0.15. When the impact weight is less than 0.05, it indicates that the roughness change of the branch has a minor impact on the pipeline network operation, and is classified as a low interference level, requiring only routine monitoring; between 0.05 and 0.15 is a medium interference level, requiring increased monitoring frequency and the development of a maintenance plan; and above 0.15 is a high interference level, indicating that the branch has seriously affected the normal operation of the pipeline network, requiring immediate cleaning or replacement measures. This classification mechanism enables maintenance personnel to rationally allocate maintenance resources according to different interference levels, prioritizing high-interference branches and achieving precise management.

[0066] In one possible implementation, for a parallel pipeline network structure, the interference levels of each branch will also affect each other. When a branch is classified as having a high interference level, the flow of the branch connected in parallel with it will increase accordingly, and its interference level assessment needs to be dynamically adjusted.

[0067] The current impedance growth rate of the branch with increased roughness is obtained and compared with the preset flow distribution rating of the branch with increased roughness to obtain the flow attenuation ratio. At the same time, the thickness of the rough layer on the inner wall of the branch with increased roughness and the pressure loss when the fluid passes through are collected. The actual flow capacity attenuation of the branch with increased roughness after impedance growth is analyzed, the degree of flow shortage in the downstream water supply area caused by the flow capacity attenuation is assessed, and the transport difference of the branch with increased roughness is determined.

[0068] The current impedance growth rate of the branch with increased roughness is obtained and compared with the preset flow distribution rating of that branch. The flow attenuation ratio is calculated by the ratio of the two. Simultaneously, data on the thickness of the rough layer on the inner wall of the branch pipe and the pressure loss during fluid flow are collected. Based on the flow attenuation ratio, rough layer thickness, and pressure loss, the Hayzen-Williams formula is used to calculate the influence coefficient of pipe roughness on flow. This influence coefficient is multiplied by the original flow capacity of the pipe to obtain the actual flow capacity after impedance increase. This actual flow capacity is compared with the rated demand flow determined during the pipeline network design of the downstream water supply area. If the actual flow capacity is lower than the rated demand flow, the difference is calculated to obtain the downstream flow shortage value. Based on the downstream flow shortage value and the flow attenuation ratio, the reduction in transport capacity caused by increased roughness in this branch is calculated and determined as the transport difference of the branch with increased roughness.

[0069] Specifically, in one implementation, the flow attenuation ratio is calculated based on the ratio of the impedance growth rate to a preset flow allocation rating. This preset flow allocation rating is determined during the initial design phase of the pipeline network based on the water demand, pipe specifications, and water pressure distribution of each area, representing the flow share the pipeline should ideally handle. When the inner wall of the pipe becomes rougher due to long-term operation, the impedance growth rate deviates from the normal range; the ratio of these two values ​​directly reflects the severity of the flow attenuation.

[0070] It should be noted that the thickness of the rough layer on the inner wall of the pipe is measured non-contactly using an ultrasonic thickness gauge. The ultrasonic waves emitted by the probe are reflected at the interface between the rough layer and the pipe wall substrate, and the thickness of the rough layer is calculated based on the echo time difference. The pressure loss is obtained by high-precision pressure sensors installed at both ends of the pipe section, which record the pressure change of the fluid before and after passing through the pipe section in real time.

[0071] Specifically, the application of the Hayzen-Williams formula in pipeline hydraulic calculations involves the comprehensive consideration of several key parameters. This formula expresses the functional relationship between flow rate and pipe roughness coefficient, pipe diameter, and hydraulic gradient, where the roughness coefficient is the core parameter reflecting the condition of the pipe's inner wall. For new cast iron pipes, the roughness coefficient is typically 130; after ten years of use, it drops to around 100; for severely corroded or scaled pipes, the roughness coefficient may drop below 60. The current roughness coefficient value can be estimated using the measured thickness of the roughened layer. Substituting the flow rate attenuation ratio, the measured roughness coefficient, and the degree of pressure loss into the Hayzen-Williams formula, the influence coefficient of pipe roughness on flow rate is calculated. This influence coefficient reflects the degree to which increased roughness weakens the pipe's water-carrying capacity, typically ranging from 0.6 to 1.0. Multiplying the influence coefficient by the original flow capacity of the pipe yields the actual flow capacity under the current impedance growth state. This actual flow capacity value directly determines the actual amount of water that the branch can supply downstream, serving as the basis for subsequent flow shortage assessments.

[0072] Preferably, the effect of temperature on water viscosity is also considered when calculating the actual flow capacity. For every 10-degree Celsius increase in water temperature, the viscosity decreases by about 20%, and the corresponding flow resistance also decreases. Therefore, in heating pipe networks, the actual flow capacity of high-temperature water will be slightly higher than that of room-temperature water.

[0073] In one possible implementation, the rated flow demand of the downstream water supply area is determined comprehensively based on the water-using population, industrial water consumption, and fire-fighting reserve requirements. Domestic water consumption is calculated at 200 liters per person per day, industrial water consumption is determined according to production process requirements, and fire-fighting reserves are set according to building area and fire risk level.

[0074] For example, in the circulating water network of a chemical industrial park, the rated demand flow of a main branch is 500 cubic meters per hour. Calculations show that the actual flow capacity of this branch is only 420 cubic meters per hour, resulting in a flow shortage of 80 cubic meters per hour. This means that the downstream cooling tower cannot receive enough circulating water, potentially leading to a decrease in heat exchange efficiency. Furthermore, the determination of this flow difference comprehensively considers both the flow shortage value and the flow attenuation ratio. The flow shortage value directly reveals the downstream water supply gap, while the flow attenuation ratio allows us to trace how much of this gap is directly caused by increased roughness.

[0075] For example, if the flow rate reduction ratio is 0.8, it means that the actual flow rate is only 80% of the rated value, and the conveying capacity has decreased by 20%. Multiplying this reduction ratio by the flow rate shortage value yields the conveying difference caused by increased roughness. This conveying difference is an important basis for developing pipeline cleaning or replacement plans.

[0076] Understandably, accurately determining the transport difference helps maintenance personnel assess the urgency and cost-effectiveness of pipeline maintenance. When the transport difference exceeds 30% of the pipeline's transport capacity, it is generally considered that immediate cleaning or replacement of the pipeline's inner wall is necessary.

[0077] S104. Based on the mapping relationship between valve opening and flow rate, analyze the expected flow rate under the target valve opening, analyze the difference between the actual flow rate and the expected flow rate under the target valve opening, obtain the flow rate drop value, evaluate whether the flow rate drop value exceeds the set limit, and identify abnormal valves with abnormal flow rates.

[0078] Based on the mapping relationship between valve opening and flow rate, the current opening value of the target valve is obtained. Linear interpolation is performed on adjacent data points in the mapping relationship to calculate the expected flow rate corresponding to this opening. Simultaneously, the actual flow rate data of the target valve control branch is read from the flow sensor. The difference between the expected flow rate and the actual flow rate data is calculated to obtain the flow rate difference value. The absolute value of the flow rate difference value is compared with a preset limit. If it exceeds the preset limit, the valve is determined to have an abnormal flow rate. For the flow rate difference value, its positive or negative sign determines the type of abnormality. A positive flow rate difference value indicates that the actual flow rate is less than the expected flow rate, identifying it as a valve blockage-type abnormality. A negative flow rate difference value indicates that the actual flow rate is greater than the expected flow rate, identifying it as a valve leakage-type abnormality.

[0079] Specifically, in one implementation, the mapping relationship between valve opening and flow rate is established using historical operating data, forming a discrete set of data points for valve opening and flow rate. When it is necessary to calculate the expected flow rate for a specific opening, a linear interpolation method is used to estimate between adjacent data points.

[0080] Specifically, the linear interpolation process involves finding adjacent data points of the target valve opening value in the mapping dataset. Assuming the target valve opening is 65%, and the mapping dataset stores discrete points corresponding to a flow rate of 40 liters per second for a 60% opening and 45 liters per second for a 70% opening, then the expected flow rate corresponding to a 65% opening is calculated to be 42.5 liters per second through linear interpolation. Actual flow data is collected in real-time by an electromagnetic flowmeter installed downstream of the valve, which updates its measurement value every second. The flow difference is obtained by subtracting the actual flow rate from the expected flow rate; this difference directly reflects the deviation between the valve's actual operating state and its ideal state. A positive flow difference indicates that the actual flow rate is less than the expected flow rate, suggesting potential blockage in the valve-controlled pipeline, such as scale buildup on the pipeline wall, valve core jamming, or filter clogging. A negative flow difference indicates that the actual flow rate is greater than the expected flow rate, suggesting potential problems such as poor valve sealing, valve body damage, or leakage in the bypass pipeline.

[0081] It should be noted that the preset limit is determined based on the normal operating fluctuation range of the pipeline network, and is usually set at 10% of the expected flow rate. Exceeding this limit indicates that the valve's operating state has deviated from the normal range.

[0082] Preferably, the preset limit can be adjusted for different types of valves. The limit for ball valves is set to 8%, the limit for butterfly valves is set to 12%, and the limit for gate valves is set to 15%.

[0083] In one possible implementation, after anomaly type identification, the system records the location number of the abnormal valve, the anomaly type, the flow drop value, and the occurrence time, forming an anomaly file for maintenance personnel to refer to.

[0084] For example, when the regulating valve of a water supply network is opened to 50%, the expected flow rate is 30 liters per second, but the actual measured flow rate is only 22 liters per second. The flow rate difference is 8 liters, which exceeds the preset limit of 3 liters. This is judged as a blockage-type abnormality, indicating that pipeline cleaning and maintenance are required.

[0085] S105. Analyze the abnormality type of the abnormal valve and the flow rate of its branch, evaluate the influence weight between the abnormality type and the flow distribution in combination with the interference level, and determine the precise coordinates of the abnormal location.

[0086] Obtain the anomaly type identifier and real-time flow data of the branch to which the abnormal valve belongs. Extract the upstream and downstream node numbers and their spatial coordinates of the valve based on the pipeline topology, and construct a set of valve-associated nodes. Using the anomaly type identifier and branch flow data, combined with a preset interference level, construct a judgment matrix using the severity of the anomaly type, the degree of flow deviation, and the interference level as judgment factors, and calculate the weight of each factor to obtain an influence weight matrix. Based on the influence weight matrix and the set of valve-associated nodes, analyze the decreasing influence of the anomaly from the valve to adjacent nodes, identify nodes with weights greater than a preset threshold within the influence range, and obtain the anomaly influence distribution vector. Using the spatial coordinates of the three nodes with the largest weights in the anomaly influence distribution vector, calculate the anomaly center location using a weighted average method to determine the precise coordinates of the anomaly location.

[0087] Specifically, in one implementation, the pipeline topology is stored in the form of an adjacency matrix, with each node recording its unique number, spatial coordinates, and connection relationships. When an abnormal valve is detected, the system extracts the valve's location information from the topology database, including its directly connected upstream water supply nodes and downstream water user nodes. S104 identifies abnormal valves through branch flow data and valve status monitoring, including blockage, leakage, and jamming anomalies. Anomaly type identification uses numerical coding, where 1 represents a blockage anomaly, 2 represents a leakage anomaly, and 3 represents a valve jamming anomaly. The branch flow data is collected in real time by an electromagnetic flowmeter installed on the pipeline, with a sampling frequency of once per second.

[0088] It should be noted that the valve-related node set includes not only directly connected nodes, but also second-level adjacent nodes, that is, other nodes connected to the directly connected nodes. The spatial coordinates of each node are represented in a geodetic coordinate system, including three dimensions: longitude, latitude, and elevation.

[0089] Specifically, the construction of the judgment matrix involves pairwise comparisons of three key judgment factors. The first factor is the severity of the anomaly type, determined by the anomaly type identifier: blockage anomalies are assigned a value of 9, leakage anomalies are assigned a value of 7, and stagnation anomalies are assigned a value of 5. The second factor is the degree of flow deviation, determined by the percentage deviation between the actual flow and the rated flow: deviation exceeding 30% is assigned a value of 9, deviation between 20% and 30% is assigned a value of 7, deviation between 10% and 20% is assigned a value of 5, and deviation less than 10% is assigned a value of 3. The third factor is the interference level, directly using the high, medium, and low interference levels determined in step S103, assigned values ​​of 9, 5, and 3 respectively. A 3×3 judgment matrix A is constructed, with matrix element a. ij This represents the ratio of the importance of the i-th factor to the j-th factor. The weights are calculated using the eigenvalue method of the Analytic Hierarchy Process (AHP): First, solve the equation (A-λ). max ·I)·W=0, thus obtaining the largest eigenvalue λ max The corresponding eigenvector W; then the eigenvector is normalized so that Σw i =1, obtaining the weight values ​​w=[w1,w2,w3] for each factor. These three weight values ​​constitute the influence weight matrix, reflecting the comprehensive impact of anomaly type, flow deviation, and disturbance level on pipeline operation. Consistency testing is performed by calculating the consistency ratio CR=(λ). maxThe matrix is ​​calculated as (n) / (n-1) / RI, where n=3 is the matrix order and RI is the average random consistency index (RI=0.58 when n=3). When CR<0.1, the matrix is ​​considered to have satisfactory consistency. The weights of the three factors are used to form an influence weight matrix, which reflects the comprehensive impact of anomalies on pipeline operation. This multi-factor comprehensive evaluation method can comprehensively consider the different dimensions of anomalies, avoiding the one-sidedness of single-indicator judgment.

[0090] Preferably, when calculating the consistency of the judgment matrix, if the consistency ratio is greater than 0.1, the comparison values ​​in the judgment matrix need to be adjusted until the consistency requirements are met.

[0091] In one possible implementation, the distribution vector of anomaly impacts is obtained based on the law of diminishing impact. Starting from the node where the anomaly valve is located, the degree of impact decreases inversely proportional to the square of the distance. This is calculated using the following formula: Node Impact Weight W node =W base ×(k / d 2 ), where W base The basic influence weight is calculated using the judgment matrix, where k is a constant of 0.8, and d is the shortest path distance from the node to the anomalous valve (in terms of pipe segments). For directly connected nodes (d=1), the node influence weight is 0.8 times the basic weight; for second-order adjacent nodes (d=2), the node influence weight is 0.2 times the basic weight (0.8 / 4=0.2). Considering the actual complexity of the pipeline network, for nodes with d=2, the system adjusts the influence weight to 0.3 times the basic weight to improve the model's sensitivity to indirect influences. For example, a regulating valve in a chemical pipeline network experiences a blockage. The upstream node connected to this valve is numbered N101, and the downstream nodes are numbered N102 and N103. The basic influence weight W is calculated using the judgment matrix. base =0.75. Therefore, the influence weight of the directly connected node N101 is 0.75 × 0.8 = 0.6, and the influence weights of N102 and N103, as direct downstream nodes, are also 0.6. The influence weight of the second-level adjacent node N201 connected to N101 is 0.75 × 0.3 = 0.225. Furthermore, the abnormal influence distribution vector contains the numbers of all affected nodes and their corresponding influence weight values. By setting a weight threshold of 0.2, nodes with weights greater than this threshold are selected as key influencing nodes.

[0092] For example, the weighted average method for calculating the anomaly center location is as follows: Select the three nodes with the largest influence weights, multiply the spatial coordinates of each node by its normalized weight, and then sum the three weighted coordinates to obtain the precise coordinates of the anomaly location. These coordinates represent the geometric center of the anomaly's influence, facilitating rapid location and handling by maintenance personnel.

[0093] It is understandable that when an abnormal valve is located at the end of the pipeline network, the influence weight of only one or two nodes may exceed the threshold. In this case, the weighted average coordinates of these nodes are used as the abnormal location.

[0094] In one embodiment, a branch valve of a heating pipeline experienced a leakage anomaly. The coordinates of the anomaly location were calculated using the above method to be 116.3874 degrees east longitude and 39.9042 degrees north latitude. This location was exactly near the intersection of two pipelines, with a deviation of less than 5 meters from the actual leak point, thus achieving precise location of the anomaly.

[0095] S106. Obtain the actual flow rate at the precise coordinates of the abnormal location, determine the fluctuation range between the expected flow rate and the actual flow rate, and determine the tracking and judgment boundary value of the valve opening and closing status based on the fluctuation range.

[0096] The precise coordinates of the pipeline node corresponding to the abnormal location are obtained. The actual flow data of that node is read from the flow monitoring device, and the expected flow value for that location is extracted from the pipeline design parameters. The difference between the actual and expected flow is calculated, and the difference is divided by the expected flow to obtain the deviation rate. Based on the statistical distribution of historical operating data, the normal fluctuation range is determined as the flow value within a preset percentage range of the deviation rate. Using the normal fluctuation range in conjunction with the deviation rate, when the deviation rate exceeds the upper percentage limit, it is determined that the valve is in an abnormal closed state; when the deviation rate is below the lower percentage limit, it is determined that the valve is in an abnormal open state, thus obtaining the tracking and judgment boundary values ​​for the valve opening and closing states.

[0097] Specifically, in one implementation, the precise coordinates of the abnormal location have been determined using the aforementioned weighted average method, and the corresponding pipeline node is equipped with an electromagnetic flowmeter and a pressure sensor. The flow monitoring equipment collects the instantaneous flow rate value once per second, and the average of ten consecutive sampled values ​​is taken as the actual flow rate data for that node. The expected flow rate value is predetermined based on the pipeline hydraulic calculation model, taking into account factors such as pipe diameter, design pressure, and water demand.

[0098] It should be noted that the deviation rate reflects the degree to which the actual operating condition deviates from the design condition. The formula for calculating the deviation rate is R = (Qactual - Qexpected) / Qexpected, where R is the deviation rate, Qactual is the actual flow rate, and Qexpected is the expected flow rate. A positive deviation rate indicates that the actual flow rate is greater than the expected flow rate; a negative deviation rate indicates that the actual flow rate is less than the expected flow rate.

[0099] Specifically, the determination of the normal fluctuation range is based on a statistical analysis of historical operating data. Flow data from the past three months of normal operation at the node is collected, and after removing outliers, the distribution characteristics of the flow deviation rate are calculated. A normal distribution is used to fit the deviation rate data, and the mean and standard deviation are calculated. The mean plus or minus twice the standard deviation is used as the upper and lower limits of the normal fluctuation range, which covers 95% of normal operating conditions. In water supply networks, due to daily and seasonal variations in water demand, the normal deviation rate typically fluctuates between -15% and +15%. When the network operation is stable, the deviation rate concentrates in the range of -5% to +5%. The normal fluctuation range determined through statistical analysis provides a scientific basis for subsequent anomaly judgments, avoiding misjudgments caused by normal fluctuations.

[0100] Preferably, a time-segmented statistical method is used for flow data in different time periods. The day is divided into peak water usage periods, off-peak periods, and low-water periods, and the normal fluctuation range of each period is statistically analyzed.

[0101] In one possible implementation, the setting of the tracking and judgment boundary values ​​takes into account the mechanical characteristics of the valve. When the deviation rate exceeds 20% of the upper limit of the normal floating range, it is determined that the valve is abnormally open; when the deviation rate is lower than 20% of the lower limit of the normal floating range, it is determined that the valve is abnormally closed.

[0102] For example, the normal deviation rate range for a certain chemical pipeline node is from -10% to +10%. When the measured deviation rate reaches +12%, exceeding the upper limit, the system determines that the valve at that location may be blocked or not properly closed. Furthermore, the tracking and judgment boundary value is dynamically adjusted according to the pipeline's operating conditions, appropriately widening the boundary range during maintenance and tightening it during critical production periods to achieve precise monitoring of valve status.

[0103] S107. By tracking and judging the boundary value of the valve opening and closing status, the roughness change rate and opening reading collected in real time are processed to evaluate whether the valve opening and closing status is normal.

[0104] By tracking and judging the boundary values ​​of the valve's opening and closing status, the real-time roughness change rate and valve opening reading are obtained. The ratio of the roughness change rate to a preset change threshold is used as the roughness influence factor. The roughness influence factor is multiplied by the opening reading to obtain a corrected opening value. This corrected opening value is compared with the tracking and judging boundary values ​​to determine if it is within the normal range. If the corrected opening value is within the boundary value range, the valve's opening and closing status is assessed as normal; if it exceeds the boundary value range, the valve's opening and closing status is assessed as abnormal.

[0105] Specifically, in one implementation, boundary values ​​are tracked and used as a benchmark for valve condition assessment. These values ​​are pre-determined based on historical valve failure data and engineering experiments and stored in the monitoring system. Roughness variation information is obtained through periodic inspection of the pipe inner wall, using an ultrasonic thickness gauge to measure the roughness layer thickness weekly and recording the thickness value for each measurement. Valve opening readings are acquired in real-time using a rotary encoder mounted on the valve stem, achieving an accuracy of 0.1%.

[0106] Specifically, the calculation process of the roughness influence factor involves a comparative analysis of the roughness change rate and a preset change threshold. First, the difference in roughness layer thickness between two adjacent measurements is calculated and divided by the measurement time interval (7 days) to obtain the weekly roughness change rate (unit: mm / week). The weekly change rate is multiplied by 4 to convert it to a monthly roughness change rate (unit: mm / month). The preset change threshold is determined based on the pipe material and service life: the threshold for new pipes is set at no more than 0.1 mm per month, and for pipes used for more than five years, the threshold is appropriately relaxed to 0.2 mm. The measured monthly roughness change rate is divided by the preset change threshold to obtain the dimensionless roughness influence factor. The calculation formula is: Roughness influence factor = Monthly roughness change rate / Preset change threshold. When the influence factor is greater than 1, it indicates that the roughness growth rate exceeds the normal level; when it is less than 1, it indicates that the roughness growth is within an acceptable range. This influence factor directly reflects the degree of impact of roughness changes on the valve's flow control capability. In heating pipe networks, due to the presence of minerals in the water, scale easily forms on the inner walls of the pipes, and the roughness influence factor typically varies between 0.8 and 1.5.

[0107] It should be noted that the opening value is corrected using a multiplicative correction method. The original opening reading is multiplied by the roughness influence factor to obtain the corrected opening value that takes into account the roughness effect. This correction method is based on fluid mechanics principles, where increased roughness reduces the effective flow area of ​​the valve.

[0108] Preferably, when the roughness influence factor exceeds 1.3, the system will issue a warning signal, indicating that pipeline cleaning and maintenance are required.

[0109] In one possible implementation, the comparison between the corrected opening value and the tracking judgment boundary value uses an interval judgment method. The boundary value includes two thresholds: an upper limit and a lower limit. If the corrected opening value falls within this interval, it is determined to be in a normal state.

[0110] For example, if the initial opening reading of a valve is 60% and the roughness influence factor is 1.2, then the corrected opening value is 72%. If the tracking and judgment boundary value range is between 50% and 80%, then the valve's condition is assessed as normal. Furthermore, abnormal conditions can be further subdivided into minor and severe abnormalities, each corresponding to different maintenance strategies, achieving refined management of the valve's operating status.

[0111] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for real-time tracking and anomaly diagnosis of valve opening and closing status, characterized in that, include: The system acquires real-time valve opening values, branch flow distribution, and pipe roughness changes, performs timestamp alignment and correlation processing, and obtains the pipeline impedance distribution matrix and the mapping function between valve opening and flow rate. Based on the pipeline impedance distribution matrix, the impedance growth rate of each branch is identified, and the correlation coefficient between the impedance growth rate and the historical roughness data sequence is calculated to identify branches with increased roughness. The flow contribution variation is calculated by increasing the impedance growth rate of the branch by increasing the roughness and the preset flow distribution rating, and the interference level of each branch is determined based on the flow contribution variation. The expected flow rate at the target valve opening is calculated based on the mapping function between valve opening and flow rate. The flow rate difference is then compared with the actual flow rate data to identify valves with abnormal flow rates. Obtain the abnormal valve's abnormality type identifier, its associated branch flow data, and interference level; construct an influence weight matrix; and combine this with the set of valve-related nodes to determine the precise coordinates of the abnormal location. Obtain the actual flow rate data and expected flow rate value at the precise coordinates of the abnormal location, calculate the deviation rate, and determine the boundary value for tracking and judging the valve opening and closing status. The valve opening and closing status is evaluated by comparing and correcting the real-time roughness change information with the opening reading through the boundary value of the valve opening and closing status tracking judgment.

2. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 1, characterized in that, The process of acquiring real-time valve opening values, branch flow distribution, and pipe roughness changes, performing timestamp alignment and correlation processing, and obtaining the mapping function between the pipeline impedance distribution matrix and valve opening and flow rate includes: Valve opening values, branch flow monitoring data, changes in pipe roughness coefficient, and node pressure values ​​are acquired by sensors, and the data are aligned according to timestamps to obtain a synchronous pipeline network operation status dataset. Based on the synchronous pipeline network operation status dataset, the pressure difference between adjacent nodes is calculated, and the Darcy-Wiesbach formula is used to calculate the impedance coefficient of each pipe segment using the pressure difference, flow rate and roughness coefficient variation to obtain the pipeline network impedance distribution matrix. By fitting the pipeline impedance distribution matrix and the real-time valve opening value using the least squares method, a mapping function between valve opening and flow rate is established.

3. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 1, characterized in that, The step of identifying the impedance growth rate of each branch based on the pipeline impedance distribution matrix, calculating the correlation coefficient between the impedance growth rate and the historical roughness data sequence, and identifying branches with increasing roughness includes: Obtain the impedance value sequence of each branch at continuous time points in the pipeline impedance distribution matrix, and perform linear regression fitting with time as the independent variable to obtain the impedance growth rate. Extract the historical roughness data sequence of the corresponding branch and calculate the roughness change rate at adjacent time points; The Pearson correlation coefficient is calculated by comparing the roughness change rate with the impedance growth rate to determine the branch with increasing roughness.

4. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 1, characterized in that, The process involves increasing the impedance growth rate of the branch by increasing roughness and calculating the flow contribution variation based on a preset flow distribution rating. The interference level of each branch is then determined based on this flow contribution variation, including: The flow deviation coefficient is calculated by obtaining the rate of increase in impedance of the branch with increased roughness and the preset flow distribution rating. The flow contribution variation is determined based on the flow deviation coefficient and the current actual flow. The influence weight value is calculated by using the change in flow contribution and the total flow of the pipeline network, and the low, medium and high interference levels are determined based on the influence weight value.

5. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 4, characterized in that, It also includes obtaining the current impedance growth rate of the roughness-increasing branch and calculating the flow attenuation ratio based on the preset flow distribution rating, collecting data on the thickness of the roughness layer on the inner wall of the pipe and the degree of pressure loss, and determining the transport difference of the roughness-increasing branch, specifically including: The current impedance growth rate of the branch with increased roughness is obtained and the preset flow distribution rating is used to calculate the flow attenuation ratio. Data on the roughness layer thickness and pressure loss are collected. The influence coefficient is calculated using the Hayzen-Williams formula based on the flow attenuation ratio, roughness layer thickness data and pressure loss degree. The influence coefficient is then multiplied by the original flow capacity of the pipeline to obtain the actual flow capacity. The downstream flow shortage value is obtained by comparing the actual flow capacity with the rated flow demand of the downstream water supply area. The transport difference of the roughness-increased branch is determined based on the downstream flow shortage value and the flow attenuation ratio.

6. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 1, characterized in that, The step of calculating the expected flow rate at the target valve opening based on the mapping function between the valve opening and the flow rate, comparing it with the actual flow rate data to obtain the flow rate difference value, and identifying valves with abnormal flow rates includes: The expected flow rate is obtained by linearly interpolating the current opening value of the target valve according to the mapping function between valve opening and flow rate, and then compared with the actual flow rate data read by the flow sensor to obtain the flow rate difference value. The abnormal valves can be identified as either clogged or leaking based on the positive or negative value of the flow drop.

7. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 1, characterized in that, The process of obtaining the abnormal valve's abnormality type identifier, its associated branch flow data, and interference level; constructing an influence weight matrix; and determining the precise coordinates of the abnormal location by combining the valve's associated node set includes: Obtain the abnormal valve's abnormal type identifier and the real-time flow data of its branch; extract the upstream and downstream node numbers and their spatial coordinates based on the pipeline network topology; and construct a set of valve-related nodes. A judgment matrix is ​​constructed by using anomaly type identifiers, branch flow data and interference levels, and the influence weight matrix is ​​obtained by calculating the weights of each factor. Based on the influence weight matrix, the decreasing influence of anomalies on adjacent nodes is analyzed, and nodes with weights greater than a preset threshold are identified to obtain the anomaly influence distribution vector. The anomaly center location is calculated using the weighted average method based on the spatial coordinates of the three nodes with the largest weights in the anomaly influence distribution vector, thus determining the precise coordinates of the anomaly location.

8. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 1, characterized in that, The process of obtaining the actual flow rate data and expected flow rate value at the precise coordinates of the abnormal location, calculating the deviation rate, and determining the boundary value for valve opening and closing status tracking includes: Obtain the actual flow data and expected flow value of the pipeline node corresponding to the precise coordinates of the abnormal location, and calculate the deviation rate; Based on the statistical distribution of historical operating data, the deviation rate range corresponding to the normal fluctuation range is determined, and the boundary value for tracking and judging the valve opening and closing status is obtained.

9. The method for real-time tracking and anomaly diagnosis of valve opening and closing status according to claim 1, characterized in that, The step of tracking and judging the boundary value of the valve opening and closing state to correct and compare the real-time roughness change information with the opening reading, and evaluating the valve opening and closing state, includes: Obtain real-time roughness change information, calculate the roughness influence factor, and multiply it by the opening value to obtain the corrected opening value; The corrected opening value is compared with the valve opening and closing status tracking and judgment boundary value to determine whether the valve opening and closing status is normal or abnormal.

Images