Intelligent identification system and method for water network leakage of power station in active service

Abstract

The invention provides an active-service power station water network leakage intelligent identification system and method, and the method comprises the steps: forming a holographic sensing layer covering a whole plant water network through the calibration and upgrading of an active-service flow meter and the arrangement of a key node electromagnetic / ultrasonic flow meter; single-point anomaly detection, real-time dynamic balance analysis and accumulated data verification are constructed based on a dynamic water balance algorithm and a topological database, and the limitation of traditional manual inspection is broken through; and a reverse traceability positioning model and a self-learning optimization mechanism are combined to realize accurate positioning of leakage points and dynamic optimization of an early warning threshold. The system effectively solves the problems that hidden leakage of a water network of an old power plant is difficult to find and the leakage amount is difficult to quantify, the operation and maintenance cost is remarkably reduced, and the utilization efficiency of water resources is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline maintenance in power stations, and in particular to an intelligent identification system and method for water network leakage in an active power station. Background Art

[0002] The water pipes of old power plants in service are mainly made of cast iron. As the years of use increase, it is difficult to find damage to the pipes, and the rust and silt in the pipes increase, causing most of the mechanical water meters installed to be unable to accurately measure or even become blocked and unusable. At present, there are widespread water pipe network leakages in old power plants in service, and there are long-term and large-scale water losses. In daily operation and maintenance management, it is necessary to conduct regular manual meter readings of the water network in the entire plant, which takes more than 6 hours each time, with high manpower and time costs, and it is impossible to detect leakages in time, resulting in a waste of water resources. Summary of the invention

[0003] In a first aspect of the present disclosure, a system for intelligently identifying water network leakage in an existing power station is provided, comprising:

[0004] A flow metering module, comprising an existing flow meter that has passed the calibration and a newly added flow meter, wherein the newly added flow meter is an electromagnetic flow meter or a pipe clamp type ultrasonic flow meter;

[0005] A data acquisition module, configured to collect real-time signals of existing flow meters and remote transmission signals of newly added flow meters;

[0006] The data processing module performs leakage judgment, rationality judgment of individual flow values, dynamic water balance judgment of the total system, and water balance judgment of accumulated data;

[0007] The warning output module generates a graded warning signal when any judgment level triggers a leakage condition;

[0008] The system topology database stores pipeline topology relationship data generated based on water network modeling.

[0009] In combination with the first aspect, in the flow metering module:

[0010] The installation location of the newly added flow meter includes at least one of the following nodes: the intersection node of the water inlet main pipe and the branch pipe of each subsystem, the inlet main pipe node of the water-using equipment cluster, the inter-stage transmission node of the multi-stage treatment system, and the confluence node of the recycled water and the feed water.

[0011] In combination with the first aspect, in the data processing module:

[0012] The rationality judgment of the single flow value is specifically as follows: for the measured value Qt of any flow meter within a continuous time period of t1, an early warning is triggered when |Qt-Qavg|>k1σ, where Qavg is the historical benchmark average, σ is the historical standard deviation, and k1 is the adjustment coefficient;

[0013] The specific dynamic water balance judgment is as follows: Set a Δt time window for the total system or subsystem, and trigger an alarm when the deviation rate δ of the input flow rate and the output flow rate is > 5%;

[0014] The specific cumulative data water balance judgment is as follows: For the daily / weekly / monthly cumulative water volume data, trigger an alarm when the deviation rate η of the input and output water volumes is > 5%.

[0015] Combined with the first aspect, the data processing module further includes: a dynamic compensation unit that performs temperature compensation on the flow measurement value according to the ambient temperature and the pipe material expansion coefficient;

[0016] a working condition correction unit that adjusts the water balance judgment threshold according to the operating load rate of the generator set.

[0017] Combined with the first aspect, the system topology database includes: a pipeline attribute data layer that records parameters such as pipe diameter, material, and service life; a flowmeter topology layer that records the spatial position relationship of each flowmeter and the affiliated subsystem; a historical leakage record layer that stores the location characteristic data of previous leakage events.

[0018] In the second aspect of the present disclosure, a method for intelligent identification of water network leakage in an active power station is provided, including the following steps:

[0019] S1. Based on engineering drawings and on-site surveys, establish a list of flowmeters including installation locations and operating conditions;

[0020] S2. Construct a weighted topology graph according to the pipeline connection relationship, and the weights include pipe segment length, pipe diameter, and material parameters;

[0021] S3. Calibrate and replace the existing flowmeters, and install electromagnetic / ultrasonic flowmeters;

[0022] S4. Collect data of existing flowmeters through a signal converter, and obtain data of newly added flowmeters through a wireless remote transmission module;

[0023] S5. Leakage analysis, specifically including: comparing the deviation between the real-time flow rate and the historical reference curve; dynamic water balance analysis, calculating the real-time deviation of the input and output flow rates of the subsystem / total system, and counting the daily / weekly / monthly cumulative water volume deviation;

[0024] S6. Determine the leakage pipe segment based on the topological relationship and the flow anomaly characteristics.

[0025] Combined with the second aspect, in step S3:

[0026] For pipelines with a pipe diameter > DN200, electromagnetic flowmeters are selected, and for pipelines that cannot be shut down, clamp-on ultrasonic flowmeters are used.

[0027] Combined with the second aspect, in step S5:

[0028] The real-time deviation calculation for dynamic water balance analysis adopts the following formula:

[0029] δ(t) = [ΣQ_in(t) - ΣQ_out(t)] / ΣQ_in(t) × 100%,

[0030] where ΣQ_in(t) is the total input flow of the system, and ΣQ_out(t) is the total output flow of the system;

[0031] When δ(t) lasts for more than T1 time > 5%, a first-level warning is generated. When the instantaneous value of δ(t) > 10%, a second-level warning is generated.

[0032] Combined with the second aspect, in the S6 step, the reverse tracing and positioning method is adopted:

[0033] Establish a flow anomaly propagation model: where v is the water flow velocity and α is the leakage coefficient;

[0034] Generate a set of suspicious pipe segments based on the topology database: Suspiciousness score = ω1·|ΔQ| + ω2·pipe age coefficient + ω3·historical leakage frequency;

[0035] Select the pipe segments with scores higher than the threshold as the objects to be investigated.

[0036] Beneficial effects: The present disclosure provides an intelligent leakage identification system and method for the water network of an active power station. Through the calibration and upgrade of existing flow meters and the layout of electromagnetic / ultrasonic flow meters at key nodes, a holographic perception layer covering the entire plant water network is formed; based on the dynamic water balance algorithm and the topology database, single-point anomaly detection, real-time dynamic balance analysis, and cumulative data verification are constructed, breaking through the limitations of traditional manual inspections; combined with the reverse tracing and positioning model and the self-learning optimization mechanism, accurate positioning of leakage points and dynamic optimization of warning thresholds are achieved. This system effectively solves the problems of difficult detection of hidden leakage in the water network of old power plants and difficult quantification of leakage volume, significantly reducing the operation and maintenance costs and improving the water resource utilization efficiency. Description of the Drawings

[0037] Figure 1 It is a schematic structural diagram of an intelligent leakage identification system for the water network of an active power station according to an embodiment of the present disclosure;

[0038] Figure 2 It is a schematic flow diagram of an intelligent leakage identification method for the water network of an active power station according to an embodiment of the present disclosure. Detailed Embodiments

[0039] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present disclosure.

[0040] The terms used in the embodiments of the present disclosure are for the purpose of describing specific embodiments only and are not intended to limit the embodiments of the present disclosure. The singular forms "a", "the", and "said" used in the embodiments of the present disclosure and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0041] It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the embodiments of the present disclosure, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining".

[0042] As Figure 1 shown, it is a schematic structural diagram of an intelligent identification system for water network leakage in an active power station according to an embodiment of the present disclosure, including:

[0043] A flow measurement module 110, including a calibrated active flowmeter and a newly added flowmeter, and the newly added flowmeter adopts an electromagnetic flowmeter or a clamp-on ultrasonic flowmeter;

[0044] A data acquisition module 120, configured to collect the signals of the active flowmeter and the remote transmission signals of the newly added flowmeter in real time;

[0045] A data processing module 130, which performs leakage judgment, rationality judgment of a single flow value, dynamic water balance judgment of the total system, and cumulative data water balance judgment;

[0046] An early warning output module 140, which generates a hierarchical early warning signal when any judgment level triggers a leakage condition;

[0047] A system topology database 150, which stores the pipeline topology relationship data generated based on water network modeling.

[0048] The flow measurement module (110) is the data acquisition basis of the system and is responsible for measuring the flow of the water network. It includes:

[0049] Existing flowmeter: Refers to the flowmeter originally installed in the power station and calibrated to be qualified, which is used to provide long-term operating water flow measurement data.

[0050] Newly added flowmeter: To improve measurement accuracy and monitoring capabilities, new flowmeters are added to the system, which are of the following two types:

[0051] Electromagnetic flowmeter: Suitable for large-diameter pipelines and can provide high-precision measurements.

[0052] Clamp-on ultrasonic flowmeter: Suitable for pipelines that cannot be shut down, with convenient installation and no need to damage the pipeline structure.

[0053] Data acquisition module (120), which is responsible for collecting and transmitting flow data to ensure real-time monitoring of the operation status of the water network.

[0054] Collect signals from existing flowmeters to ensure compatibility with the original system.

[0055] Collect remote signals from newly added flowmeters, and wireless communication, industrial bus or other data transmission methods can be used for remote monitoring.

[0056] Data processing module (130), which performs core data analysis and leakage judgment, including the following three levels of analysis logic:

[0057] Rationality judgment of single flow value: Analyze whether the readings of individual flowmeters are abnormal. For example, if the measured value of a certain flowmeter deviates too much from the historical average over a period of time, abnormal detection is triggered.

[0058] Total system dynamic water balance judgment: Calculate the deviation between the input flow and the output flow of the entire water network or subsystem within a specific time window (such as 5 minutes, 10 minutes). If the deviation exceeds the threshold (such as 5%), it is determined that there is an abnormality.

[0059] Cumulative data water balance judgment: Analyze the cumulative input and output water volumes over a long period (such as daily, weekly, monthly). If the deviation exceeds the threshold (such as 5%), the leakage risk is further confirmed.

[0060] Early warning output module (140), when the data processing module detects an abnormality at any level (such as the flow deviation exceeding the threshold), this module will generate early warning signals of different levels, such as:

[0061] Level 1 early warning: Small deviation, indicating that there is a trend of leakage and suggesting inspection.

[0062] Level 2 early warning: Severe deviation, requiring immediate investigation of the cause of leakage.

[0063] Early warning signals can be transmitted to operation and maintenance personnel through the SCADA system, text messages, APP push or control center alarms, etc. for timely handling.

[0064] The system topology database (150) stores the topology structure and key parameters of the water network, providing basic data for leakage analysis, including:

[0065] The pipeline attribute data layer: records information such as the pipe diameter, material, and service life of the pipeline.

[0066] The flowmeter topology layer: stores the installation location of the flowmeter and its relationship in the system.

[0067] The historical leakage record layer: saves the data of past leakage events, including the occurrence time, location, and repair situation, for system optimization and trend analysis.

[0068] Beneficial effects: Through flow monitoring, data collection, data analysis, and early warning response, the intelligent identification system realizes the efficient identification of water network leakage in active power stations, can improve the utilization rate of water resources, reduce ineffective losses, and lower operating costs.

[0069] Furthermore, in the flow measurement module, the installation locations of newly added flowmeters include at least one of the following nodes: the intersection node of the main inlet pipe and branch pipes of each subsystem, the inlet header node of the water-using equipment cluster, the inter-stage transmission node of the multi-stage treatment system, and the confluence node of recycled water and make-up water.

[0070] Specifically, the intersection node of the main inlet pipe and branch pipes of each subsystem refers to the intersection of the main water supply pipe and the branch pipes. Installing a flowmeter at this position can effectively monitor the water inlet situation of each subsystem and ensure that the water supply volume of each subsystem meets the expectations.

[0071] The inlet header node of the water-using equipment cluster refers to the inlet position of the common water supply main pipe for multiple water-using equipment. Installing a flowmeter at this position can monitor the water volume change of the entire equipment cluster and timely detect abnormal water use or unexpected leakage of the equipment.

[0072] The inter-stage transmission node of the multi-stage treatment system refers to the water flow transmission position between different treatment levels in the water treatment system. The flowmeter at this position can ensure that the input and output flows of each treatment level meet the process requirements and timely detect abnormal changes.

[0073] The confluence node of recycled water and make-up water refers to the confluence position of recycled water and new make-up water. Installing a flowmeter at this position can be used to monitor the utilization situation of recycled water and the make-up volume of make-up water to ensure the stability of the overall water cycle.

[0074] Furthermore, in the data processing module:

[0075] The specific judgment of the rationality of the single flow value is as follows: for the measured value Qt of any flowmeter within a continuous time period t1, when |Qt - Qavg| > k1σ, an alarm is triggered, where Qavg is the historical reference average value, σ is the historical standard deviation, and k1 is an adjustment coefficient;

[0076] The specific dynamic water balance judgment is as follows: for the total system or subsystem, a Δt time window is set, and when the deviation rate δ of the input flow and the output flow is > 5%, an alarm is triggered;

[0077] The specific cumulative data water balance judgment is as follows: for the daily / weekly / monthly cumulative water volume data, when the deviation rate η of the input and output water volumes is > 5%, an alarm is triggered.

[0078] Specifically, this rationality judgment method is used to detect short-term abnormal conditions of a single flowmeter, such as sensor drift, pipeline blockage, or sudden leakage.

[0079] When the deviation of the measured value Qt of the flowmeter within a continuous time period t1 exceeds k1 times the standard deviation σ, that is, there is a significant difference from the historical reference mean value Qavg, the system will trigger an alarm to promptly check the equipment or pipeline network.

[0080] This method can effectively identify sudden anomalies, reduce false alarms caused by measurement errors or short-term fluctuations, and simultaneously improve the detection ability for real abnormal situations.

[0081] The dynamic water balance judgment is to set a Δt time window and calculate the deviation between the total input flow and the total output flow of the entire system or subsystem within this time period. When the deviation rate δ exceeds 5%, the system will trigger an alarm, indicating the existence of water leakage or abnormal water use. This method is applicable to the balance detection within a short time range and can be used to discover water volume losses caused by abnormal equipment operation or local leakage.

[0082] The cumulative data water balance judgment is applicable to long-term water volume analysis. By statistically analyzing the input water volume and output water volume on a longer time scale such as daily, weekly, and monthly, when the cumulative deviation rate η exceeds 5%, an alarm is triggered. This method can identify long-term slowly developing water leakage problems, such as pipeline aging and concealed leakage, thus providing a more comprehensive water volume monitoring ability for the system.

[0083] Furthermore, the data processing module further includes: a dynamic compensation unit that performs temperature compensation on the flow measurement value according to the ambient temperature and the pipe material expansion coefficient;

[0084] A working condition correction unit that adjusts the water balance judgment threshold according to the operating load rate of the generator set.

[0085] Specifically, the dynamic compensation unit is used to eliminate the influence of environmental temperature changes on the measured values of the flowmeter, especially for pipelines with a large coefficient of thermal expansion of the pipe material, such as the thermal expansion effect of plastic pipes or metal pipes. In an environment with large temperature changes, the pipe diameter will change slightly with the temperature, affecting the flow measurement accuracy. The dynamic compensation unit calculates the temperature compensation value by obtaining the ambient temperature in real time and combining the coefficient of expansion of the pipe material, thereby correcting the flow measurement data and improving the measurement accuracy.

[0086] The operating condition correction unit is used to dynamically adjust the warning threshold for the water balance judgment according to the operating load of the generator set. When the unit load is high, there will be short-term water volume fluctuations. Therefore, appropriately increasing the deviation threshold can reduce false alarms. When the unit load is low, the water volume fluctuation is small, and the threshold should be lowered to improve the detection sensitivity in order to more accurately detect leakage problems. This method can optimize the warning threshold of the system according to the operating conditions, avoiding false alarms or missed alarms caused by unreasonable setting of fixed thresholds.

[0087] Furthermore, the system topology database includes: a pipeline attribute data layer for recording parameters such as pipe diameter, material, and service life; a flowmeter topology layer for recording the spatial position relationship of each flowmeter and its affiliated subsystem; and a historical leakage record layer for storing the position characteristic data of previous leakage events.

[0088] Specifically, the system topology database is used to store the water network structure and historical operation data. Among them, the pipeline attribute data layer contains the basic information of each pipeline, such as pipe diameter, material, service life, etc., providing a basis for pipeline life assessment and leakage analysis.

[0089] The flowmeter topology layer is used to record the spatial position information of all flowmeters and their affiliated subsystems, ensuring that the water network monitoring system can perform data analysis based on the topological structure. The historical leakage record layer stores all the occurred leakage events, including information such as occurrence time, location, and leakage volume, providing historical data support for the system to optimize the leakage detection model and improve the prediction and warning capabilities.

[0090] As Figure 2 shown, it is a schematic flow chart of a method for intelligent identification of water network leakage in an active power station according to an embodiment of the present disclosure, including:

[0091] S1. Based on engineering drawings and on-site surveys, establish a list of flowmeters including installation locations and operating conditions;

[0092] S2. Construct a weighted topology graph according to the pipeline connection relationship, and the weights include pipe segment length, pipe diameter, and material parameters;

[0093] S3. Calibrate and replace the existing flowmeters, and install electromagnetic / ultrasonic flowmeters;

[0094] S4. Collect the data of existing flow meters through a signal converter and obtain the data of newly added flow meters through a wireless remote transmission module;

[0095] S5. Leakage analysis, specifically including: comparing the deviation between the real-time flow rate and the historical reference curve; dynamic water balance analysis, calculating the real-time deviation of the input and output flow rates of the subsystem / total system, and statistically analyzing the daily / weekly / monthly cumulative water volume deviation;

[0096] S6. Determine the leaky pipe section based on the topological relationship and flow anomaly characteristics.

[0097] Further, in step S3:

[0098] Select electromagnetic flow meters for pipelines with a diameter > DN200, and use clamp-on ultrasonic flow meters for pipelines that cannot be shut down.

[0099] Specifically, in the existing power plant water network leakage detection system, the selection of flow meters needs to consider the pipe diameter and construction conditions. For pipelines with a diameter greater than DN200, electromagnetic flow meters are preferentially selected due to their high measurement accuracy and adaptability to different water qualities.

[0100] Electromagnetic flow meters rely on fluid media with relatively high conductivity and are suitable for accurate measurement of large-diameter pipelines. For pipelines that cannot be shut down, it is difficult to implement traditional insertion-type or pipeline-cutting-installed flow meters. Therefore, clamp-on ultrasonic flow meters are used.

[0101] This flow meter measures the flow velocity through the time difference of ultrasonic signal propagation, does not require damage to the pipeline structure, is suitable for installation without stopping water, and can complete the upgrade of flow monitoring equipment in a short time, improving the comprehensiveness and accuracy of data collection.

[0102] Further, in step S5:

[0103] The real-time deviation calculation of dynamic water balance analysis adopts the following formula:

[0104] δ(t) = [ΣQ in(t) - ΣQout(t)] / ΣQ in(t) × 100%,

[0105] In the formula, ΣQ in(t) is the total input flow rate of the system, and ΣQout(t) is the total output flow rate of the system;

[0106] When δ(t) lasts for more than T1 time > 5%, a first-level warning is generated. When the instantaneous value of δ(t) > 10%, a second-level warning is generated.

[0107] Specifically, dynamic water balance analysis is used to monitor the flow balance of the water network system in real time. ΣQ i n(t) in the calculation formula represents the total input flow at the current moment, ΣQout(t) represents the total output flow, and δ(t) represents the current relative deviation. When δ(t) continuously exceeds 5% within T1 time, it indicates that there is a relatively stable leakage trend, and the system triggers a first-level warning to prompt relevant personnel to conduct a preliminary investigation.

[0108] When the instantaneous value of δ(t) exceeds 10%, it indicates that sudden leakage has occurred, such as pipeline rupture or equipment failure. At this time, the system triggers a second-level warning and requires emergency maintenance. By setting warning thresholds at different levels, this method realizes double monitoring of long-term slow leakage and sudden leakage, and improves the system's ability to identify abnormalities.

[0109] Furthermore, the reverse tracing and positioning method is adopted in step S6:

[0110] Establish a flow anomaly propagation model: where v is the water flow velocity and α is the leakage coefficient;

[0111] Generate a set of suspicious pipe segments based on the topology database: Suspect degree score = ω1·|ΔQ| + ω2·pipe age coefficient + ω3·historical leakage frequency;

[0112] Select the pipe segments with scores higher than the threshold as the objects for investigation.

[0113] Specifically, the reverse tracing and positioning method uses the flow anomaly propagation model to analyze the flow loss in the water flow. In the equation, represents the change of flow rate over time, represents the propagation of flow rate in space, and -αQ represents the flow rate attenuation caused by leakage.

[0114] Through this model, the leakage points in the process of water flow propagation along the pipeline can be estimated. Combining with the topology database, the system further calculates the suspect degree score. ω1·|ΔQ| represents the degree of abnormal flow rate change, ω2·pipe age coefficient reflects the degree of pipeline aging, and ω3·historical leakage frequency represents the possibility of leakage occurring in this pipe segment in the past. The pipe segments with scores higher than the set threshold are selected as the key objects for investigation to narrow the scope of maintenance and improve the efficiency of leakage location.

[0115] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit it; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. An intelligent identification system for water network leakage in an active power station, characterized in that, Including: A flow measurement module, which includes qualified in-service flow meters and newly added flow meters. The newly added flow meters are electromagnetic flow meters or clamp-on ultrasonic flow meters. A data acquisition module, configured to collect in-service flow meter signals and remote transmission signals of newly added flow meters in real time. A data processing module, which performs leakage judgment, rationality judgment of individual flow values, dynamic water balance judgment of the total system, and cumulative data water balance judgment. An early warning output module, which generates hierarchical early warning signals when leakage conditions are triggered at any judgment level. A system topology database, which stores pipeline topology relationship data generated based on water network modeling.

2. The system according to claim 1, wherein In the flow measurement module: The installation positions of the newly added flow meters include at least one of the following nodes: the intersection node of the inlet main pipe and branch pipe of each subsystem, the inlet header node of the water-using equipment cluster, the inter-stage transmission node of the multi-stage treatment system, and the confluence node of reclaimed water and makeup water.

3. The system according to claim 1, wherein In the data processing module: The rationality judgment of the individual flow value is specifically as follows: for the measured value Qt of any flow meter within the continuous t1 time period, when |Qt - Qavg| > k1σ is satisfied, an early warning is triggered, where Qavg is the historical reference average value, σ is the historical standard deviation, and k1 is an adjustment coefficient. The dynamic water balance judgment is specifically as follows: for the total system or a subsystem, a Δt time window is set. When the deviation rate δ of the input flow and the output flow > 5%, an early warning is triggered. The cumulative data water balance judgment is specifically as follows: for the daily / weekly / monthly cumulative water volume data, when the deviation rate η of the input and output water volumes > 5%, an early warning is triggered.

4. The system according to claim 3, wherein The data processing module further includes: a dynamic compensation unit, which performs temperature compensation on the flow measurement value according to the ambient temperature and the pipe material expansion coefficient. A working condition correction unit, which adjusts the water balance judgment threshold according to the operating load rate of the generator set.

5. The system according to claim 1, wherein The system topology database includes: a pipeline attribute data layer, which records parameters such as pipe diameter, material, and service life; a flow meter topology layer, which records the spatial position relationship of each flow meter and the affiliated subsystem; and a historical leakage record layer, which stores the location characteristic data of previous leakage events.

6. An intelligent identification method for water network leakage in an existing power station, based on the system described in claim 1, characterized in that, Including the following steps: S1. Based on engineering drawings and on-site surveys, establish a list of flow meters including installation positions and working conditions. S2. Construct a weighted topology graph according to the pipeline connection relationship, and the weights include pipe section length, pipe diameter, and material parameters. S3. Calibrate and replace the in-service flow meters, and install electromagnetic / ultrasonic flow meters. S4. Collect data of in-service flow meters through signal converters, and obtain data of newly added flow meters through wireless remote transmission modules. S5. Leakage analysis, specifically including: comparing the deviation between the real-time flow and the historical reference curve. Dynamic water balance analysis, calculating the real-time deviation of the input and output flows of the subsystem / total system, and statistically analyzing the daily / weekly / monthly cumulative water volume deviation. S6. Determine the leakage pipe section based on the topology relationship and flow anomaly characteristics.

7. The method according to claim 6, wherein In step S3: For pipelines with a pipe diameter > DN200, electromagnetic flow meters are selected, and for pipelines that cannot be shut down, clamp-on ultrasonic flow meters are used.

8. The method according to claim 6, wherein In step S5: The real-time deviation calculation of the dynamic water balance analysis uses the following formula: δ(t) = [ΣQin(t) - ΣQout(t)] / ΣQin(t) × 100%, Where ΣQin(t) is the total input flow rate of the system, and ΣQout(t) is the total output flow rate of the system; Generate a first-level warning when δ(t) lasts for T1 time > 5%, and generate a second-level warning when the instantaneous value of δ(t) > 10%.

9. The method according to claim 6, characterized in that, In the S6 step, the reverse traceability positioning method is adopted: Establish a flow anomaly propagation model: where v is the water flow velocity and α is the leakage coefficient; Generate a set of suspicious pipe segments based on the topology database: Suspiciousness score = ω1·|ΔQ| + ω2·Pipe age coefficient + ω3·Historical leakage frequency; Select the pipe segments with scores higher than the threshold as the objects for investigation.

Images