Intelligent Leakage Monitoring and Early Warning System for Water Supply Networks Adapted to Harsh Operating Conditions in Electrolytic Aluminum Plants

By employing anti-interference acquisition, dual-mode transmission and error correction, operating condition linkage processing, and leakage identification modules in the water supply network of an electrolytic aluminum plant, the problems of magnetic temperature deviation and process disturbance confusion in leakage monitoring of the water supply network of an electrolytic aluminum plant have been solved, achieving more reliable leakage identification and section location.

CN123110550APending Publication Date: 2026-11-17HENAN DONGDA METALLURGICAL TECH SERVICE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611277466.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Under the complex operating conditions of the water supply network in electrolytic aluminum plants, existing leak monitoring technologies suffer from residual measurement biases caused by the combined effects of magnetic fields and temperature. Normal process disturbances can easily be confused with leak anomalies, and the source-end magnetic temperature correction information lacks a continuous correlation with the downstream leak sample determination, affecting the reliability of leak identification and section location.

Method used

An anti-interference acquisition module is used to reduce power frequency electromagnetic interference and correct magnetic temperature deviation. A dual-mode transmission and error correction module is used to restore continuous parameters. A working condition linkage processing module is used to filter out false fluctuations. A leak identification module is used to form multi-dimensional physical features and combine them with the pipeline topology to determine the leak section. An early warning and location module is used for early warning and location.

Benefits of technology

It improves the reliability of leak identification and section location in the water supply network of electrolytic aluminum plants, and reduces the interference of residual magnetic temperature deviation and normal process disturbance on leak judgment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123110550A_ABST
    Figure CN123110550A_ABST
Patent Text Reader

Abstract

This invention discloses an intelligent leak monitoring and early warning system for water supply networks adapted to the harsh operating conditions of electrolytic aluminum plants. It belongs to the field of pipeline and water supply network monitoring technology. The method of this invention is as follows: Pressure, flow, temperature, and pipeline vibration data of the water supply network are collected. Target network operating parameters and magnetic temperature correction information are generated through electromagnetic shielding and magnetic temperature coupling compensation. Error-corrected network operating parameters are obtained through dual-mode transmission and electromagnetic disturbance correction. These parameters are then associated with corresponding time periods such as pole switching, anode rise / fall, rectification and pressure regulation, or equipment start / stop conditions. Corresponding anomaly judgment thresholds are called or updated, and false fluctuations are filtered out. Multi-dimensional physical features are constructed from standard network operating parameters and magnetic temperature correction information, and the leak result is determined by corresponding network leak feature samples. The leak section is output based on the early warning level and network topology. The beneficial effects of this invention are that it reduces the interference of residual magnetic temperature deviation and normal process disturbances on leak judgment, improving the reliability of leak identification and section location.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of pipeline and water supply network monitoring technology, specifically to an intelligent leak monitoring and early warning system for water supply networks adapted to the harsh working conditions of electrolytic aluminum plants. Background Technology

[0002] Electrolytic aluminum production requires the continuous operation of pipelines for cooling circulating water, pressure-stabilizing pure water, and heat dissipation water supply for rectifier equipment. These pipelines are distributed throughout the electrolysis workshop, rectification area, and open-air areas of the plant. The on-site monitoring environment is complex due to the influence of high-current equipment, high-temperature heat radiation, alumina dust, and trace corrosive media. Electrode switching, anode raising and lowering, rectification and voltage regulation, and equipment start-up and shutdown can also cause transient changes in pressure or flow.

[0003] Existing pipeline leak monitoring technologies typically collect operational data such as pressure, flow, temperature, acoustics, or vibration, which are then uploaded to a server via a communication link. Leaks are identified through fixed or dynamic thresholds, time-series characteristics, historical samples, or multi-parameter joint analysis, and abnormal sections are located by combining pipeline topology. Some solutions also employ environmental compensation, data error correction, or operational condition information to reduce the impact of measurement deviations and normal operation adjustments.

[0004] However, under the complex operating conditions of the water supply network in an electrolytic aluminum plant, electromagnetic shielding alone may still retain residual measurement deviations caused by the combined effects of magnetic field and temperature. If the network operation parameters are not established in accordance with the production conditions at the same time and the anomaly judgment threshold is not adjusted accordingly, normal process disturbances can easily lead to leak identification. At the same time, if the source-end magnetic temperature correction information does not continue to participate in the downstream leak sample judgment, there is a lack of continuous correlation between data acquisition calibration and leak identification, which affects the reliability of leak identification and section location.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] This invention provides an intelligent leak monitoring and early warning system for water supply networks adapted to the harsh operating conditions of electrolytic aluminum plants. It addresses the problems of residual magnetic temperature deviation after shielding easily entering subsequent leak judgments under the combined effects of strong power frequency electromagnetic fields, high temperatures, and periodic production conditions in the water supply networks of electrolytic aluminum plants; pressure or flow fluctuations caused by normal production operations easily being confused with actual leak anomalies; and the lack of continuous correlation between source-end magnetic temperature correction information and back-end leak sample judgment, which affects the reliability of leak identification and section location.

[0007] The technical solution of this invention is as follows: an intelligent leak monitoring and early warning system for water supply networks adapted to the harsh operating conditions of electrolytic aluminum plants, comprising the following modules: an anti-interference acquisition module: used to acquire pressure, flow, temperature, and pipeline vibration data of the water supply network; reduce power frequency electromagnetic interference through a multi-layer electromagnetic shielding structure; correct the magnetic temperature interference deviation after shielding by calling a specially calibrated magnetic temperature coupling compensation coefficient based on the magnetic field and temperature of the monitoring point; and generate target network operating parameters and magnetic temperature correction information; a dual-mode transmission and error correction module: used to select a spread spectrum penetration transmission node or a dedicated network transmission gateway to transmit the target network operating parameters based on the steel structure shielding conditions; perform data identification and error correction when periodic electromagnetic disturbances are detected; and generate error-corrected network operating parameters; and an operating condition linkage processing module: used to obtain the operating parameters within the time period corresponding to the error-corrected network operating parameters. The system uses time-series data of production conditions such as pole switching, anode raising / lowering, rectification and pressure regulation, or equipment start-up and shutdown to determine the current production condition and pressure and flow process disturbances. It then calls or updates the anomaly judgment threshold corresponding to the current production condition, filters out corresponding false fluctuations in the error-correcting pipeline operating parameters, and generates standard pipeline operating parameters. A leak identification module extracts pressure time-series fluctuation gradients, flow day-night operating condition differences, zoned production-sales differences, and pipeline corrosion vibration noise from the standard pipeline operating parameters. These are used in conjunction with the magnetic temperature correction information as multi-dimensional physical features. The leak identification result is determined based on the correspondence between the multi-dimensional physical features and pipeline leak feature samples in the operating condition database. An early warning and location module determines the early warning level based on the leak identification result, identifies the leaking pipeline section based on the pipeline topology, and outputs early warning and location information.

[0008] Optionally, the anti-interference acquisition module is further configured to: install pressure sensors, flow sensors, temperature sensors, and vibration detection units at the water supply network monitoring point, forming a protective acquisition space through a corrosion-resistant and explosion-proof housing and a multi-layer copper foil composite electromagnetic shielding structure installed inside the housing; acquire pressure, flow, temperature, and pipeline vibration data at the monitoring point, and associate the pressure, flow, temperature, and pipeline vibration data with the corresponding acquisition time; and calibrate the pressure, flow, temperature, and pipeline vibration data by calling the corresponding magnetothermal coupling compensation coefficient according to the magnetic field environment and temperature environment of the monitoring point, thereby obtaining the target pipeline network operating parameters.

[0009] Optionally, the dual-mode transmission and error correction module is further used to: set up LoRa spread spectrum penetration transmission nodes in the steel structure shielded area of ​​the electrolysis workshop or rectifier room, and set up industrial 4G private network transmission gateways in the open-air long-distance pipeline network of the plant area; select the corresponding transmission link according to the area where the monitoring point is located to upload the target pipeline network operation parameters; detect the periodic electromagnetic disturbances generated by rectification voltage regulation or electrolysis cell start-up and shutdown, and when the periodic electromagnetic disturbances occur, perform interference data identification, error correction and completion, and breakpoint continuation on the transmitted data to obtain the continuous error-corrected pipeline network operation parameters.

[0010] Optionally, the operating condition linkage processing module is further configured to: divide the operating condition database into a pipeline topology process unit, a production operating condition time sequence unit, a magnetic temperature interference correction unit, and a pipeline leakage characteristic sample unit; acquire pole switching operating condition data, anode rise and fall operating condition data, rectification and pressure regulation data, or equipment start and stop data within the time period corresponding to the error correction pipeline operating parameters; determine the current production operating condition based on the production operating condition time sequence data, and call the anomaly judgment threshold corresponding to the current production operating condition to filter process disturbances in the pressure and flow changes in the error correction pipeline operating parameters.

[0011] Optionally, the anti-interference acquisition module is further used to: obtain the magneto-thermal coupling compensation coefficients corresponding to different magnetic field environments and temperature environments through high-temperature aging, power frequency electromagnetic compatibility, and dust corrosion durability calibration; correlate the pressure, flow, temperature, and pipeline vibration data collected through the multi-layer copper foil composite electromagnetic shielding structure with the current magnetic field environment and temperature environment, and correct the magneto-thermal interference deviation that still exists after shielding according to the corresponding magneto-thermal coupling compensation coefficient; and output the corrected pressure, flow, temperature, and pipeline vibration data as the target pipeline network operating parameters to the dual-mode transmission and error correction module.

[0012] Optionally, the operating condition linkage processing module is further configured to: determine the operating condition status based on the production operating condition timing data corresponding to electrode switching, anode raising / lowering, rectification and pressure regulation, or equipment start-up and shutdown; acquire the process disturbance changes in pressure and flow under the operating condition status, and update the anomaly judgment threshold used for leak anomaly judgment accordingly; use the updated anomaly judgment threshold to identify false fluctuations in pressure or flow corresponding to the current operating condition in the error correction pipeline operating parameters, and filter the false fluctuations to obtain standard pipeline operating parameters for use by the leak identification module.

[0013] Optionally, the leak identification module is further configured to: extract pressure time-series fluctuation gradient, flow day-night difference, and zone production-sales difference from the standard pipeline network operating parameters, and obtain pipeline corrosion vibration noise and magnetic temperature correction information; use the pressure time-series fluctuation gradient, flow day-night difference, zone production-sales difference, pipeline corrosion vibration noise, and magnetic temperature correction information as multi-dimensional physical features for determining water supply network leaks; determine the leak identification result based on the correspondence between the multi-dimensional physical features and pipeline network leak feature samples in the operating condition database, wherein the leak identification result includes high-temperature corrosion hidden leaks, thermal expansion seepage leaks, or pressure pipe bursts.

[0014] Optionally, the leak identification module is further configured to: determine the pressure time-series fluctuation gradient from the standard pipeline operating parameters, and obtain the magnetic temperature correction information; combine the pipeline corrosion vibration noise characterization to represent the pressure and vibration changes of the pipeline under the current magnetic temperature conditions; match the pressure changes, vibration changes, and regional production and sales differences with the pipeline leak feature samples; when the combined features of the pressure changes, vibration changes, and regional production and sales differences match the high-temperature corrosion dark leak samples, determine the leak identification result as a high-temperature corrosion dark leak, and distinguish it from the identification results of thermal expansion leakage or pressure pipe burst.

[0015] Optionally, the early warning and location module is further configured to: obtain the leakage amount, pressure fluctuation amplitude, and affected core equipment range corresponding to the leakage identification result; determine the early warning level based on the leakage amount, pressure fluctuation amplitude, and affected core equipment range; associate the leakage identification result with the pipeline topology process data in the operating condition database to determine the pipeline topology segment where the leakage occurred; and combine the early warning level and the pipeline topology segment to generate the early warning and location information and output it.

[0016] Optionally, the early warning and positioning module is further configured to: classify the leakage identification results into early warning levels corresponding to minor corrosion leakage, moderate pipeline leakage, or major pressure pipe burst based on the leakage amount, pressure fluctuation amplitude, and affected core equipment range; match corresponding differentiated operation and maintenance push or emergency repair prompt strategies according to the early warning level; and associate the pipeline topology section with the corresponding early warning level and operation and maintenance prompts for output.

[0017] The beneficial effects of this invention are as follows:

[0018] This invention first uses a multi-layer electromagnetic shielding module to reduce power frequency electromagnetic interference through an anti-interference acquisition module. Then, based on the magnetic field and temperature at the monitoring point, it calls a specially calibrated magnetic-temperature coupling compensation coefficient to correct the magnetic-temperature interference deviation after shielding, ensuring that the target pipeline operating parameters and magnetic-temperature correction information have a clear source. A dual-mode transmission and error correction module restores continuous pipeline operating parameters, and a working condition linkage processing module establishes a relationship between the corrected pipeline operating parameters and corresponding time periods such as pole switching, anode rise / fall, rectification and pressure regulation, or equipment start / stop conditions. It calls or updates corresponding anomaly judgment thresholds to filter out false fluctuations in pressure or flow. Furthermore, the standard pipeline operating parameters and magnetic-temperature correction information are combined to form multi-dimensional physical characteristics and correlated with leakage characteristic samples. Combined with the warning level and pipeline topology, the leak section is determined, thereby reducing the interference of residual magnetic-temperature deviation and normal process disturbances on leak judgment and improving the reliability of leak identification and section location in the electrolytic aluminum plant's water supply network. Attached Figure Description

[0019] Figure 1This is a schematic diagram of the overall architecture and data flow of an intelligent leak monitoring and early warning system for water supply networks provided in one embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram of anti-interference acquisition, magnetic temperature compensation and dual-mode transmission deployment provided in one embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of production condition timing alignment and pressure / flow pseudo-fluctuation filtering provided in an embodiment of the present invention;

[0022] Figure 4 This is a schematic diagram of the five-dimensional physical feature composition and leakage feature sample matching provided in one embodiment of the present invention;

[0023] Figure 5 This is a schematic diagram illustrating the location and hierarchical early warning of leaking pipeline network topology sections according to an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0025] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0026] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0027] The water supply network involved in this invention can be an electrolytic cell cooling circulating water network, a pure water pressure-stabilized water supply network, or a rectifier equipment heat dissipation network. The system can operate in an industrial water supply environment comprised of the electrolysis workshop, rectifier room, and open-air network within the plant area. Data processing is collaboratively completed by acquisition equipment located at monitoring points in the water supply network, a dual-mode transmission link, an industrial server, and an operating condition database. The industrial server can employ a dual-machine hot standby mode to meet the data access requirements of a continuous production environment. The continuous operating time is configured according to the electrolytic aluminum production system (e.g., based on...). The continuous production cycle configuration (derived from the continuous production system configuration for electrolytic aluminum) is used to improve the continuity of data processing without changing the data flow between modules.

[0028] The industrial server can also support high-frequency, second-level concurrent data access and processing across multiple production networks. The data access cycle is determined by the on-site acquisition configuration and the server's concurrent task configuration (e.g., according to...). The data collection cycle is concurrently accessed from each monitoring point (the sampling cycle is consistent with the example sampling cycle used in the subsequent working condition time sequence). The dual-machine hot standby mode is used to maintain the continuity of data access and processing during continuous production.

[0029] As mentioned earlier, the water supply network of the electrolytic aluminum plant is simultaneously affected by strong power frequency electromagnetic interference, continuous high-temperature heat radiation, alumina dust, trace corrosion from fluorinated media, and periodic fluctuations in production conditions. The rectifier system operates in a high-current environment (e.g., the rectified current is...). The current level (derived from the operating condition configuration of the electrolytic aluminum rectifier system) is used to characterize the operating background of the rectifier equipment; the magnetic field environment of the monitoring terminal is characterized by magnetic induction intensity, not expressed in current units. Meanwhile, pole switching, anode raising and lowering, rectifier voltage regulation, and equipment start-up and shutdown can cause normal transient changes in pressure or flow. If the above production conditions lack a corresponding relationship with leakage detection, normal process disturbances can easily be included in leakage identification.

[0030] To address this, the present invention provides an intelligent leak monitoring and early warning system for water supply networks adapted to the harsh operating conditions of electrolytic aluminum plants. The system uses an anti-interference acquisition module to reduce power frequency electromagnetic interference at the source and correct magnetic temperature interference deviations that still exist after shielding. A dual-mode transmission and error correction module restores continuous network operating parameters. An operating condition linkage processing module establishes a relationship between the corrected network operating parameters and the corresponding production conditions for the time period and filters out corresponding false fluctuations. A leak identification module then uses standard network operating parameters and magnetic temperature correction information to form multi-dimensional physical features and establishes a correspondence with network leak feature samples. Finally, an early warning and location module completes the early warning classification and network topology segment location.

[0031] The following is combined Figures 1 to 5 This invention is described in detail.

[0032] Example 1:

[0033] like Figure 1 As shown in the figure, this embodiment of the invention provides an intelligent leak monitoring and early warning system for water supply networks adapted to the harsh operating conditions of electrolytic aluminum plants. In this first embodiment, the system includes an anti-interference acquisition module, a dual-mode transmission and error correction module, an operating condition linkage processing module, a leak identification module, and an early warning and location module.

[0034] Among them, the anti-interference acquisition module is used to collect pressure, flow, temperature and pipeline vibration data of water supply network. It reduces power frequency electromagnetic interference through multi-layer electromagnetic shielding structure, and corrects the magnetic temperature interference deviation after shielding by calling the specially calibrated magnetic temperature coupling compensation coefficient according to the magnetic field and temperature of the monitoring point, and generates target network operation parameters and magnetic temperature correction information.

[0035] The dual-mode transmission and error correction module is used to select a spread spectrum penetration transmission node or a private network transmission gateway to transmit the target pipeline operation parameters according to the steel structure obstruction conditions. When periodic electromagnetic disturbances are detected, data identification and error correction are performed to generate error-corrected pipeline operation parameters.

[0036] The operating condition linkage processing module is used to acquire the production operating condition time sequence data of pole switching, anode raising and lowering, rectification and pressure regulation or equipment start-up and shutdown within the time period corresponding to the error correction pipeline operating parameters, determine the current production operating condition and pressure and flow process disturbances, call or update the abnormal judgment threshold corresponding to the current production operating condition, filter the corresponding false fluctuations in the error correction pipeline operating parameters, and generate standard pipeline operating parameters.

[0037] The leak identification module is used to extract pressure time-series fluctuation gradient, flow day-night operating condition difference, regional production and sales difference, and pipeline corrosion vibration noise from the standard pipeline network operating parameters, and use them and the magnetic temperature correction information as multi-dimensional physical features. The leak identification result is determined according to the correspondence between the multi-dimensional physical features and the pipeline network leak feature samples in the operating condition database.

[0038] The early warning and location module is used to determine the early warning level based on the leakage identification results, determine the leaking pipeline section in combination with the pipeline topology, and output early warning and location information.

[0039] Specifically, the anti-interference acquisition module is located at the front end of the water supply network monitoring data. This module receives pressure, flow rate, temperature, and pipeline vibration data. First, it reduces power frequency electromagnetic interference entering the acquisition circuit through a multi-layer electromagnetic shielding structure. Then, based on the current magnetic field and temperature environment at the monitoring point, it calls the specially calibrated magnetic-temperature coupling compensation coefficient. The corrected pressure, flow rate, temperature, and pipeline vibration data form the target network operating parameters and enter the dual-mode transmission and error correction module. The actual magnetic-temperature environment, compensation coefficient version, and actual correction amount form magnetic-temperature correction information and are saved synchronously. For pure water pressure-stabilized water supply networks, the acquisition terminal can also synchronously record conductivity data as auxiliary data for water quality operation status (e.g., recording conductivity at a certain acquisition moment). (The conductivity data is derived from the conductivity detection data of the pure water system. The conductivity data is used for operational status tracing and is not involved in the sample matching of the five-dimensional leakage characteristics described in this embodiment.)

[0040] The acquisition terminal of the anti-interference acquisition module can adopt a corrosion-resistant and explosion-proof aluminum alloy shell, with a multi-layer copper foil composite electromagnetic shielding structure inside the shell. The core circuit can use wide-temperature industrial-grade components, and the surface of the sensing probe can be coated with polytetrafluoroethylene for corrosion and dust resistance. ℃ to ℃); the magnetic induction intensity immunity and calibration range are derived from the strong magnetic interference immunity design and specific calibration range of the monitoring terminal (e.g., according to to (Includes anti-interference design and specific calibration). The above structure is designed to provide physical protection and electromagnetic shielding for pressure, flow, temperature, and pipeline vibration signals before they enter the magnetotemperature correction stage.

[0041] The dual-mode transmission and error correction module receives the target pipeline network operation parameters and selects a spread spectrum penetration transmission node or a dedicated network transmission gateway based on the steel structure obstruction conditions of the monitoring point's area. LoRa spread spectrum penetration transmission is used in steel structure obstruction areas such as electrolysis workshops or rectifier rooms, while industrial 4G dedicated network transmission is used for long-distance open-air pipelines in the plant area. During transmission, the monitoring point identifier, original acquisition time, data sequence number, and valid status are retained. When periodic electromagnetic disturbances occur due to rectification voltage regulation or electrolytic cell start-up and shutdown, the module identifies, corrects, completes, and resumes interrupted transmission data based on the original acquisition timestamp, data sequence number, and terminal cache records, generating error-corrected pipeline network operation parameters. The steady-state timed reporting interval is determined by the communication task configuration and link stability requirements (e.g., configured as...). (The reporting interval, which originates from the communication task configuration of the dual-mode transmission module, is used for communication task scheduling and is not involved in leakage determination.)

[0042] The operating condition linkage processing module receives the error-correcting pipeline network operating parameters and retrieves the production operating condition time-series data for pole switching, anode rise / fall, rectification and pressure regulation, or equipment start / stop within the time period corresponding to the error-correcting pipeline network operating parameters from the operating condition database. The operating condition database includes a pipeline network topology process unit, a production operating condition time-series unit, a magnetic temperature interference correction unit, and a pipeline network leakage characteristic sample unit. The pipeline network topology process unit stores the topological relationships between monitoring points, zones, pipe sections, and water users; the production operating condition time-series unit stores production events and their start and end times; the magnetic temperature interference correction unit stores the magnetic temperature environment, compensation coefficient, actual correction amount, and version; and the pipeline network leakage characteristic sample unit stores leakage type, operating condition label, environmental label, and corresponding characteristic samples. The operating condition linkage processing module determines the current production operating condition and corresponding pressure and flow process disturbances based on the production operating condition time-series data, calls or updates the anomaly judgment threshold corresponding to the current production operating condition, and filters only pseudo-fluctuations corresponding to the time-series and disturbance patterns of the current production operating condition to form standard pipeline network operating parameters.

[0043] The leak identification module receives standard pipeline operating parameters and magnetic temperature correction information. Within the same monitoring point or pipeline section and analysis window, it generates five types of physical characteristics: pressure time-series fluctuation gradient, flow rate difference between day and night, regional production and sales difference, pipeline corrosion vibration noise, and magnetic temperature correction information. These characteristics are then correlated with samples of high-temperature corrosion leaks, thermal expansion leaks, or pressure bursts in the pipeline leak characteristic sample unit. The pipeline leak characteristic sample unit can also retain high-temperature aging valve leakage samples as reserve samples for operation and maintenance analysis. When the reserve samples have a high degree of correspondence, their results can be used as auxiliary operation and maintenance markers or references for verification, without directly changing the three types of leak identification results (high-temperature corrosion leaks, thermal expansion leaks, or pressure bursts) output in this embodiment.

[0044] The early warning and location module uses the leak identification result as the entry point to obtain the corresponding leak amount, pressure fluctuation amplitude, and affected core equipment range. Based on the currently effective grading rules, it determines the early warning level, then associates the leak identification result with pipeline topology process data to determine the pipeline topology segment where the leak occurred. Finally, it links the early warning level, pipeline topology segment, and maintenance prompts to form early warning and location information. This information can be output to the monitoring or maintenance interface of the industrial server. The delivery status returned by the interface and subsequent manual confirmation results can be written back to the server so that the confirmed threshold version and sample version can be used in the next operating cycle.

[0045] Based on the above system, the target pipeline operating parameters formed by the anti-interference acquisition module are restored into continuous error-corrected pipeline operating parameters by the dual-mode transmission and error correction module. The operating condition linkage processing module further utilizes the production operating conditions of the corresponding time period to remove normal pressure and flow disturbances. The leak identification module then uses the standard pipeline operating parameters and the magnetic temperature correction information retained at the source end together for leak sample matching. Finally, the early warning and location module implements the leak identification results into the early warning level and the physical pipeline section, thus forming a continuous operation relationship from acquisition, transmission, operating condition filtering, leak identification to early warning and location.

[0046] Example 2:

[0047] To provide a clearer and more complete explanation of the above embodiments, the present invention also provides an embodiment two. For example... Figure 2 As shown, this embodiment uses the same water supply network monitoring point and the same analysis cycle, so that the data generated in the previous processing stage continues to be used as the input for the next processing stage, thus illustrating the data transfer relationship between the modules.

[0048] In this second embodiment, the anti-interference acquisition module is further used for:

[0049] Pressure sensors, flow sensors, temperature sensors, and vibration detection units are installed at monitoring points in the water supply network, and a data acquisition and protection space is formed by a corrosion-resistant and explosion-proof housing and a multi-layer copper foil composite electromagnetic shielding structure installed inside the housing.

[0050] Acquire pressure, flow rate, temperature, and pipeline vibration data at the monitoring points, and associate the pressure, flow rate, temperature, and pipeline vibration data with the corresponding acquisition time.

[0051] Based on the magnetic field and temperature environment of the monitoring point, the corresponding magnetic-temperature coupling compensation coefficient is used to calibrate the pressure, flow rate, temperature, and pipeline vibration data to obtain the target pipeline network operating parameters.

[0052] The pressure sensor, flow sensor, temperature sensor, and vibration detection unit are located at the same monitoring point in the water supply network, and each physical quantity is associated with a unified monitoring point identifier and the original acquisition time. A corrosion-resistant and explosion-proof housing, together with a multi-layer copper foil composite electromagnetic shielding structure, forms a protective acquisition space. A polytetrafluoroethylene (PTFE) anti-corrosion and dust-proof coating on the sensor probe surface isolates dust and trace corrosive media. Wide-temperature industrial-grade circuitry adapts to continuous high-temperature thermal radiation environments. The magnetic field environment is obtained from the monitoring point environmental record or a specialized calibration process, while the temperature environment is obtained from the current reading of the temperature sensor.

[0053] For example, a monitoring point is located inside a steel structure electrolysis workshop, and the current magnetic induction intensity is recorded as follows: The current temperature is ℃; the lie in to Within the example calibration coverage area, the ℃ is located ℃ to Within the example wide-temperature design range of ℃, both are used together to determine the version of the magnetic temperature coupling compensation coefficient to be invoked subsequently.

[0054] At the same acquisition time, pressure, flow, temperature, and pipeline vibration data are stored in their corresponding units and valid states, and the original acquisition time remains unchanged during subsequent retransmissions, reordering, or breakpoint continuation. If any data is missing, exceeds its range, or has an abnormal time sequence, the record retains its original value and is marked as invalid, and is not directly included in the magnetic temperature correction or leakage characteristic calculation. For example, the pressure obtained at the above acquisition time after shielding at this monitoring point is... At the same time, effective flow, temperature and pipeline vibration records are obtained. All of these records are associated with the same original acquisition time, so that the subsequent production condition sequence can be established according to that time.

[0055] Based on the above-mentioned data acquisition protection and time correlation, the anti-interference data acquisition module continues to execute the magnetic temperature correction relationship obtained from the special calibration.

[0056] In this second embodiment, the anti-interference acquisition module is further used for:

[0057] The magnetothermal coupling compensation coefficients corresponding to different magnetic field and temperature environments were obtained through high-temperature aging, power frequency electromagnetic compatibility and dust corrosion durability calibration.

[0058] The pressure, flow, temperature and pipeline vibration data collected through the multi-layer copper foil composite electromagnetic shielding structure are correlated with the current magnetic field environment and temperature environment, and the magnetic temperature interference deviation that still exists after shielding is corrected according to the corresponding magnetic temperature coupling compensation coefficient.

[0059] The corrected pressure, flow rate, temperature, and pipeline vibration data are output as the target pipeline network operating parameters to the dual-mode transmission and error correction module.

[0060] The magnetotemperature coupling compensation coefficient was obtained through high-temperature aging, power frequency electromagnetic compatibility, and dust corrosion durability calibration, and is saved according to monitoring point, magnetic field range, temperature range, and physical quantity type. When the current magnetic field and temperature are between adjacent valid calibration points, the piecewise interpolation results from the same calibration version can be used; when the current environment exceeds the valid calibration range, the boundless extrapolation results are not used. For any physical quantity among pressure, flow rate, temperature, or pipeline vibration, the residual magnetotemperature deviation after shielding can be corrected using the following formula:

[0061]

[0062] In equation (1), The first sample collected after passing through a multi-layer copper foil composite electromagnetic shielding structure. For a type of physical quantity, its unit is consistent with the corresponding physical quantity; For the corrected operating parameters of the same type of target pipeline network, the units are... Consistent; To monitor the magnetic induction intensity at the monitoring point, the unit is... ; Temperature at the monitoring point, in °C; The actual correction amount with the same dimensions is determined after calling the magnetic temperature coupling compensation coefficient according to the current magnetic field environment and temperature environment. Its source is the special calibration of high temperature aging, power frequency electromagnetic compatibility and dust corrosion durability.

[0063] The specialized calibration table records the correction values ​​for the corresponding channels according to different magnetic field and temperature environments (e.g., when the magnetic induction intensity is...). Temperature is Within the calibration range of ℃, the correction amount for the pressure channel is: ); after shielding the pressure Substituting into equation (1), we obtain the corrected pressure. The actual one used , ℃ The correction amount and the corresponding compensation version together form the magnetic temperature correction information at that acquisition time.

[0064] The corrected pressure, flow rate, temperature, and pipeline vibration data are used to construct the target pipeline network operating parameters according to the original acquisition time, including example pressure values. Upon entering the dual-mode transmission and error correction module, the magnetic temperature correction information is simultaneously written into the magnetic temperature interference correction unit and retained until the leakage identification stage. Therefore, magnetic temperature correction is not only used to change the acquired values, but also to preserve the current environment and actual correction background for subsequent samples.

[0065] Once the target pipeline network operating parameters are generated, they are transmitted by the dual-mode transmission and error correction module based on the conditions of the monitoring area.

[0066] In this second embodiment, the dual-mode transmission and error correction module is further used for:

[0067] LoRa spread spectrum penetration transmission nodes are set up in the steel structure shielded areas of the electrolysis workshop or rectifier room, and industrial 4G private network transmission gateways are set up in the open-air long-distance pipeline network of the plant area.

[0068] Select the corresponding transmission link based on the area where the monitoring point is located to upload the target pipeline operation parameters;

[0069] The system detects periodic electromagnetic disturbances generated by rectifier voltage regulation or electrolytic cell start-up and shutdown. When the periodic electromagnetic disturbances occur, the system performs interference data identification, error correction and completion, and breakpoint continuation on the transmitted data to obtain continuous error-corrected pipeline network operating parameters.

[0070] The deployment locations of the LoRa spread spectrum penetration transmission nodes and industrial 4G private network transmission gateways are determined by the steel structure obstruction conditions in the factory area, the distribution of monitoring points, and the transmission distance. Data is preferentially carried by LoRa spread spectrum penetration transmission nodes in areas obstructed by steel structures, and by industrial 4G private network transmission gateways in open-air long-distance pipeline networks. The steady-state reporting interval of the communication task is derived from the dual-mode transmission task configuration and link stability requirements (e.g., set to...). The steady-state reporting interval is used for transmitting task scheduling and is recorded separately from the original acquisition time of the target pipeline network operating parameters. The above reporting interval is an example configuration of a millisecond-level steady-state timed reporting mechanism.

[0071] When periodic electromagnetic disturbances occur, the dual-mode transmission and error correction module, in conjunction with the rectification and voltage regulation or electrolytic cell start-up and shutdown events, checks the data sequence number, original acquisition timestamp, duplicate records, and missing records. If a data sequence number gap or transmission interruption is detected, the original record in the terminal buffer is read first and retransmitted, restoring the order according to the original acquisition timestamp and sequence number; if the retransmission is not completed, the missing flag is retained, and transmission continues from the breakpoint after the link is restored.

[0072] For example, the aforementioned monitoring point is located in an area obscured by a steel structure. The target pressure is uploaded via LoRa spread spectrum transmission nodes. If a data sequence number gap occurs during a rectification and voltage regulation event, the system retrieves the record from the terminal cache and replenishes it according to the original acquisition time. The replenished record, together with other valid records, forms the error-correcting pipeline operation parameters. If the record is no longer in the cache, subsequent processing retains the missing position and does not use speculative values ​​to create continuous data.

[0073] After the error correction pipeline operation parameters reach the industrial server, the operating condition linkage processing module reads the operating condition database according to the same time base.

[0074] In this second embodiment, the working condition linkage processing module is further used for:

[0075] The operating condition database is divided into pipeline topology process unit, production operating condition timing unit, magnetic temperature interference correction unit and pipeline leakage characteristic sample unit.

[0076] Acquire polarization condition data, anode rise and fall condition data, rectification and pressure regulation data, or equipment start and stop data within the time period corresponding to the error correction pipeline network operating parameters;

[0077] The current production condition is determined based on the production condition time series data, and the abnormal judgment threshold corresponding to the current production condition is called to filter the process disturbances in the pressure and flow changes in the error correction pipeline operation parameters.

[0078] The pipeline topology process unit stores monitoring points, pipeline zones, upstream and downstream pipe sections, water supply objects, and their relationships; the production operation time sequence unit stores polarity switching, anode rise and fall, rectification and pressure regulation, or equipment start and stop events, as well as the start time, end time, and status of the events; the magnetic temperature interference correction unit stores the magnetic induction intensity, temperature, compensation version, and actual correction amount of the monitoring points; the pipeline leakage characteristic sample unit stores multi-dimensional physical characteristic samples of high-temperature corrosion leaks, thermal expansion leaks, and pressure bursts, and can also store high-temperature aging valve leakage samples as auxiliary operation and maintenance samples. The four units are linked through monitoring point or pipeline section identifiers, original acquisition time, production event time, and data version.

[0079] like Figure 3 As shown, the correspondence between production condition time-series data and error-correcting pipeline network operating parameters is based on the original acquisition time. The sampling period is determined by on-site initialization and acquisition equipment configuration according to the dynamic change rate of the pipeline network (e.g., configured as...). (This is derived from the acquisition configuration of high-frequency, second-level data access); the allowable time deviation between production events and pipeline network data is jointly determined by the sampling period and the accuracy of the production condition timestamp (for example, when the sampling period is...). The accuracy of the production condition timestamp is: At that time, the allowable time deviation is taken as ).

[0080] For example, if the error correction recording window of the aforementioned monitoring point overlaps with a polarization event window, and the maximum deviation between the recording time and the event boundary does not exceed... If the condition linkage processing module determines that the record segment is the time period corresponding to the polarization condition, it calls the abnormal judgment threshold version corresponding to the polarization condition; if a corresponding relationship cannot be established, the pressure or flow abnormality of the segment is deleted without using the polarization condition threshold.

[0081] Once the corresponding time period and current production conditions are determined, the condition linkage processing module further updates and applies the anomaly judgment threshold under that condition.

[0082] In this second embodiment, the working condition linkage processing module is further used for:

[0083] The operating status is determined based on the production operating condition timing data corresponding to pole switching, anode raising and lowering, rectification and voltage regulation, or equipment start-up and shutdown.

[0084] The process disturbance changes of pressure and flow under the operating conditions are obtained, and the anomaly detection threshold used for leakage anomaly detection is updated accordingly.

[0085] The updated anomaly detection threshold is used to identify false fluctuations in pressure or flow rate in the error correction pipeline operating parameters that correspond to the current operating conditions, and the false fluctuations are filtered to obtain standard pipeline operating parameters for use by the leak identification module.

[0086] Different anomaly detection thresholds for pressure and flow rate changes are maintained for different production conditions. These thresholds are sourced from historical valid non-leakage data, water supply network design limits, on-site initialization results, and manual confirmation records. During operation, only data confirmed to be leak-free and corresponding to specific production conditions are used to update the threshold version. During updates, newly added confirmed non-leakage data under the same production condition are included in the valid non-leakage data statistics for that condition. The normal process disturbance range for pressure and flow rate changes is redefined, and a corresponding anomaly detection threshold version is generated (e.g., the pressure change threshold for the current pole-switching condition is...). The threshold for flow change is Both are determined by statistical analysis of historical valid non-leakage samples under pole-switching conditions and by combining them with on-site initialization configuration.

[0087] The identification of pseudo-fluctuations satisfies both the time period correspondence and the process disturbance pattern correspondence. For example, pressure occurs within the time period corresponding to the pole switching condition. relative to the pressure reference value for that period change Traffic flow during the same period was Change to ,change The aforementioned pressure changes are lower than... Example pressure change threshold, flow change below The example flow rate change threshold is used, and the time location is consistent with the confirmed polarization process disturbance record. Therefore, the pressure and flow rate change is marked as a pseudo-fluctuation corresponding to the current production condition and filtered out. If the pressure or flow rate change does not conform to the confirmed disturbance pattern of the polarization condition, even if it is within the polarization event period, it is retained as an abnormal record and entered into the leakage identification module.

[0088] After filtering out spurious fluctuations, the standard pipeline operating parameters retain effective pressure, flow rate, temperature, and pipeline vibration records. Following the aforementioned operating cycle, in addition to the filtered process disturbance records, two adjacent effective standard pressures are respectively... and Both continue to serve as inputs for the leak identification module to form the pressure time-series fluctuation gradient.

[0089] like Figure 4 As shown, after the standard pipeline operating parameters are formed, the leak identification module establishes five-dimensional physical characteristics within the same analysis window.

[0090] In this second embodiment, the leakage identification module is further used for:

[0091] The pressure time-series fluctuation gradient, flow rate difference between day and night, and regional production and sales difference are extracted from the standard pipeline network operating parameters, and pipeline corrosion vibration noise and magnetic temperature correction information are obtained.

[0092] The pressure time-series fluctuation gradient, the difference between day and night flow conditions, the difference between regional production and sales, pipeline corrosion vibration noise, and magnetic temperature correction information are used as multi-dimensional physical features for determining leakage in water supply networks.

[0093] The leakage identification result is determined based on the correspondence between the multidimensional physical characteristics and the pipeline leakage characteristic samples in the operating condition database. The leakage identification result includes high-temperature corrosion hidden leakage, thermal expansion seepage leakage, or pressure pipe burst.

[0094] The pressure time-series fluctuation gradient is used to characterize the rate of change of adjacent valid pressure samples after load condition filtering. For two adjacent valid samples, it can be calculated using the following formula:

[0095]

[0096] In equation (2), The current effective standard pressure, in units of ; The previous effective standard pressure, in units of ; The time interval between adjacent valid samples, in units of Its source is the aforementioned sampling period configuration (e.g., ); This represents the pressure time-series fluctuation gradient, in units of... Substituting the previous effective standard pressure of 0.480 MPa and the current effective standard pressure of 0.472 MPa into equation (2), the current pressure time-series fluctuation gradient is obtained as follows: Adjacent records with missing communication flags do not directly participate in the calculation of equation (2) across the missing segment.

[0097] The difference in flow rate between day and night is determined by the average standard flow rate of two traceable production periods within the same statistical period. The day and night statistical window is derived from production shifts, on-site operating rhythms, or manual configuration (e.g., the daytime window is set to...). to Night window set By the next day The two together constitute (Statistical period). Its calculation can be expressed as:

[0098]

[0099] In equation (3), Standard flow rate, unit: ; and These are daytime windows and nighttime windows, respectively; The difference in flow rate between day and night, in units of For example, the daily average standard flow rate within the same statistical period is... The average standard flow rate at night is Both originate from the standard flow statistics records of the same pipeline zone. Substituting them into equation (3) yields the difference in flow between day and night conditions as follows: .

[0100] The regional production and sales difference is based on the regional inflow, regional outflow, and known legal water use or replenishment flow within the same time window. Its calculation can be expressed as:

[0101]

[0102] In equation (4), Inflow within the same time window, in units of ; For the outflow of a partition within the same time window, the unit is ; Known legal water use or replenishment flow rate, in units of ; The difference between production and sales by region, in units of For example, the inflow to a partition within the same time window is... , partition outflow The known legal water use or replenishment flow rate is The three originate from the same zone flow record associated with the pipeline topology process unit. Substituting this into equation (4) yields the zone production and sales difference as follows: When legal water use or replenishment data is unavailable, the difference between inflow and outflow is used as an approximate zonal flow imbalance, not as a direct determination of leakage.

[0103] Pipeline corrosion vibration noise is formed from valid pipeline vibration records within the same analysis window. The analysis window is derived from the sampling configuration, leakage sample statistical window, and on-site initialization results (e.g., the analysis window is 60 s). In this embodiment, the root mean square of the vibration amplitude within this window is used as the vibration characteristic (e.g., the current window is 2.8 mm / s, derived from valid vibration data from the same monitoring point). Magnetic temperature correction information is formed by using the actual applied correction amount and its environmental records within the same analysis window to create traceable characteristics. The sample unit is converted into dimensionless statistical values ​​according to the specifically calibrated effective correction range.

[0104] Specifically, the absolute values ​​of the actual applied corrections for each effective physical quantity channel within the same analysis window are taken, and the maximum absolute correction within the effective range of the corresponding channel's specific calibration is used as the normalization scale to obtain the dimensionless correction ratio for each channel. The arithmetic mean of these dimensionless correction ratios for each effective channel is then taken as the statistical value of the magnetic temperature correction information for that analysis window. The magnetic field environment, temperature environment, and compensation version are used to determine the corresponding effective calibration range and maintain the traceability of the statistical value. The magnetic temperature correction information in the pipeline leakage characteristic samples uses the same processing caliber. For example, in the current example, it is 0.40, dimensionless, derived from the actual correction amount and calibration range stored in the magnetic temperature interference correction unit.

[0105] To enable comparison of the five-dimensional physical characteristics of different units with leakage feature samples, the pipeline network leakage feature sample unit stores a corresponding effective scale for each feature dimension. This effective scale is derived from the effective range of similar leakage samples or the sample dispersion (e.g., the pressure time-series fluctuation gradient scale is...). The standard for the difference in flow rate between day and night is: The regional production and sales difference scale is Pipeline corrosion vibration noise scale is The magnetic temperature correction information scale is (Dimensionless). The corresponding distance between the current five-dimensional physical characteristics and the leakage category samples can be calculated using the following formula:

[0106]

[0107] In equation (5), For the current number 3D physical characteristics; Leakage category The The center or representative value of the dimension sample template, its unit and corresponding Consistent; For the first A dimensionless effective scale, whose unit is consistent with the corresponding characteristic; Leakage category The dimensionless matching distance. The leak category confirmation threshold is determined jointly by leak characteristic sample statistics, specific calibration, and manual configuration, and versions are saved by category (e.g., the confirmation threshold for high-temperature corrosion dark leaks is...). The threshold for confirming thermal expansion and leakage is: The threshold for confirming a pressure pipe rupture is: (These are dimensionless example values). The system selects the leakage category with the smallest matching distance, and confirms the corresponding leakage category only when the minimum matching distance is not higher than the corresponding category confirmation threshold and all key features are valid.

[0108] For example, using the aforementioned pressure time-series fluctuation gradient Flow rate difference between day and night Regional production and sales differences Pipeline corrosion vibration and noise and magnetic temperature correction information statistics The corresponding five-dimensional example values ​​for the high-temperature corrosion dark leak sample center are as follows: , , , and The source is the statistical analysis of confirmed high-temperature corrosion hidden leak samples in the pipeline leakage characteristic sample unit. Substituting the aforementioned five effective scales into equation (5), the example matching distance for high-temperature corrosion hidden leaks is obtained as follows: lower than Example confirmation thresholds; when the matching distance for both thermal expansion leakage and pressure pipe rupture is higher than their respective confirmation thresholds, high-temperature corrosion dark leakage can be used as the current candidate leakage category. If the matching distance for all categories is higher than the corresponding confirmation threshold, or if key features are missing, a pending review status is output without mandatory classification.

[0109] In addition to transparent sample matching, the system can also achieve sample correspondence using a supervised classifier while maintaining the aforementioned five-dimensional physical feature inputs and three types of leakage result outputs. The training data for the supervised classifier comes from labeled leakage samples and non-leakage production condition data confirmed by manual verification; if feature fusion weights are used, the weights are derived from the statistical analysis of labeled samples and the results of manual review (for example, the initial weights for the five features are all 1). (Dimensionless), and subsequent adjustments will only be made based on confirmed samples. This optional implementation does not change the relationship between the magnetic temperature correction information and the five-dimensional physical characteristics, the corresponding time period operating condition filtering, and the continuity of the leaked samples.

[0110] After the five-dimensional physical features are formed and the sample correspondence is completed, the leak identification module further performs combined verification of high-temperature corrosion dark leaks.

[0111] In this second embodiment, the leakage identification module is further used for:

[0112] The pressure time-series fluctuation gradient is determined from the standard pipeline network operating parameters, and the magnetic temperature correction information is obtained. Combined with the pipeline corrosion vibration noise, the pressure change and vibration change of the pipeline under the current magnetic temperature condition are characterized.

[0113] The pressure changes, vibration changes, and regional production and sales differences are matched together with the pipeline leakage characteristic samples;

[0114] When the combined characteristics of pressure change, vibration change and regional production and sales difference match the high-temperature corrosion dark leak sample, the leak identification result is determined to be a high-temperature corrosion dark leak, and it is distinguished from the identification results of thermal expansion leakage or pressure pipe burst.

[0115] Among them, pressure changes directly adopt the aforementioned pressure time-series fluctuation gradient and retain the magnetic temperature correction information of the same analysis window; vibration changes adopt the pipeline corrosion vibration noise of the same window; and the regional production and sales difference adopts the same time window result obtained by equation (4). When the above three types of physical evidence are matched with the high-temperature corrosion leak samples, the system simultaneously checks the matching results of thermal expansion leakage and pressure burst pipe samples to avoid determining high-temperature corrosion leaks based solely on a single pressure drop or a single vibration anomaly.

[0116] For example, the current pressure time-series fluctuation gradient is Pipeline corrosion vibration noise is The regional production and sales difference is Furthermore, the same window contains complete magnetic temperature correction information; the high-temperature corrosion dark leakage matching distance obtained by equation (5) is lower than its If the example confirms the threshold, but neither thermal expansion leakage nor pressure rupture meets its respective confirmation criteria, then the leak identification result can be determined as a high-temperature corrosion hidden leak. If magnetic temperature correction information is missing, regional production and sales differences are unavailable, or both types of leak samples simultaneously meet the confirmation criteria, then the high-temperature corrosion hidden leak will not be confirmed solely based on pipeline vibration changes, and a pending verification status will be output.

[0117] like Figure 5 After the leak identification results are generated, the early warning and location module continues to use the same analysis window and the same pipeline object to determine the early warning level and the leaking pipeline section.

[0118] In this second embodiment, the early warning positioning module is further used for:

[0119] Obtain the leakage amount, pressure fluctuation amplitude, and affected core equipment range corresponding to the leakage identification results;

[0120] The warning level is determined based on the leakage amount, pressure fluctuation range, and the scope of affected core equipment.

[0121] The leak identification results are correlated with the pipeline topology process data in the operating condition database to determine the pipeline topology section where the leak occurred.

[0122] The warning level and the pipeline topology section are combined to generate the warning location information and then output.

[0123] The leakage amount is obtained from the water balance record of the pipeline network corresponding to the leakage identification result. When the inflow, outflow and known legal water use or replenishment data of the zone are complete, the zone production and sales difference obtained by equation (4) can be used as the basis for estimating net water loss. The pressure fluctuation range is determined by the maximum and minimum effective pressure values ​​of the same analysis window. The scope of the affected core equipment is determined by the topological relationship between the candidate leakage section in the pipeline network topology process unit and the water use objects for electrolytic cell cooling, pure water pressure stabilization supply or rectifier heat dissipation. The early warning classification rules are derived from the design limits of the water supply network, the safety management rules of the electrolytic aluminum plant, historical valid events and manual configuration, and are saved according to the rule version.

[0124] For example, the current regional production-to-sales gap is Within the same analysis window, it can serve as the basis for estimating net water loss; the maximum effective pressure within this window is... The minimum value is Both are derived from standard pipeline operating parameters, therefore the pressure fluctuation range is... The pipeline topology process unit displays that the current candidate section affects a cooling water supply branch of an electrolytic cell. The early warning and location module calls the current effective classification rule version to determine the early warning level by comprehensively considering the above leakage amount, pressure fluctuation amplitude, and impact range of core equipment. Then, it determines the candidate leaking pipeline section based on the topological correspondence between monitoring points, pipeline zones, upstream and downstream pipeline sections, and water users.

[0125] After the warning level and pipeline topology section are determined, the warning and location module further associates differentiated operation and maintenance prompts.

[0126] In this second embodiment, the early warning positioning module is further used for:

[0127] Based on the leakage amount, pressure fluctuation range, and the scope of affected core equipment, the leakage identification results are classified into warning levels corresponding to minor corrosion leakage, moderate pipeline leakage, or major pressure pipe burst.

[0128] The corresponding differentiated operation and maintenance push or emergency repair prompt strategy is matched according to the warning level;

[0129] The pipeline topology section is associated with the corresponding early warning level and operation and maintenance prompts and output.

[0130] The terms "minor corrosion leakage," "moderate pipeline leakage," and "major pressure pipe burst" are used to characterize the different urgency levels of the three-tiered early warning system. Their levels are determined comprehensively based on the leakage volume, pressure fluctuation amplitude, and the scope of affected core equipment, and do not replace the high-temperature corrosion seepage, thermal expansion leakage, or pressure pipe burst leakage categories already identified by the leakage identification module. Maintenance prompts for minor corrosion leakage can be continuous monitoring and on-site verification; for moderate pipeline leakage, they can be increased inspection and maintenance; and for major pressure pipe burst, they can be emergency repair prompts. These differentiated maintenance push notifications or emergency repair prompts only apply to monitoring and maintenance personnel and do not send automatic control commands to pumps, valves, electrolytic cells, or rectifier equipment. For example, if the current effective classification rule comprehensively determines the aforementioned 3 m³ / h net water loss estimate, 0.018 MPa pressure fluctuation amplitude, and the impact range of a cooling water supply branch for an electrolytic cell as the early warning level corresponding to a moderate pipeline leakage, then the system matches the moderate-level increased inspection and maintenance prompts and generates early warning location information together with the candidate leaking pipeline topology segment.

[0131] The early warning location information should at least be associated with the leak identification result, early warning level, pipeline topology section, monitoring point or zone, analysis time, sample version, threshold version, and maintenance prompts. The delivery status returned by the maintenance interface is written back to the industrial server; if delivery fails, the original early warning location information is retained and transmission retry continues, without changing the already formed leak identification result and early warning level. Only events that have undergone manual confirmation or other effective confirmation processes are used for subsequent anomaly judgment thresholds or leak feature sample version updates.

[0132] When a common anomaly occurs in the system, each module rolls back according to data validity and traceability requirements. If input data is missing, exceeds the range, timestamp is reversed, or sensor status is abnormal, the original record is retained and marked as invalid. If the current magnetic field or temperature exceeds the effective range of the specific calibration, the applicable range of the magnetic-temperature coupling compensation coefficient is not expanded. If the LoRa or industrial 4G link is interrupted and cached records cannot be retrieved, the missing flag is retained. If the error correction pipeline operating parameters cannot establish a corresponding time period relationship with the production operating condition time sequence, the anomaly judgment threshold corresponding to the current production operating condition is not called to delete pressure or flow anomalies. If the number of leakage samples is insufficient, the matching distance is higher than the confirmation threshold, or multiple categories simultaneously meet the confirmation conditions, a pending verification status is output. If the pipeline topology cannot uniquely determine the leakage section, monitoring point or zone-level location information is output. After recovery, data validity, time period correspondence, and version checks are performed again from the most recent valid original acquisition time before proceeding to subsequent processing.

[0133] For example, if the LoRa link continuously loses records during periodic electromagnetic disturbances and the terminal cache cannot be replenished, the system retains the missing flag in the error correction network operation parameters, and equation (2) does not cross the missing segment to calculate the pressure time-series fluctuation gradient; after the link is restored, processing continues from the most recent valid original acquisition time. As another example, if the error correction network operation parameters of a certain round cannot establish a corresponding time period relationship with the polarization condition, then the data of that round will not use the polarization condition anomaly judgment threshold for filtering, but will enter the five-dimensional physical feature formation and leakage sample correspondence in an unfiltered state.

[0134] During the system engineering deployment process, a working condition survey can be conducted first based on the magnetic induction intensity, workshop temperature, production rhythm and pipeline operation conditions of the target electrolytic aluminum plant. Based on this, the monitoring terminal layout, sensor parameter matching, magnetic temperature compensation coefficient calibration, and special industrial performance tests of hardware such as high temperature aging, power frequency electromagnetic compatibility and dust corrosion durability can be completed.

[0135] Subsequently, based on the steel structure obstruction of the electrolysis workshop and rectifier room, the equipment distribution location, and the distribution of long-distance open-air pipelines, LoRa spread spectrum penetration transmission nodes and industrial 4G private network transmission gateways were set up. The data identification, error correction and completion, and breakpoint resume functions under periodic electromagnetic disturbances were debugged, and network connectivity and data integrity tests were conducted on the predetermined monitoring area.

[0136] Based on the electrolytic aluminum production process, pipeline topology process unit, production condition timing unit, magnetic temperature interference correction unit, and pipeline leakage characteristic sample unit are established to form corresponding data table structure and data cleaning rules. Confirmed production condition timing records, magnetic temperature correction records, and leakage characteristic samples such as electrode switching, anode raising and lowering, rectification and voltage regulation, and equipment start-up and shutdown are entered.

[0137] After completing the above data preparation, the anomaly judgment threshold, leakage sample confirmation threshold and multi-dimensional physical feature effective scale corresponding to each production condition are initialized or debugged. When a supervised classifier is used as an optional implementation, confirmed leakage samples and non-leakage production condition data can be used for training or calibration, and the multi-feature fusion weights are adjusted according to sample statistics and manual review results.

[0138] Finally, the system can be connected to the actual water supply network for on-site commissioning. Long-term trial operation will be conducted under various production conditions, including pole switching, anode raising / lowering, rectification and pressure regulation, and equipment start-up and shutdown. Based on subsequent manual confirmation results, the anti-interference configuration, anomaly detection threshold version, leakage sample version, or optional classifier parameters will be iteratively updated. The above engineering process is intended to establish unified data, time, and version relationships among the modules in the actual electrolytic aluminum plant environment; it does not imply that deployment, training, or trial operation has been completed at a specific customer site.

[0139] Accordingly, this invention uses an anti-interference acquisition module to collect pressure, flow, temperature, and pipeline vibration data in the strong power frequency electromagnetic, high temperature, dust, and trace corrosion environment of an electrolytic aluminum plant. After multi-layer electromagnetic shielding, it uses a specially calibrated magneto-temperature coupling compensation coefficient according to the magnetic field and temperature of the monitoring point to generate target pipeline network operating parameters and magneto-temperature correction information. A dual-mode transmission and error correction module selects a LoRa spread spectrum penetration transmission node or an industrial 4G private network transmission gateway based on the steel structure obstruction conditions, and completes data identification, error correction and completion, and breakpoint resumption under periodic electromagnetic disturbances to generate error-corrected pipeline network operating parameters. A working condition linkage processing module links the error-corrected pipeline network operating parameters with corresponding time-period polarity changes, anode rise / fall, rectification and pressure regulation, or... The system establishes a relationship between equipment start-up and shutdown sequences, calls up or updates anomaly judgment thresholds based on pressure and flow process disturbances under the current production conditions, and filters out corresponding pseudo-fluctuations to form standard pipeline network operating parameters. Furthermore, the leak identification module uses pressure time-series fluctuation gradients, flow day-night operating condition differences, regional production and sales differences, pipeline corrosion vibration noise, and magnetic temperature correction information as multi-dimensional physical features, and establishes a correspondence with pipeline network leak feature samples in the operating condition database. The early warning and location module combines leakage amount, pressure fluctuation amplitude, core equipment influence range, and pipeline network topology to output early warning and location information, thereby forming a continuous monitoring and early warning processing chain of source-end disturbance rejection, production operating condition decoupling, leak sample identification, and physical pipeline network segment location.

[0140] The above description is merely a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart leak monitoring and early warning system for water supply networks adapted to the harsh operating conditions of electrolytic aluminum plants, characterized in that, Includes the following modules: Anti-interference acquisition module: Used to collect pressure, flow, temperature and pipeline vibration data of water supply network. It reduces power frequency electromagnetic interference through multi-layer electromagnetic shielding structure. According to the magnetic field and temperature of the monitoring point, it calls the specially calibrated magnetic temperature coupling compensation coefficient to correct the magnetic temperature interference deviation after shielding and generates target network operation parameters and magnetic temperature correction information. Dual-mode transmission and error correction module: used to select a spread spectrum penetration transmission node or a private network transmission gateway to transmit the target pipeline operation parameters according to the steel structure obstruction conditions; when periodic electromagnetic disturbances are detected, data identification and error correction are performed to generate error-corrected pipeline operation parameters. Operating condition linkage processing module: used to acquire production operating condition time sequence data of pole switching, anode raising and lowering, rectification and pressure regulation or equipment start-up and shutdown within the time period corresponding to the error correction pipeline operating parameters, determine the current production operating condition and pressure and flow process disturbances, call or update the abnormal judgment threshold corresponding to the current production operating condition, filter the corresponding false fluctuations in the error correction pipeline operating parameters, and generate standard pipeline operating parameters. Leakage identification module: used to extract pressure time-series fluctuation gradient, flow day-night operating condition difference, regional production and sales difference and pipeline corrosion vibration noise from the standard pipeline network operating parameters, and use them and the magnetic temperature correction information as multi-dimensional physical features, and determine the leakage identification result according to the correspondence between the multi-dimensional physical features and the pipeline network leakage feature samples in the operating condition database; Early warning and location module: used to determine the early warning level based on the leakage identification results, determine the leaking pipeline section in combination with the pipeline topology, and output early warning and location information.

2. The system according to claim 1, characterized in that, The anti-interference acquisition module is further used for: Pressure sensors, flow sensors, temperature sensors, and vibration detection units are installed at monitoring points in the water supply network, and a data acquisition and protection space is formed by a corrosion-resistant and explosion-proof housing and a multi-layer copper foil composite electromagnetic shielding structure installed inside the housing. Acquire pressure, flow rate, temperature, and pipeline vibration data at the monitoring points, and associate the pressure, flow rate, temperature, and pipeline vibration data with the corresponding acquisition time. Based on the magnetic field and temperature environment of the monitoring point, the corresponding magnetic-temperature coupling compensation coefficient is used to calibrate the pressure, flow rate, temperature, and pipeline vibration data to obtain the target pipeline network operating parameters.

3. The system according to claim 1, characterized in that, The dual-mode transmission and error correction module is further used for: LoRa spread spectrum penetration transmission nodes are set up in the steel structure shielded areas of the electrolysis workshop or rectifier room, and industrial 4G private network transmission gateways are set up in the open-air long-distance pipeline network of the plant area. Select the corresponding transmission link based on the area where the monitoring point is located to upload the target pipeline operation parameters; The system detects periodic electromagnetic disturbances generated by rectifier voltage regulation or electrolytic cell start-up and shutdown. When the periodic electromagnetic disturbances occur, the system performs interference data identification, error correction and completion, and breakpoint continuation on the transmitted data to obtain continuous error-corrected pipeline network operating parameters.

4. The system according to claim 1, characterized in that, The working condition linkage processing module is further used for: The operating condition database is divided into pipeline topology process unit, production operating condition timing unit, magnetic temperature interference correction unit and pipeline leakage characteristic sample unit. Acquire polarization condition data, anode rise and fall condition data, rectification and pressure regulation data, or equipment start and stop data within the time period corresponding to the error correction pipeline network operating parameters; The current production condition is determined based on the production condition time series data, and the abnormal judgment threshold corresponding to the current production condition is called to filter the process disturbances in the pressure and flow changes in the error correction pipeline operation parameters.

5. The system according to claim 2, characterized in that, The anti-interference acquisition module is further used for: The magnetothermal coupling compensation coefficients corresponding to different magnetic field and temperature environments were obtained through high-temperature aging, power frequency electromagnetic compatibility and dust corrosion durability calibration. The pressure, flow, temperature and pipeline vibration data collected through the multi-layer copper foil composite electromagnetic shielding structure are correlated with the current magnetic field environment and temperature environment, and the magnetic temperature interference deviation that still exists after shielding is corrected according to the corresponding magnetic temperature coupling compensation coefficient. The corrected pressure, flow rate, temperature, and pipeline vibration data are output as the target pipeline network operating parameters to the dual-mode transmission and error correction module.

6. The system according to claim 4, characterized in that, The working condition linkage processing module is further used for: The operating status is determined based on the production operating condition timing data corresponding to pole switching, anode raising and lowering, rectification and voltage regulation, or equipment start-up and shutdown. The process disturbance changes of pressure and flow under the operating conditions are obtained, and the anomaly detection threshold used for leakage anomaly detection is updated accordingly. The updated anomaly detection threshold is used to identify false fluctuations in pressure or flow rate in the error correction pipeline operating parameters that correspond to the current operating conditions, and the false fluctuations are filtered to obtain standard pipeline operating parameters for use by the leak identification module.

7. The system according to claim 1, characterized in that, The leakage detection module is further used for: The pressure time-series fluctuation gradient, flow rate difference between day and night, and regional production and sales difference are extracted from the standard pipeline network operating parameters, and pipeline corrosion vibration noise and magnetic temperature correction information are obtained. The pressure time-series fluctuation gradient, the difference between day and night flow conditions, the difference between regional production and sales, pipeline corrosion vibration noise, and magnetic temperature correction information are used as multi-dimensional physical features for determining leakage in water supply networks. The leakage identification result is determined based on the correspondence between the multidimensional physical characteristics and the pipeline leakage characteristic samples in the operating condition database. The leakage identification result includes high-temperature corrosion hidden leakage, thermal expansion seepage leakage, or pressure pipe burst.

8. The system according to claim 7, characterized in that, The leakage detection module is further used for: The pressure time-series fluctuation gradient is determined from the standard pipeline network operating parameters, and the magnetic temperature correction information is obtained. Combined with the pipeline corrosion vibration noise, the pressure change and vibration change of the pipeline under the current magnetic temperature condition are characterized. The pressure changes, vibration changes, and regional production and sales differences are matched together with the pipeline leakage characteristic samples; When the combined characteristics of pressure change, vibration change and regional production and sales difference match the high-temperature corrosion dark leak sample, the leak identification result is determined to be a high-temperature corrosion dark leak, and it is distinguished from the identification results of thermal expansion leakage or pressure pipe burst.

9. The system according to claim 1, characterized in that, The early warning and positioning module is further used for: Obtain the leakage amount, pressure fluctuation amplitude, and affected core equipment range corresponding to the leakage identification results; The warning level is determined based on the leakage amount, pressure fluctuation range, and the scope of affected core equipment. The leak identification results are correlated with the pipeline topology process data in the operating condition database to determine the pipeline topology section where the leak occurred. The warning level and the pipeline topology section are combined to generate the warning location information and then output.

10. The system according to claim 9, characterized in that, The early warning and positioning module is further used for: Based on the leakage amount, pressure fluctuation range, and the scope of affected core equipment, the leakage identification results are classified into warning levels corresponding to minor corrosion leakage, moderate pipeline leakage, or major pressure pipe burst. The corresponding differentiated operation and maintenance push or emergency repair prompt strategy is matched according to the warning level; The pipeline topology section is associated with the corresponding early warning level and operation and maintenance prompts and output.