Pipe network operation and maintenance management system based on data analysis

Abstract

The invention discloses a pipe network operation and maintenance management system based on data analysis, which relates to the technical field of municipal infrastructure operation and maintenance management and comprises an equipment feature modeling module, a group anomaly identification module, a common-mode risk tracing module, a dynamic risk modeling module, a regulation and control strategy optimization module and a closed-loop self-adaptive updating module. Based on historical operation data and real-time acquisition parameters, feature coding is performed on the model, batch, position and time sequence of the sensor, and an equipment attribute mapping matrix is constructed. According to the method, the equipment attribute mapping matrix and the group anomaly identification mechanism are constructed, and external induction factor positioning and the dynamic risk weight map are combined, so that intelligent adjustment of regulation and control parameters and anomaly interference avoidance are realized, a closed-loop optimization process is constructed, and the identification accuracy, the control safety and the operation toughness of a pipe network system in a multi-disturbance scene are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of municipal infrastructure operation and maintenance management, and in particular to a pipe network operation and maintenance management system based on data analysis. Background Art

[0002] The "Data Analysis-Based Pipeline Network Operation and Maintenance Management System" refers to a digital management platform that utilizes multi-source data collection, fusion, and intelligent analysis technologies to comprehensively monitor, evaluate, and provide decision support for the operating status, failure risks, and maintenance needs of urban or industrial pipeline systems (such as water supply and drainage, gas, and heat). This system uses integrated sensors to collect key operating parameters such as pressure, flow, temperature, and water quality in real time. Combining historical operation and maintenance records, Geographic Information System (GIS) data, and environmental factors, the system employs data mining, anomaly detection, trend prediction, and intelligent scheduling algorithms to achieve precise management and optimization of pipeline network operational efficiency, fault warnings, hidden danger location, and maintenance planning. This improves the intelligence level of pipeline network operation and maintenance, reduces the cost of manual intervention, and ensures the safe, efficient, and sustainable operation of public infrastructure.

[0003] The existing technology has the following deficiencies: Existing data-analysis-based pipeline network operation and maintenance management systems typically adopt a single-point anomaly tolerance strategy. This means that when an operating parameter collected by a sensor experiences a sudden change or deviates from the expected threshold, the system typically treats it as an individual device failure, short-term interference, or occasional fluctuation, thereby avoiding malfunctions caused by misjudgments. However, in actual applications, a large number of sensor devices deployed in the same subsystem or geographic area often have the same model, production batch, or hardware architecture, resulting in a high degree of homology. When such sensors encounter drastic changes in the electromagnetic environment (such as lightning strikes, electrical equipment startup and shutdown), firmware defects (such as logical errors and temperature drift failures), or when attackers implement batch control operations through specific protocol interfaces, group synchronization offsets or consistency distortions are very likely to occur.

[0004] Because existing systems lack a mechanism for identifying consistent anomalies across common device characteristics, in these situations, the system still treats these fluctuations as "single-point fluctuations from multiple independent devices," failing to identify the underlying systemic common-mode risk sources. This can lead to erroneous operational status judgments and scheduling decisions. For example, if multiple pressure sensors simultaneously report inflated data due to common-mode interference, the system may mistakenly believe that the pipeline section is under-pressure and initiate pressure-replenishing measures, further exacerbating the risk of overpressure and potentially causing equipment damage, energy leaks, or even pipe bursts, with extremely serious consequences.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a pipeline network operation and maintenance management system based on data analysis. By constructing an equipment attribute mapping matrix and a group anomaly recognition mechanism, combined with the positioning of external inducing factors and a dynamic risk weight map, it can realize intelligent adjustment of control parameters and avoidance of abnormal interference, build a closed-loop optimization process, and improve the recognition accuracy, control safety and operation resilience of the pipeline network system in multiple disturbance scenarios, so as to solve the problems in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solutions: a pipeline network operation and maintenance management system based on data analysis, including an equipment feature modeling module, a group anomaly identification module, a common mode risk tracing module, a dynamic risk modeling module, a control strategy optimization module, and a closed-loop adaptive update module: The device feature modeling module encodes the sensor model, batch, location, and timing based on historical operation data and real-time acquisition parameters, and constructs a device attribute mapping matrix; The group anomaly recognition module uses the device attribute mapping matrix to identify homologous sensors, extract their operating parameters per unit time, calculate the cross-point difference factor and time series synchronization factor, and generate the group anomaly preliminary judgment vector; The common mode risk tracing module collects electromagnetic interference, firmware upgrade, and remote access records based on the group anomaly initial judgment vector, analyzes their temporal relationship with anomaly offsets, identifies common mode risk inducing factors, and determines their corresponding inducing mechanisms; The dynamic risk modeling module reconstructs the parameter weights within the impact area based on the influence range of common risk-inducing factors, and generates a dynamic risk weight map based on the operating parameter trends to represent the priority relationship between control objectives and resources; The control strategy optimization module adjusts valve openings, pump start and stop sequences, and alarm thresholds within the target area based on a dynamic risk weight map to mitigate interference from common-mode offset sensors and improve control accuracy and safety. The closed-loop adaptive update module writes the control process data into the equipment attribute mapping matrix, updates the group anomaly initial judgment vector and dynamic risk weight map, and realizes closed-loop optimization of anomaly identification and control parameters.

[0008] Preferably, constructing the device attribute mapping matrix includes the following steps: Collect operating parameter records, operation and maintenance records, geographic information, and environmental information covering the sensor's operating cycle to form a unified data view; Feature encoding of sensor hardware information, manufacturing information, deployment environment information, and runtime information based on data views; Convert all feature codes into structured fields and embed them into the device attribute mapping matrix in the form of a two-dimensional array; The mapping matrix is ​​filled with missing values, logical conflicts are eliminated, and feature validity is verified to complete the construction of the structured data foundation.

[0009] Preferably, generating a group abnormality preliminary judgment vector includes the following steps: Sensors with the same model, production batch, and deployment area are selected to form a homologous device set; Extract the operating parameter data of the device set within a unified time window, perform time alignment and missing data completion; Calculate the cross-point data difference factor and timing synchronization factor between devices in the set to obtain behavioral consistency analysis results; A group abnormality preliminary judgment vector representing the abnormal state of synchronization deviation is generated based on the set threshold.

[0010] Preferably, identifying common mode risk inducing factors and determining their corresponding inducing mechanisms includes the following steps: Based on the time period marked by the group anomaly initial judgment vector, the corresponding electromagnetic interference records, firmware upgrade information and remote access logs are extracted; Compare the occurrence time of electromagnetic interference records, firmware upgrade information, and remote access logs with the time sequence of group offset start time; Unify the timelines of the three types of events, construct a time series correlation map, and analyze the impact and credibility of each event on the abnormal deviation; Based on the degree of event overlap and impact, determine the common risk factors that cause group deviation.

[0011] Preferably, generating a dynamic risk weight map comprises the following steps: Based on the spatial influence range of common mode risk-inducing factors, calibrate the set of controlled equipment within the risk intervention area; Extract the pressure, flow, and temperature parameters of the devices in the set within a fixed period before and after the common mode event, and calculate the operational response strength; Calculate the comprehensive risk weight value of each device based on the control criticality of the device in the pipeline network; A dynamic risk weight map is constructed based on the comprehensive risk weight values ​​of all equipment and their deployment locations, and the risk gradient channels are marked with weight differences.

[0012] Preferably, adjusting the valve opening, pump start and stop sequence, and alarm threshold within the target area includes the following steps: Extract nodes with risk weight scores higher than 75 points in the dynamic risk weight diagram and calibrate the control intervention equipment list; Adjust the opening setting value of the electric valve according to the risk level of the control point and execute it step by step according to the preset range; Adjust the start and stop sequence and operating speed parameters of the pump group according to the risk level; Temporarily modify the threshold parameters of the alarm model and add homologous device comparison rules to control the alarm triggering conditions.

[0013] Preferably, after writing the operating data during the control execution process into the device attribute mapping matrix, the following steps are included: Write pressure value, flow rate, temperature and control device response status data to the corresponding fields; Extract the operating parameter data of the same-source sensor and calculate the cross-point data difference factor and time series synchronization factor; Generate an updated group anomaly initial judgment vector based on the cross-point data difference factor and the time series synchronization factor; Refresh the dynamic risk weight map based on the updated preliminary judgment vector and adjust the risk level and response priority of each node.

[0014] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention achieves accurate identification of homologous device groups by constructing a device attribute mapping matrix and systematically integrating multi-dimensional attributes such as sensor model, batch, deployment location, and operating sequence; then, through the analysis of difference factors and synchronization factors, a preliminary judgment mechanism for group abnormal deviation is established for the first time, and combined with external induced events such as electromagnetic interference, firmware updates, and remote access behaviors, the location and mechanism identification of common mode risk sources are achieved. On this basis, a quantitative mapping of operating status to control strategy is achieved through a dynamic risk weight map, enabling the system to dynamically adjust valves, pump groups, and alarm thresholds to avoid interference from false data and ensure the accuracy and safety of control. Finally, through the continuous writing back of operating data and iterative updating of models, a data-driven closed-loop optimization mechanism is constructed to achieve the system's continuous adaptability and self-evolution capabilities to complex abnormal situations. This solution not only improves the accuracy and efficiency of abnormal responses, but also significantly enhances the resilience and stability of the pipeline network system in a multi-disturbance environment, and has a wide range of practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0016] Figure 1 This is a module diagram of the pipe network operation and maintenance management system based on data analysis of the present invention. DETAILED DESCRIPTION

[0017] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0018] The present invention provides Figure 1 The data analysis-based pipeline network operation and maintenance management system shown in the figure includes an equipment feature modeling module, a group anomaly identification module, a common mode risk tracing module, a dynamic risk modeling module, a control strategy optimization module, and a closed-loop adaptive update module: The device feature modeling module, based on full historical operating data and real-time collected parameters, performs multi-dimensional feature encoding on the sensor device model, production batch, installation location, and operating sequence, generating a device attribute mapping matrix to provide structured basic data for abnormal consistency analysis. Based on in-depth processing of the full historical operating data and real-time collected parameters of the sensor devices deployed in the pipe network system, a data infrastructure is constructed to support abnormal consistency identification. Specifically, it is a set of device attribute mapping matrices containing multiple dimensional fields. This process includes the following steps: Raw data and associated information covering the entire pipeline network operation cycle are collected to form a raw dataset for device attribute encoding. This dataset includes operational parameter records for various sensors from installation to the current time period. Specifically, this includes pressure curves for pressure sensors, flow rate records for flow sensors, temperature response logs for temperature sensors, and dissolved oxygen, conductivity, and turbidity values ​​reported by water quality monitoring equipment. It also includes complete operational and maintenance records for the equipment throughout its lifecycle, including repair times, parts replacement details, downtime duration, commissioning cycles, the specific time of each failure, and the handling methods of maintenance personnel. Furthermore, geographic information must be extracted, including the exact coordinates of the sensor device within the urban pipeline network, the pipe segment number, upstream and downstream pipe diameters, the water or heating zone identification, and influencing environmental factors such as terrain elevation, surrounding building density, and manhole cover closure. After cleaning, timestamp alignment, and anomaly removal, this data is indexed into a complete structure. Using the sensor's unique identification code as the primary key, a unified multi-source data view is constructed.

[0019] Based on the above data view, each sensor device is characterized in four areas: hardware information, manufacturing information, deployment environment information, and runtime information. Hardware information includes the product model, rated range, sampling accuracy, signal output method (e.g., 4–20mA analog, Modbus digital communication, NB-IoT wireless transmission), operating temperature range, and power input specifications. Manufacturing information includes the manufacturer name, production plant number, product batch number, production date, and product certificate number. Deployment environment information includes the material type of the pipeline in which the device is located (e.g., ductile iron, polyethylene, steel), burial depth, well location, connected valve or pump group number, pipeline installation year, historical water hammer frequency, and the distribution of surrounding high-voltage equipment. Runtime information includes the number of device starts and stops, daily valid data upload rate, frequency of abnormal data segments, response delay to data changes with adjacent devices, and the maximum daily data variance over the past month. All of these features are converted into structured fields, numbered and categorized according to a unified field encoding standard, and a field dictionary is established to ensure consistency across different devices.

[0020] The aforementioned features are embedded in a structured form within a device attribute mapping matrix. This mapping matrix is ​​constructed as a two-dimensional array, with each row corresponding to a unique sensor device and each column corresponding to a specific coded field. The matrix has 48 fields, including 12 fields for hardware specifications, 6 fields for manufacturing information, 18 fields for deployment environment, and 12 fields for runtime. Each field clearly defines its unit, data type, and value range. For example, the "well number" field in the deployment environment is a string field whose value is a city number plus a geographic coordinate code; the "signal output mode" field is a categorical field with a value range limited to analog, voltage-type digital, serial communication, or wireless communication; and the "average daily data upload rate" field is a floating-point field with a value range of 0 to 1. All numeric fields are unit-normalized, categorical fields are one-hot encoded, and Boolean fields are binary encoded. The generated matrix is ​​exported to the data repository in CSV format and used through the data interface for subsequent querying, feature filtering, and device grouping.

[0021] The constructed device attribute mapping matrix undergoes multiple layers of validation to ensure data consistency and integrity. First, each field is checked for missing values. Null values ​​in Boolean fields are zero-filled, and missing values ​​in numeric fields are interpolated using the historical mean. Second, fields are screened for logical conflicts. For example, a sensor should not have a "disabled" status in the "Operational Status" field while also having sampling records from the past week. Third, statistical analysis is performed on each field value to identify outliers. For example, devices with a "Daily Abnormal Segment" exceeding a reasonable threshold are marked as "Feature Deviation Abnormal." Finally, the mapping matrix is ​​validated for usability. Three representative deployment areas are selected, and attribute vectors of all sensor devices are extracted. Their feature similarity distribution curves are compared to confirm the matrix's ability to distinguish between homologous and non-homogeneous devices in real-world data. Once constructed, the device attribute mapping matrix allows for field-level incremental updates when device status changes (such as model changes, redeployment, or communication method changes). This ensures the mapping matrix's continued validity throughout the pipeline network lifecycle and provides a highly accurate and interpretable structured data foundation for subsequent population anomaly identification.

[0022] The core function of this step is to provide a unified, complete, and comparable structured data foundation for subsequent anomaly consistency identification. By encoding key sensor device attributes such as model, production batch, installation location, and operating sequence in multiple dimensions, the system transforms dispersed, heterogeneous, and unstructured data into a device attribute matrix with semantic associations and logical structure. This matrix not only encodes and abstracts the static properties and dynamic behavior of each sensor device but also provides technical support for group feature comparisons between homologous devices. In practical applications, sensors of the same model or production batch often have similar response characteristics and potential common defects, while operating environment and time series information determine their sensitivity to external perturbations or systemic risks. Therefore, by standardizing and modeling this information, this step enables the system to identify structural homology relationships between devices in large-scale sensor data and conduct synchronized group behavior analysis based on this information. If the system subsequently detects abnormal excursions on multiple devices at the same time, it can quickly query their attribution in the device attribute mapping matrix to determine whether they share common attributes or deployment correlations, thereby avoiding misinterpreting group system failures as individual fluctuations of multiple independent devices. Therefore, the construction of this structured matrix is ​​not only the basis of data management, but also a key prerequisite for ensuring the logical rigor of the recognition algorithm and the credibility of the judgment results.

[0023] The group anomaly recognition module, supported by the device attribute mapping matrix, identifies homologous sensors of the same model, production batch, and deployment area, obtains their operating parameter data per unit time, calculates the cross-point data difference factor and time series synchronization factor, and generates a group anomaly initial judgment vector to identify highly consistent abnormal deviation behaviors of multiple homologous sensors within the same time period. Relying on the constructed device attribute mapping matrix, we carry out the identification of abnormal sensor group offset behavior. This identification process focuses on the behavior comparison of homologous devices and includes device screening, parameter extraction, behavior consistency analysis, and the generation of initial abnormality judgment vectors. It clearly realizes the determination of whether there is a group synchronous offset of multiple sensors in the same time period. The specific steps are as follows: All registered sensor devices are conditionally screened using the model field, production batch field, and deployment area field recorded in the device attribute mapping matrix. The specific screening criteria are: the model fields are completely consistent, for example, all are "PS-2020B" pressure sensors; the production batch field is the same production serial number, such as "B0421"; the geographical location corresponding to the deployment area field is continuous, specifically, the urban pipe network number is continuous or the geographical coordinates fall within the same subnet range, for example, all devices are located in pressure pipes numbered "Z001" to "Z008". Through the combined conditions of the three types of fields, the qualified devices are divided into a set of homologous devices with structural consistency and similar operating environments. The devices within this set are highly consistent in construction, manufacturing, and deployment, providing a basis for comparing group operating behaviors.

[0024] For all devices in the same source device set, extract the complete data segment within a unit time window from their corresponding operating data. The time window can be a fixed duration, such as 30 consecutive minutes, and the specific value is set according to the frequency of change of the target parameter. The extracted data parameters include but are not limited to: the output pressure value of the pressure sensor (in MPa), the measured temperature value of the temperature sensor (in degrees Celsius), and the instantaneous flow rate of the flow sensor (in cubic meters per hour). During the extraction process, unified alignment is performed through timestamps to ensure that the start and end time of the data sequence of each device is consistent, and the interval spacing is unified to one record per minute. If data is found to be missing at certain time points, such as the sensor does not upload records in a certain minute, the valid data of the device in the five minutes before and after is referred to, and the gap is filled by linear interpolation, thereby ensuring the data integrity of the subsequent processing links.

[0025] After data preparation is completed, the synchronization and difference analysis of the operating parameters between each device in the homologous device set is performed. For difference analysis, the numerical difference of the parameter values ​​between the two devices at the same time point is calculated in units of one minute, the absolute value is recorded, and the average difference over the entire time period is accumulated as a measure of the degree of offset between devices. For synchronization analysis, observe whether the parameter fluctuation direction of each device at multiple consecutive time points is consistent. For example, if four of the five devices show a continuous upward trend within three consecutive time points, it is recorded as a "synchronous rising behavior"; this statistical process is accumulated multiple times within the entire time window, and the proportion of synchronous change behavior is finally calculated. The above two analysis results together form a description of the response behavior of the device set within the specified time window, reflecting whether the overall state of change is coordinated and synchronized.

[0026] The analysis results are converted into a preliminary judgment vector for identifying abnormal group deviations. A judgment threshold is set. For example, if more than 80% of the devices in the device set exhibit a synchronization rate exceeding 70% within the time window, and their average parameter differences significantly deviate from the normal range of the historical data for the set (e.g., exceeding the set ±15%), the time window is marked as a "suspicious group deviation period." Accordingly, a preliminary judgment vector is generated for this time period, with a vector value of "1" indicating suspected synchronization deviation and "0" indicating no abnormality. This preliminary judgment vector, indexed by the time period and bound to the device set identifier, is used for subsequent process calls to identify potential common mode induction mechanisms and can serve as a precursor to triggering regulatory response mechanisms.

[0027] The above steps effectively analyze the consistent operating behavior of sensors of the same model and manufacturing source deployed in the same area over the same time period, thereby determining whether there are synchronization deviation anomalies that are not individual but rather have a systematic background. This approach effectively breaks through the traditional approach of isolated treatment of single-point sensor anomalies, improving the accuracy of multi-device joint behavior recognition and the forward-looking nature of the response, providing the technical prerequisites for subsequent treatment of common-mode interference, electromagnetic interference, or human manipulation risks.

[0028] This step analyzes the operational behavior of groups of sensor devices with the same model and production batch, deployed in adjacent areas, to identify highly consistent abnormal excursions within a specific time period, effectively identifying potential systemic risks or common-mode interference issues. Traditional pipe network operation and maintenance systems typically use single-point anomaly detection logic. When a sensor's data fluctuates beyond a threshold, the system considers it an isolated device failure or a sporadic anomaly, lacking the ability to analyze behavioral consistency across multiple devices. However, in practical applications, a large number of sensors often originate from the same production line, share the same hardware architecture, communication protocol, and internal circuit design, and are concentrated in a specific area. These sensors are prone to collective synchronization distortion when exposed to external electromagnetic interference, batch configuration errors, or remote manipulation. Therefore, this step filters the key attributes recorded in the device attribute mapping matrix to identify a set of devices with structural homology. The system then extracts their operational data per unit time. Parameter differences and trend consistency indicators are calculated across devices at a granularity of minutes or seconds to determine whether similar excursions occur simultaneously across a large area and across multiple devices. The resulting initial group anomaly vector can not only serve as input for subsequent common-mode risk analysis, but also as a warning signal in the real-time monitoring system, reminding dispatchers that the current anomaly may not be an isolated incident, but rather a potential wide-area collaborative distortion phenomenon, thereby avoiding erroneous control decisions and resource misallocation, and improving the system's perception capabilities and the accuracy of operation and maintenance responses.

[0029] The common mode risk tracing module collects electromagnetic interference records, firmware upgrade information, and remote access behavior logs based on the time period corresponding to the group anomaly initial judgment vector. It analyzes the temporal correlation between the anomaly offset starting point and the above external events to determine whether there are common mode risk inducing factors and identify their corresponding inducing mechanisms. To accurately identify the potential common-mode triggering factors that cause highly consistent abnormal deviations in multiple sensor devices within the same time period, it is necessary to retrieve relevant external environmental data, device software status change information, and network access behavior records based on the time period marked by the group abnormality initial judgment vector generated above, and comprehensively analyze the temporal relationship between the abnormal starting point and the external factors, so as to determine whether there is a traceable common-mode risk event and further clarify its triggering mechanism. This process includes the following steps: Based on the time period marked as "synchronous anomaly present" in the group anomaly initial judgment vector, the start and end times of that time period and the device set identifier are extracted. Specifically, within the time window where the initial judgment vector value is "1," the start and end minutes are determined and the corresponding device attributes, including sensor number, deployment area number, product model, and batch number, are located. Subsequently, based on the deployment area number, all electrical interference logs within that area are retrieved. These logs are sourced from interference monitoring equipment deployed in substations, distribution cabinets, and key power control points. Data fields in these logs include the time of interference occurrence, interference type (e.g., surge, voltage sag, electromagnetic pulse), interference intensity level (in volts / meter or amperes / meter), and interference range. Each event in the log is timestamped with an accuracy of at least one second, along with the location of the interference source and the impact radius. By comparing the time of the interference event with the start time of the group offset, it is possible to determine whether there is any overlap or premature occurrence of high-intensity electromagnetic interference.

[0030] Within the same time period, all firmware upgrade or configuration modification events are retrieved from the sensor device maintenance record platform. This type of information is regularly archived and saved by the device management platform. Each record contains the device's unique number, upgrade operator identification, operation time, pre-upgrade version number, post-upgrade version number, upgrade content summary, and upgrade completion status. Special attention should be paid to records involving measurement accuracy parameter adjustments, signal processing logic optimization, or communication protocol version changes in the upgrade content. Compare the time the upgrade operation occurs with the abnormal offset time. If the two are within the same time window, or the upgrade operation occurs no more than five minutes before the abnormality, it is considered a potential cause of internal configuration changes. In particular, when multiple devices complete firmware upgrades in a short period of time and data synchronization offsets occur immediately after the upgrade, it is necessary to further correlate the upgrade content with changes in device response characteristics to determine whether the abnormality is caused by program logic mismatch or parameter initialization error.

[0031] Obtain the remote access behavior logs involving the above-mentioned sensor devices within the same time period and analyze the legitimacy of the access behavior. The log contains the remote access initiating IP address, access timestamp, operation type (such as remote data reading, parameter distribution, remote restart), access account ID, device response status and log verification value. During the screening process, focus on the following two situations: First, whether there are a large number of centralized access requests initiated from abnormal IP addresses; second, whether there are operations that perform batch configuration changes through privilege escalation. If it is found that such behavior occurs before the group abnormal offset time point, and the access objects are concentrated at the set of devices where the abnormality occurs, it can be preliminarily determined that there may be abnormal manipulation or configuration abuse. In addition, for records in the access log with the operation type of "remote parameter batch distribution", it is necessary to further check whether the operation is successfully executed and whether the distribution content involves sensor parameter changes, such as threshold settings, filter period adjustments and other key control fields.

[0032] By combining data from these three sources, time series matching and causal analysis were performed. The timelines of electromagnetic interference records, firmware upgrade information, and remote access logs were unified, creating a temporal correlation map anchored by "group offset time." This map identifies electromagnetic events, upgrades, and remote access behaviors as event nodes, with their occurrence time, impact targets, and impact level indicated. The events are arranged chronologically, and the time difference between each event and the anomaly's origin is plotted. By analyzing the temporal sequence of these events, the degree of overlap in impact areas, and the number of affected devices, we determine whether there is a clear triggering event. For example, if an electromagnetic interference record occurs within two minutes of the anomaly, the interference radius covers the locations of all anomalous devices, the interference intensity exceeds the warning level, and thereafter, device parameters exhibit synchronized offsets, then the interference event is considered a high-probability common-mode risk source. Furthermore, if a firmware upgrade occurs during the same period but only affects a small number of devices, or if the remote access logs contain no suspicious activity, then electromagnetic interference is considered the primary cause. After the analysis is completed, the type, time, scope of influence and credibility of the common mode inducing factors will be recorded to provide a clear basis for the subsequent risk response and regulatory measures.

[0033] This step, after identifying multiple sensors with highly consistent abnormal excursions within the same time period, further analyzes whether there are external triggers that may have caused the group anomaly. This prevents misidentification of systemic risks as individual device failures, improving the accuracy of anomaly identification and the reliability of operational decisions. Specifically, this step locates the time period marked in the initial judgment vector for the group anomaly and extracts three types of key data covering that period: electromagnetic interference records, such as strong electromagnetic field changes caused by lightning strikes, sudden starts and stops of electrical equipment, and transformer operations; sensor firmware upgrade information, including version updates, parameter configuration changes, program logic reloads, and other operations that may affect device measurement or communication; and remote access behavior logs, including remote reading and writing, parameter distribution, and start / stop control operations performed by external personnel or systems on sensors via network interfaces. By aligning these data along a timeline and accurately comparing them with the starting point of the group anomaly excursion, it is determined whether the three types of events occurred within a short period before the excursion, whether they cover all devices experiencing the anomaly, and whether they have the potential to affect their data measurement, signal transmission, or logic execution, thereby determining whether there are common-mode risk triggers. If an electromagnetic interference event or a firmware upgrade operation is found to closely coincide with a large-scale device deviation, and occurs close in time to the point where the deviation occurred, this event can be inferred as a triggering mechanism. This analysis process plays a key role in transitioning from "result identification" to "cause tracing." It not only improves the ability to control the source of the anomaly, but also provides a basis for subsequent control optimization, preventing ineffective or excessive responses triggered by misidentification, and ensuring the stability and security of the pipeline network system.

[0034] The dynamic risk modeling module reconstructs the weights of operating parameters within the affected area based on the impact range of common-mode risk-inducing factors. It combines the changing trends of multi-point pressure, flow, and temperature to generate a dynamic risk weight map to represent the priority of control objectives and the distribution relationship of control resources between various operating nodes. To respond to collective sensor anomalies caused by common-mode risk triggers, it is necessary to conduct real-time assessments of key operating parameters within the affected area and construct a dynamic risk weight map with structural logic from the perspective of risk propagation to provide a basis for the formulation of regulatory strategies. This process involves risk impact identification, operational data assessment, risk weight quantification, and map construction. The specific steps are as follows: Based on the identified common mode risk inducing factors, clarify their impact areas and object ranges in the spatial dimension. Specific operations include extracting the impact radius of the interference event, the location of the electromagnetic source, the time period of action, and the impact intensity level. If it is an electromagnetic interference event, obtain the coordinate point where the interference occurs and its affected radius, usually in meters, and use the GIS geographic information system to determine whether it covers a specific water supply trunk or thermal return water trunk. For firmware upgrade risk inducements, mark the corresponding physical deployment points and device channels based on the device numbers listed in the upgrade log and the device attribute mapping matrix. Remote access operations are located based on the target device numbers listed in the access control record. In this way, a clear set of controlled devices and a physical deployment range are formed, marked as a "risk intervention area."

[0035] For all sensor devices within the aforementioned "risk intervention zone," fixed sampling periods are set before and after a common-mode event (e.g., 30 minutes before and after the event). The instantaneous pressure value (in MPa) of the pressure sensor, the flow rate (in cubic meters per hour) of the flow sensor, and the medium temperature (in degrees Celsius) of the temperature sensor are collected. The time series of these parameters are organized by minute, and the average, maximum, and magnitude of change for each device in the two periods before and after the event are quantitatively analyzed and compared with the device's historical statistical values ​​over the previous 72 hours. For example, if a pressure sensor's pressure rise exceeds 1.5 times its 72-hour average fluctuation within 5 minutes of the event, it is marked as a "significant change device." This change flag is used as a risk response factor for the device's current status, forming a "controlled area operation response list" that lists each device's number, parameter type, magnitude of change, response intensity level (scored on a scale of 0-100), and a Boolean flag indicating whether it has crossed a set safety threshold.

[0036] A comprehensive risk weighting assessment is performed based on the status of each device in the "Operational Response List" and its position and role in the pipeline network. Specifically, this assessment includes two factors: First, the operational response intensity, which measures the actual degree of fluctuation for each parameter. For example, a sudden increase in pressure of 1.2 MPa, a sudden decrease in flow rate of 40 cubic meters per hour, or a temperature drop of 10 degrees Celsius is assigned a corresponding risk score. Second, the device's control criticality, which measures whether the device is located in the main water supply pipeline, directly connected to a primary pumping station or master control valve, or provides services to multiple end users downstream. These two pieces of information are combined to form a comprehensive weighting consisting of the "fluctuation intensity score + control importance score." For example, if a node's operational response intensity score is 85 and its control criticality score is 90, its final node risk weight is (85 + 90) / 2 = 87.5. This weighting represents the risk level and regulatory intervention priority for the device under the impact of this common mode risk.

[0037] A complete dynamic risk weight map is constructed based on all nodes with defined weights. This map uses physical deployment points as nodes, with each node labeled with a device number, geographic coordinates, operating parameter type, current parameter value, comprehensive risk weight, and status. Edges in the map represent physical connections in pipelines, such as the pipe path from a main water supply valve to a downstream distribution valve, with direction markers representing fluid flow. If the difference in risk weight between two nodes exceeds a set threshold (e.g., 20 points), the edge is labeled a "risk gradient channel" to identify the risk propagation path. High-weight nodes in the graph are prioritized by control strategies. Valve opening control, pump start-up sequencing, and alarm threshold adjustments in their areas are prioritized based on this graph structure. The map is also refreshed every five minutes to ensure real-time reflection of operational status changes during continuous events.

[0038] After identifying common-mode risk triggers, this step further quantitatively assesses each operating node within their impact area and constructs a structured "dynamic risk weight map" to clarify each node's control priority and resource allocation relationship under the current abnormal scenario. Because pipeline networks are composed of numerous sensors, valves, pumping stations, and control devices, each node exhibits significant differences in structural location, functional attributes, operating status, and risk exposure. Consequently, the response of each node to anomalies such as electromagnetic interference, firmware upgrade anomalies, or remote control anomalies can vary. Without dynamic identification and hierarchical classification of these response differences, the system may experience issues such as erroneous control instructions, unbalanced resource allocation, or incorrect risk transfer paths. This step calculates the response strength of each node within the affected area, combining operational parameters such as pressure change, flow rate fluctuation, and temperature excursion trends before and after the anomaly. Weighted scores are then assigned based on the node's criticality within the water or heating system, such as whether it is a backbone node or whether it connects to multiple user terminals. All nodes are then constructed into a network structure diagram with flow relationships and weight distribution. In this diagram, nodes with higher weights are prioritized for regulation and intervention, and resource allocation should be tilted more towards these high-risk nodes. The resulting dynamic risk weight map not only improves the accuracy of the system's response to anomalies but also provides real-time updates and trend tracking capabilities. This can be used to guide key control actions such as valve opening adjustment, pump start-up and shutdown sequence optimization, and alarm mechanism sensitivity setting. It is an important foundation for achieving intelligent scheduling and closed-loop risk management.

[0039] The control strategy optimization module adjusts the valve opening, pump start and stop sequence, and alarm thresholds of the analysis model within the target area based on the results of the dynamic risk weight map to avoid data interference from common mode offset sensors and ensure the accuracy and safety of system control behavior; Based on the high-risk nodes and propagation paths identified in the dynamic risk weight map, parameter adjustments are carried out around the operating control points in the target area. Specifically, optimization is performed on valve opening control, pump start and stop timing and operating rate management, and threshold setting strategies in the system alarm model. These adjustments aim to eliminate interference caused by common-mode offset sensors on pipeline network operation and ensure the stability of control logic and the accuracy of response behavior. The entire process includes the following steps: Extract all node information with risk weight scores higher than 75 points in the dynamic risk weight map, and identify the corresponding control equipment number, pipe section, upstream and downstream connection relationship, and equipment type (such as regulating valve, booster pump, pressure measuring point). Based on the map, the system determines the specific equipment list that requires control intervention and sets the action priority for each node. Among them, nodes with a score of 90 points or above are marked as first-level control points and are given priority for adjustment operations; nodes with a score between 75 and 89 points are regarded as second-level control points and are coordinated with the response of the first-level nodes. For example, the main water supply pipe section node numbered P-006 is located at the center of common mode offset influence, and its corresponding electric butterfly valve and outlet variable frequency pump are determined to be first-level control points and enter the state to be adjusted.

[0040] For electric valves at primary and secondary control points, the opening parameters are reconfigured based on the node's location and the water supply stability requirements of the downstream area. If the current opening is fully open (100%), the system automatically adjusts the opening setting to 80% after confirming sufficient downstream flow and elevated pressure. This adjustment is done in 5% increments over five minutes to prevent transient hydraulic fluctuations. For example, valve V-102, located at the transition point between the main and branch pipelines, had a dynamic profile showing downstream pressure exceeding 0.85 MPa for 15 consecutive minutes (the design upper limit is 0.8 MPa). The system then gradually reduced the opening from 100% to 80% to prevent excessive flow release. For control points in the same area located downstream and directly connected to high-risk nodes, such as V-104 and V-105, the opening is reduced to 90% and 85%, respectively, implementing a localized flow dispersion and pressure control strategy.

[0041] At nodes involving water pumps, the start / stop sequence and operating mode of the pump group are adjusted. If the current scheduling policy originally planned to start the main supply pump P-201 and return pump P-202 in 5 minutes, the system will determine whether to delay or cancel the start operation based on the risk weight assessment results of the return water area in the map. For example, if the branch pressure connected to P-202 remains high due to a false sensor alarm, the system determines that the current pressure replenishment will create an overpressure risk. Therefore, its start time is delayed by 15 minutes and it remains in a low-speed preheating state. Meanwhile, the main supply pump P-201 is operated in a slow start mode, with the starting speed set to be reduced from 1450 rpm to 1150 rpm. Before starting the pump, real-time pressure data within the pipeline network is verified to ensure that all monitoring points do not exceed the operating limit. The operating status of the entire pump group is controlled by the map feedback cycle, with each control cycle lasting 10 minutes. The system will dynamically adjust the pump group operating parameters after the next risk assessment update.

[0042] Temporarily reset the parameters of the alarm model currently deployed at each node to prevent invalid alarms from being frequently triggered by distorted data output by common-mode offset sensors. Taking the pressure monitoring point as an example, the original alarm upper limit was 0.85 MPa and the lower limit was 0.25 MPa. If the sampling value of a node exceeds the upper limit for three consecutive times, and its corresponding sensor has been identified as an object affected by common-mode offset, the system will temporarily raise the upper limit to 0.95 MPa and add a "homologous device comparison rule" to the alarm conditions. That is, if more than 70% of the same type of sensors show consistent offset, the alarm information will not be triggered immediately, but will be transferred to the manual review process. This mechanism can minimize the risk of regulatory errors caused by group false alarms without affecting the speed of abnormal response. The adjusted alarm threshold will be maintained for 30 minutes. If the data returns to normal within this time period, the original setting will be restored.

[0043] This step, after identifying and quantifying the impact of common-mode risk, allows targeted adjustments to key control parameters in the pipe network control system based on the node risk levels and operational status feedback provided by the dynamic risk weighting map. This helps mitigate control misjudgments and response failures caused by collective sensor offsets. Common-mode offset sensors often output a wide range of consistent, distorted data within a short period of time, such as a simultaneous increase in pressure or decrease in flow rate. If the system indiscriminately relies on this data to perform control operations, it can easily lead to erroneous valve openings, pump startups and shutdowns, and frequent system alarms. This can lead to water supply imbalances, increased energy consumption, and even engineering risks such as pipe bursts and thermal fluctuations. Therefore, this step reads the control device status of the high-risk node in the dynamic risk weight map and accurately adjusts the corresponding valve opening, for example, reducing the opening from 100% to 80% to weaken the false pressure compensation behavior caused by false high pressure; configures the peak-shaving or slow-start strategy for the start-stop sequence of the pump group, for example, delaying the two groups of pumps originally planned to start simultaneously by 5 minutes and increasing the speed in stages to prevent water hammer impact caused by instantaneous flow surges; at the same time, for the alarm logic driven by sensor data in the analysis model, dynamically increase the upper and lower limits of the alarm threshold, or introduce time lag and group consistency judgment mechanism to avoid frequent triggering of invalid alarms due to false anomalies. Through these adjustment methods, the system no longer relies entirely on the currently collected abnormal data, but makes a comprehensive judgment based on risk distribution, structural characteristics and operating trends to achieve a more robust, safe and intelligent control response. This step not only improves the system's adaptability to data anomalies, but also provides key regulatory support for ensuring the continuity, safety and stability of water supply, heating and other pipeline systems.

[0044] After completing the control operation, the closed-loop adaptive update module writes the operating data during the execution process into the device attribute mapping matrix, re-extracts the cross-point data difference factors and timing synchronization factors of the same-source sensors, continuously analyzes the offset behavior per unit time, and updates the group anomaly initial judgment vector and dynamic risk weight map to achieve anomaly consistency identification and continuous closed-loop optimization of control parameters; To achieve continuous closed-loop optimization of anomaly consistency identification and control parameter settings, after completing the control operation guided by the dynamic risk weight map, all types of real-time operating data during the control execution process must be fully written into the device attribute mapping matrix. Based on the updated matrix, the response behavior of the same-source sensor in the current time period is re-analyzed to further update the group anomaly initial judgment vector and the dynamic risk weight map to ensure that the system has the ability to continuously perceive, feedback correct, and adaptively adjust. The process includes the following steps: All operating data during this round of regulation and control will be synchronously written into the device attribute mapping matrix. The operating data includes the real-time pressure value of the pressure sensor (in MPa), the flow rate change of the flow sensor (in cubic meters per hour), the medium temperature of the temperature sensor (in degrees Celsius), and the response status parameters of the control equipment (such as electric valves and pump groups), such as the actual valve opening percentage, the pump group start-up delay time, the inverter start-stop frequency, etc. Each type of data is marked with a unified timestamp, and the time accuracy is not less than the minute level to ensure that the data time series is complete and comparable, and is classified into the corresponding mapping matrix row item according to the unique identification number of the device. At the same time, to ensure data quality, the system will perform format verification and outlier screening on all data before writing to ensure that there is no repeated sampling, missing format or severely offset values.

[0045] Based on the updated device attribute mapping matrix, re-screen the set of homologous sensors with the same model, the same production batch, and adjacent deployment locations within the current time period, and extract their corresponding operating parameter data. During the data preparation phase, ensure that the extracted time period covers the entire process of control execution, for example, from 5 minutes before the start of regulation to 15 minutes after the end of regulation, to form a complete behavioral observation window with a minimum length of not less than 20 minutes. Align the operating data of all homologous devices within this time period, and calculate the cross-point data difference factor to measure the numerical offset amplitude of different devices at the same time point. At the same time, extract the time series synchronization factor to evaluate the consistency of the change trend between devices, such as whether there is a pressure increase or flow decrease trend at the same time within a certain time period. These factors together form the basis of the group behavior characteristics of the current period.

[0046] Based on the difference factors and synchronization factors extracted above, the initial judgment vector of group anomalies in the current time period is regenerated. If, in the current time period, more than 70% of the devices in the same source sensor show a highly consistent offset direction for 10 consecutive minutes (such as a synchronous increase in pressure and an amplitude exceeding twice the historical average), then the time period is marked as a potential abnormal area, and the result is converted into a new initial judgment vector to be used as the input for the next round of common mode risk identification. At the same time, the new behavioral characteristics are compared and analyzed with the previous round of behavioral characteristics to determine whether the changes in device responses before and after the control operation have significantly converged, such as whether the average difference factor has decreased by more than 30% and whether the standard deviation of the synchronization factor has decreased to less than 60% of the previous value. If the above indicators change significantly, the current control strategy is recorded as an effective control measure; otherwise, it is prompted that the strategy model parameters need to be adjusted.

[0047] The dynamic risk weight map is refreshed synchronously based on the updated group abnormality initial judgment vector. The specific operation is: re-evaluate the risk level of each node. If the response parameters tend to be stable due to the control operation, the risk score value will be reduced; if some nodes still have abnormal fluctuations or even expansion trends after the control, the score value will be increased. The edge weights between the nodes in the graph are also adjusted accordingly to reflect the changes in the risk propagation path. For example, after the control, the risk score of an upstream node is reduced from 88 to 72, and the system adjusts it from the first-level response area to the second-level response area. The corresponding pump group control authority and response priority are also lowered accordingly. The entire map refresh process is automatically executed after each control operation, forming a risk dynamic reconstruction mechanism based on data feedback, ensuring that each subsequent round of control strategy is based on the current latest operating status, and realizing the integration of data-driven identification updates, control corrections and behavior prediction capabilities.

[0048] This step implements a closed-loop feedback mechanism for data, identification, and control during the intelligent pipe network operation and maintenance management process. By writing back, analyzing, and updating the operational data generated during the control execution phase, the system is equipped with self-learning, self-correction, and dynamic optimization capabilities. In traditional pipe network control systems, control actions are often executed on a one-time basis, triggering control commands based on the current state. However, this lacks real-time evaluation of execution results and strategic adjustments to the mechanism, which can easily lead to delayed control effects and slow response to subsequent state changes. In this step, by writing operational parameter data such as pressure, flow, and temperature during control execution into the device attribute mapping matrix, the system obtains a complete correspondence between control actions and results. Based on this, a reanalysis of synchronization and variability is performed on homologous sensors, extracting cross-point data variability factors and time-series synchronization factors to determine whether the sensor population exhibits convergence or continues to experience abnormal deviations under the current control measures. If fluctuations in most device parameters decrease and the deviation trend weakens, the current control measures are effective. Otherwise, further optimization of the control logic or reassessment of risk sources is necessary. At the same time, by updating these analysis results into the group anomaly initial judgment vector and dynamic risk weight map, the system can reconstruct the current risk distribution pattern and node response priorities, thereby adopting more accurate and reasonable scheduling decisions in the next round of control. This process enables continuous tracking of common mode anomaly responses, rolling identification of risk status, and adaptive adjustment of control plans. It is a key support link for building highly stable and intelligent urban or industrial pipeline network operation and management capabilities.

[0049] Through the above-mentioned data analysis-based pipe network operation and maintenance management system, accurate identification of common mode anomalies and intelligent optimization of control responses are achieved, effectively making up for the structural shortcomings of existing systems in dealing with the problem of offset of homologous sensor groups. The present invention achieves accurate identification of homologous device groups by constructing a device attribute mapping matrix and systematically integrating multi-dimensional attributes such as sensor model, batch, deployment location and operating sequence; then, through the analysis of difference factors and synchronization factors, an initial judgment mechanism for group abnormal offset is established for the first time, and combined with external induced events such as electromagnetic interference, firmware updates and remote access behavior, the location and mechanism identification of common mode risk sources are achieved. On this basis, a quantitative mapping of operating status to control strategy is achieved through a dynamic risk weight map, enabling the system to dynamically adjust valves, pump groups and alarm thresholds, avoid false data interference, and ensure control accuracy and safety. Finally, through the continuous write-back of operating data and iterative model updates, a data-driven closed-loop optimization mechanism is constructed to achieve the system's continuous adaptability and self-evolution capabilities to complex abnormal situations. This solution not only improves the accuracy and efficiency of abnormal response, but also significantly enhances the resilience and stability of the pipeline network system in a multi-disturbance environment, and has broad practical application value.

[0050] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

Claims

1. A pipe network operation and maintenance management system based on data analysis, characterized by: It includes equipment feature modeling module, group anomaly identification module, common mode risk tracing module, dynamic risk modeling module, control strategy optimization module and closed-loop adaptive update module: The device feature modeling module encodes the sensor model, batch, location, and timing based on historical operation data and real-time acquisition parameters, and constructs a device attribute mapping matrix; The group anomaly recognition module uses the device attribute mapping matrix to identify homologous sensors, extract their operating parameters per unit time, calculate the cross-point difference factor and time series synchronization factor, and generate the group anomaly preliminary judgment vector; The common mode risk tracing module collects electromagnetic interference, firmware upgrade, and remote access records based on the group anomaly initial judgment vector, analyzes their temporal relationship with anomaly offsets, identifies common mode risk inducing factors, and determines their corresponding inducing mechanisms; The dynamic risk modeling module reconstructs the parameter weights within the impact area based on the influence range of common risk-inducing factors, and generates a dynamic risk weight map based on the operating parameter trends to represent the priority relationship between control objectives and resources; The control strategy optimization module adjusts valve openings, pump start and stop sequences, and alarm thresholds within the target area based on a dynamic risk weight map to mitigate interference from common-mode offset sensors and improve control accuracy and safety. The closed-loop adaptive update module writes the control process data into the equipment attribute mapping matrix, updates the group anomaly initial judgment vector and dynamic risk weight map, and realizes closed-loop optimization of anomaly identification and control parameters.

2. The pipe network operation and maintenance management system based on data analysis according to claim 1 is characterized in that: Building the device attribute mapping matrix includes the following steps: Collect operating parameter records, operation and maintenance records, geographic information, and environmental information covering the sensor's operating cycle to form a unified data view; Feature encoding of sensor hardware information, manufacturing information, deployment environment information, and runtime information based on data views; Convert all feature codes into structured fields and embed them into the device attribute mapping matrix in the form of a two-dimensional array; The mapping matrix is ​​filled with missing values, logical conflicts are eliminated, and feature validity is verified to complete the construction of the structured data foundation.

3. The pipe network operation and maintenance management system based on data analysis according to claim 1 is characterized in that: Generating a group anomaly initial judgment vector includes the following steps: Sensors with the same model, production batch, and deployment area are selected to form a homologous device set; Extract the operating parameter data of the device set within a unified time window, perform time alignment and missing data completion; Calculate the cross-point data difference factor and timing synchronization factor between devices in the set to obtain behavioral consistency analysis results; A group abnormality preliminary judgment vector representing the abnormal state of synchronization deviation is generated based on the set threshold.

4. The pipe network operation and maintenance management system based on data analysis according to claim 1, characterized in that: Identifying common risk triggers and determining their corresponding triggering mechanisms involves the following steps: Based on the time period marked by the group anomaly initial judgment vector, the corresponding electromagnetic interference records, firmware upgrade information and remote access logs are extracted; Compare the occurrence time of electromagnetic interference records, firmware upgrade information, and remote access logs with the time sequence of group offset start time; Unify the timelines of the three types of events, construct a time series correlation map, and analyze the impact and credibility of each event on the abnormal deviation; Based on the degree of event overlap and impact, determine the common risk factors that cause group deviation.

5. The pipe network operation and maintenance management system based on data analysis according to claim 1, characterized in that: Generating a dynamic risk weight map involves the following steps: Based on the spatial influence range of common mode risk-inducing factors, calibrate the set of controlled equipment within the risk intervention area; Extract the pressure, flow, and temperature parameters of the devices in the set within a fixed period before and after the common mode event, and calculate the operational response strength; Calculate the comprehensive risk weight value of each device based on the control criticality of the device in the pipeline network; A dynamic risk weight map is constructed based on the comprehensive risk weight values ​​of all equipment and their deployment locations, and the risk gradient channels are marked with weight differences.

6. The pipe network operation and maintenance management system based on data analysis according to claim 1, characterized in that: Adjusting valve openings, pump start and stop sequences, and alarm thresholds within the target area involves the following steps: Extract nodes with risk weight scores higher than 75 points in the dynamic risk weight diagram and calibrate the control intervention equipment list; Adjust the opening setting value of the electric valve according to the risk level of the control point and execute it step by step according to the preset range; Adjust the start and stop sequence and operating speed parameters of the pump group according to the risk level; Temporarily modify the threshold parameters of the alarm model and add homologous device comparison rules to control the alarm triggering conditions.

7. The pipe network operation and maintenance management system based on data analysis according to claim 1, characterized in that: After writing the operating data during the control execution process into the device attribute mapping matrix, the following steps are included: Write pressure value, flow rate, temperature and control device response status data to the corresponding fields; Extract the operating parameter data of the same-source sensor and calculate the cross-point data difference factor and time series synchronization factor; Generate an updated group anomaly initial judgment vector based on the cross-point data difference factor and the time series synchronization factor; Refresh the dynamic risk weight map based on the updated preliminary judgment vector and adjust the risk level and response priority of each node.

Images