Drainage pipeline multi-parameter sensing trenchless repair dynamic regulation and control method and system

Abstract

The invention discloses a drainage pipeline multi-parameter sensing trenchless repair dynamic regulation and control method and system. Through a fusion technology path of multi-source comprehensive data acquisition, edge calculation real-time cross validation of abnormity, pipeline state simulation analysis of defect levels, intelligent algorithm dynamic adjustment of repair parameters and geographic information visualization platform full-process monitoring, closed-loop management from abnormity identification to repair regulation and control is realized. A comprehensive data set is formed through multi-source data acquisition, an edge computing technology is utilized to quickly identify abnormities, a simulation technology is combined to evaluate the severity of defects, then process parameters are optimized and repaired based on defect levels, and a regulation and control instruction is generated through linkage of a visual platform and a pump station dispatching system. And finally, a pipeline repair and operation regulation and control scheme is formed through integration, and the repair effect and the system stability are ensured. According to the method, the pipeline abnormity processing accuracy and the repairing efficiency are remarkably improved, and a technical guarantee is provided for safe and stable operation of an urban drainage system.

Description

Technical Field

[0001] This invention relates to the field of drainage pipeline repair and management technology, and in particular discloses a method and system for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing. Background Technology

[0002] Currently, drainage pipeline repair methods face several deep-seated challenges. Many traditional technologies struggle to fully perceive and respond promptly to the complexity and variability of pipeline operations, particularly lacking sufficient understanding of the internal conditions of the pipeline under different operating circumstances, resulting in a lack of targeted repair processes. Furthermore, adjustments to repair techniques often rely on manual experience, lacking adaptability to real-time environmental changes, making it difficult to achieve optimal repair results and potentially even triggering secondary problems.

[0003] A deeper technical challenge lies in how to effectively coordinate the dynamic perception of pipeline conditions with real-time adjustments to the repair process. This core factor directly determines the accuracy and efficiency of the repair work. Due to the complex operating environment of pipelines, such as varying water levels and unpredictable flow rates, it is difficult to determine the extent of pipeline defects and the urgency of repair without real-time access to this crucial information. Furthermore, without immediate feedback on this information, adjustments to repair process parameters may be delayed or inaccurate. For example, failure to adjust material dosage or curing time based on actual water flow conditions during the repair process can ultimately lead to poor repair results or even failure.

[0004] Therefore, how to achieve real-time perception of the pipeline's operating status during drainage pipeline repair, and dynamically adjust the repair process based on this information, has become a critical issue that urgently needs to be addressed. Summary of the Invention

[0005] This invention provides a method and system for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing, aiming to solve at least one defect in the prior art.

[0006] One aspect of the present invention relates to a method for dynamic control of trenchless repair of drainage pipelines by multi-parameter sensing, comprising the following steps: S100: Collects data on water level, flow rate, water quality, and structural settlement inside drainage pipes using various sensor devices to form a multi-source integrated data set; S200: Based on a multi-source integrated data set, edge computing technology is used to perform real-time cross-validation of various types of data, identify abnormal situations and generate anomaly identification results; S300. If the anomaly identification result exceeds the preset safety range, the anomaly data will be analyzed using pipeline state simulation technology, and the defect level will be determined in combination with the pipeline operating environment. S400: Determine the repair process requirements based on the defect level, use intelligent algorithms to dynamically adjust the pipeline lining repair process parameters, and generate an optimized repair parameter scheme. S500 uses a geographic information visualization platform to monitor the optimized repair parameter scheme in real time throughout the entire process and links the pump station scheduling system to generate control instructions. S600 integrates control commands and monitoring data to form the final pipeline repair and operation control scheme, ensuring repair effectiveness and system stability.

[0007] Further, step S100 includes: S110: Receive water level data, flow data, water quality data and sedimentation data uploaded by sensors, extract the collection timestamp and geospatial coordinate information to generate the original multi-source heterogeneous data stream, wherein the sensors are distributed at each node of the drainage pipe network; S120. Perform time-series alignment and interpolation completion based on the original multi-source heterogeneous data stream to obtain a standardized multi-dimensional feature vector sequence; S130. Map the standardized multidimensional feature vector sequence to construct a local correlation matrix, and calculate the cross-correlation function value of the local correlation matrix to obtain the weighted correlation feature map; S140. Based on the weighted correlation feature map, feature fusion and structured encapsulation are performed to form a multi-source integrated data set.

[0008] Further, step S200 includes: S210: Receive the multi-source integrated data set loaded onto the edge computing node, map the multi-source integrated data set to the logical topology of the drainage network, and parse to obtain the local state data stream; S220. Perform physical mechanism analysis on the local state data stream to extract the water level-discharge relationship curve and water quality sedimentation characteristic sequence; S230. Calculate the data consistency verification coefficient based on the water level-flow rate relationship curve and the water quality sedimentation characteristic sequence, and determine the mutation feature vector based on the data consistency verification coefficient. S240. If the mutation feature vector exceeds the preset abnormal deviation threshold, lock the abnormal data source and generate an anomaly identification result, wherein the anomaly identification result includes specific fault attributes.

[0009] Further, step S300 includes: S310. Obtain abnormal data and pipeline operating environment data corresponding to the abnormal identification results that exceed the preset safety range, wherein the pipeline operating environment data includes soil type and groundwater level; S320. Construct a local hydraulic structure coupling matrix based on abnormal data and pipeline operating environment data. The local hydraulic structure coupling matrix is ​​used to drive the finite element analysis engine. S330. Calculate the pipe wall stress distribution tensor and fluid pressure fluctuation vector based on the local hydraulic structure coupling matrix, and extract the simulation damage feature set from them; S340. Determine the defect level based on the simulated damage feature set.

[0010] Further, step S400 includes: S410. Obtain the damage depth data corresponding to the defect level, and match the base resin viscosity value based on the damage depth data. S420. The rheology of the mixed fluid is simulated based on the viscosity of the base resin to determine the design thickness of the inner liner. S430. Based on the design thickness of the inner liner and the rheological properties of the mixed fluid, the overturning pressure threshold and curing heating rate are determined, and the residual stress distribution is derived. S440. If the residual stress distribution satisfies the interface peeling limit, then an optimized repair parameter scheme is generated.

[0011] Further, step S500 includes: S510. Obtain the spatial coordinates of the construction pipe section based on the optimized repair parameter scheme, and load the three-dimensional pipe network topology structure in the geographic information visualization platform based on the spatial coordinates of the construction pipe section. S520: Receives real-time curing temperature and lining pressure values, and renders the real-time curing temperature and lining pressure values ​​onto the three-dimensional pipeline topology to generate a visual monitoring layer. S530. Extract the areas of abnormal pressure fluctuations in the visualization monitoring layer and calculate the hydrostatic pressure increment by combining the real-time liquid level height of the upstream pump station's sump. S540. If the sum of the hydrostatic pressure increment and the lining pressure value exceeds the process safety limit, a control command for pump station linkage containing the target frequency setting will be generated.

[0012] Further, step S600 includes: S610. Obtain the control commands of the pump station linkage and the monitoring data of real-time curing temperature and lining pressure, and generate a time-series associated dataset by aligning the control commands and monitoring data according to the timestamp. S620. Input the time-series correlation dataset into the pipeline hydraulic and material solidification coupling model to simulate the dynamic response curves of fluid impact and solidification rate. S630. Extract the risk period from the dynamic response curve, and calculate the parameters that meet the structural strength constraints for the risk period to generate an optimized configuration set for impact-resistant repair operation. S640: Based on the impact-resistant repair operation optimization configuration set, reconstruct the construction time window and flow load distribution table, and output the final pipeline repair and operation control scheme that integrates hydraulic scheduling constraints and material solidification requirements.

[0013] Another aspect of the present invention relates to a multi-parameter sensing trenchless repair dynamic control system for drainage pipelines, used to execute the above-described multi-parameter sensing trenchless repair dynamic control method for drainage pipelines, comprising: The multi-source integrated data set generation module is used to collect data related to water level, flow rate, water quality, and structural settlement inside drainage pipes through various sensor devices, and form a multi-source integrated data set. The anomaly identification result generation module is used to perform real-time cross-validation of various types of data based on a multi-source comprehensive dataset and edge computing technology to identify anomalies and generate anomaly identification results. The defect level determination module is used to analyze the abnormal data using pipeline state simulation technology and determine the defect level in combination with the pipeline operating environment if the anomaly identification result exceeds the preset safety range. The repair parameter scheme generation module is used to determine the repair process requirements based on the defect level, dynamically adjust the pipeline lining repair process parameters using intelligent algorithms, and generate an optimized repair parameter scheme. The control instruction generation module is used to monitor the optimized repair parameter scheme in real time throughout the entire process with the help of the geographic information visualization platform, and to generate control instructions in conjunction with the pump station scheduling system. The pipeline repair and operation control scheme generation module is used to integrate control commands and monitoring data to form the final pipeline repair and operation control scheme, ensuring repair effectiveness and system stability.

[0014] The beneficial effects achieved by this invention are as follows: This invention provides a multi-parameter sensing, trenchless repair and dynamic control method and system for drainage pipelines. Addressing the business scenario of identifying and optimizing repairs for multi-dimensional anomalies such as water level, flow rate, water quality, and structural settlement during drainage pipeline operation, this invention achieves closed-loop management from anomaly identification to repair control through a fusion of technologies including multi-source integrated data acquisition, real-time cross-validation of anomalies using edge computing, pipeline state simulation analysis of defect levels, intelligent algorithms for dynamic adjustment of repair parameters, and full-process monitoring via a geographic information visualization platform. First, this invention forms a comprehensive dataset through multi-source data acquisition. It then uses edge computing technology to quickly identify anomalies and combines this with simulation technology to assess the severity of defects. Subsequently, it optimizes repair process parameters based on defect levels and generates control commands through linkage between the visualization platform and the pump station scheduling system. Finally, it integrates these into a pipeline repair and operation control scheme, ensuring repair effectiveness and system stability. The core innovation of this invention lies in the deep integration of multi-source data and intelligent algorithms, significantly improving the accuracy and efficiency of pipeline anomaly handling and repair, providing technical assurance for the safe and stable operation of urban drainage systems. Attached Figure Description

[0015] Figure 1This is a flowchart illustrating an embodiment of the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines according to the present invention. Figure 2 This is a functional block diagram of an embodiment of the multi-parameter sensing trenchless repair dynamic control system for drainage pipelines of the present invention.

[0016] Explanation of icon numbers: 10. Multi-source integrated data set generation module; 20. Anomaly identification result generation module; 30. Defect level determination module; 40. Repair parameter scheme generation module; 50. Control command generation module; 60. Pipeline repair and operation control scheme generation module. Detailed Implementation

[0017] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0018] like Figure 1 As shown, the first embodiment of the present invention proposes a dynamic control method for trenchless repair of drainage pipelines with multi-parameter sensing, comprising the following steps: Step S100: Collect data on water level, flow rate, water quality, and structural settlement inside the drainage pipe using various sensor devices to form a multi-source integrated data set.

[0019] Using drainage pipelines as the monitoring target, multiple types of sensor devices are deployed at key nodes inside the pipeline (such as pipe section intersections, historical damage points, and low-lying waterlogged sections) to simultaneously collect four types of core operational data, forming a multi-source comprehensive data set: 1. Water level data: Real-time water level height and rate of change in the pipeline are collected by a liquid level sensor, reflecting the pipeline's drainage capacity and risk of siltation.

[0020] 2. Flow data: Instantaneous flow and cumulative flow are collected by electromagnetic flow meters or ultrasonic flow meters to quantify the water flow efficiency of the pipeline.

[0021] 3. Water quality data: Water quality sensors collect indicators such as pH value, suspended solids concentration, and chemical oxygen demand (COD) to determine the corrosiveness of wastewater and the trend of pipe scaling.

[0022] 4. Structural settlement data: The deformation of the inner wall of the pipeline, the displacement of joints, and the settlement amplitude are collected by tilt sensors and strain sensors, reflecting the integrity of the pipeline structure.

[0023] All data is associated with the collection timestamp, sensor geographic coordinates, and pipe section number, and is packaged in a standardized format to provide a comprehensive and traceable data source for subsequent cross-validation.

[0024] Core value: By collecting multi-dimensional parameters, it achieves full-state perception of pipeline "hydraulic characteristics + water quality characteristics + structural characteristics", breaking the limitations of single parameter monitoring.

[0025] Step S200: Based on the multi-source integrated data set, edge computing technology is used to perform real-time cross-validation on various types of data, identify abnormal situations, and generate abnormal identification results.

[0026] By leveraging edge computing units deployed near pipeline monitoring nodes, real-time cross-validation is performed on multi-source integrated data sets, eliminating measurement errors from single sensors and the problem of data silos. 1. Data consistency verification: Compare the correlation between water level and flow rate data for the same pipe section (e.g., rising water level corresponds to a decrease in flow rate, which is consistent with the hydraulic characteristic change pattern caused by siltation), and eliminate contradictory data.

[0027] 2. Abnormal threshold verification: Compare the data of each parameter with the preset normal range (such as water level ≤ 0.8 times pipe diameter, settlement amplitude ≤ 5mm) and mark the abnormal data that exceeds the range.

[0028] 3. Trend mutation verification: Analyze the time-series change trend of data (such as sudden drop in flow rate, sudden increase in sedimentation rate) to identify non-gradual abnormal fluctuations.

[0029] Through the above verification, real and valid abnormal data are filtered out, and anomaly identification results are generated, including information such as abnormal parameter type, occurrence location, anomaly degree, and duration. If there is no abnormal data, continuous monitoring can continue without triggering subsequent processes.

[0030] Core value: Edge computing enables local real-time data processing, reducing cloud transmission pressure, and cross-validation improves the accuracy of anomaly identification, avoiding false positives and false negatives.

[0031] Step S300: If the anomaly identification result exceeds the preset safety range, the abnormal data is analyzed using pipeline state simulation technology, and the defect level is determined in combination with the pipeline operating environment.

[0032] If the anomaly identification results exceed the preset safety range (e.g., settlement amplitude > 10mm, water level full pipe duration > 30 minutes), then pipeline state simulation technology (e.g., pipeline structure simulation based on finite element analysis, hydraulic dynamics simulation model) will be activated to conduct in-depth analysis of the anomaly data. 1. Structural Simulation: Input pipeline settlement and deformation data to simulate the impact of defects on pipeline load-bearing capacity and sealing performance, and determine whether there is a risk of rupture or leakage.

[0033] 2. Hydraulic Simulation: Input water level and flow rate data to simulate the impact of siltation and blockage on the drainage capacity of the pipeline and predict the risk range of water accumulation and overflow.

[0034] Based on the pipeline operating environment (such as surrounding geological conditions, traffic load, and sewage composition), abnormal defects are classified into three defect levels according to their degree of impact and urgency of repair: Level I (Emergency Defect): The pipeline is at risk of rupture and leakage, or structural instability, and requires immediate repair.

[0035] Level II (Moderate Defect): Reduced drainage capacity of the pipeline and localized siltation, requiring repair within a specified period.

[0036] Level III (Minor Defects): Minor deformation and scaling of the pipeline, which can be included in periodic maintenance.

[0037] Core value: By quantifying the impact of abnormal defects through simulation technology, we can avoid the bias of empirical judgment and provide a scientific basis for the selection of repair processes.

[0038] Step S400: Determine the repair process requirements based on the defect level, use intelligent algorithms to dynamically adjust the pipeline lining repair process parameters, and generate an optimized repair parameter scheme.

[0039] Based on the defect level, the corresponding trenchless repair process requirements are determined (e.g., Level I defects are adapted to CIPP (Cured In Place Pipe) overturning lining repair, Level II defects are adapted to localized point repair, and Level III defects are adapted to high-pressure cleaning repair). Intelligent algorithms (such as genetic algorithms and fuzzy control algorithms) are used to dynamically adjust the repair process parameters, generating an optimized repair parameter scheme. 1. For CIPP lining repair: Adjust the lining material thickness, curing temperature, and curing time according to the pipe diameter and deformation. 2. For localized spot repair: Adjust the pressure of the repair airbag, the amount of resin injected, and the curing time according to the location and size of the defect; 3. For high-pressure cleaning and repair: Adjust the cleaning pressure, nozzle speed, and propulsion speed according to the degree of siltation.

[0040] 4. The algorithm optimization goal is "optimal repair effect + lowest cost + shortest construction period", ensuring that the parameters are adapted to the specific defect characteristics and avoiding over-repair or under-repair.

[0041] Core value: Intelligent algorithms enable precise adaptation of repair parameters, improving the success rate of trenchless repair and reducing repair costs.

[0042] Step S500: Use the geographic information visualization platform to monitor the optimized repair parameter scheme in real time throughout the entire process, and link the pump station scheduling system to generate control instructions.

[0043] The optimized repair parameter scheme is imported into the Geographic Information System (GIS) platform to achieve real-time monitoring of the entire repair process. 1. Visual display: Mark the location of defects, repair process, parameter configuration, and construction progress on the platform, and support management personnel to view remotely.

[0044] 2. Real-time monitoring: By using sensors to provide feedback on pipeline status data during the repair process (such as lining curing temperature and sealing of the repaired area), construction parameters can be dynamically adjusted.

[0045] Simultaneously, the system coordinates with the pumping station dispatching system to generate control commands based on changes in drainage capacity caused by pipeline defects: for example, for pumping stations upstream of silted pipe sections, the pumping head is lowered to avoid pipeline overload; for pumping stations around leaking pipe sections, the operating hours are adjusted to reduce sewage erosion of the soil.

[0046] Core value: The visualization platform enables transparent management of the repair process, and the coordinated control of pump stations ensures the stable operation of the pipeline system during the repair period, avoiding secondary disasters.

[0047] Step S600: Based on control commands and monitoring data, integrate them to form the final pipeline repair and operation control plan to ensure repair effectiveness and system stability.

[0048] By integrating repair parameter schemes, visualized monitoring data, and pump station control commands, a final pipeline repair and operation control scheme is formed, which includes two core components: 1. Repair Implementation Plan: Clearly define the construction process, personnel and equipment allocation, quality acceptance standards, and safety assurance measures; 2. Operation and control plan: Clarify the pump station scheduling rules, pipeline monitoring frequency, and subsequent maintenance plan.

[0049] After the plan is implemented, pipeline operation data will be continuously collected to verify the repair effect (such as whether the defects have been eliminated and whether the drainage capacity has been restored), ensuring that the pipeline system returns to a stable operating state and forming a complete closed loop of "monitoring-repair-control-verification".

[0050] Core value: The integrated solution enables coordinated management and control of repair and operation, ensuring the long-term effectiveness of repairs and improving the overall operational stability of the drainage pipeline system.

[0051] The multi-parameter sensing trenchless repair dynamic control method for drainage pipelines provided in this embodiment has the core advantage of dynamically adapting multi-parameter sensing and trenchless repair. It not only achieves accurate diagnosis of pipeline defects, but also reduces the impact of repair on urban operations through intelligent control. It is applicable to trenchless repair projects in various scenarios such as municipal drainage networks and industrial park sewage networks.

[0052] Furthermore, the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines provided in this embodiment includes step S100 as follows: Step S110: Receive water level data, flow data, water quality data and sedimentation data uploaded by the sensors, extract the collection timestamp and geospatial coordinate information to generate the original multi-source heterogeneous data stream, wherein the sensors are distributed at each node of the drainage pipe network.

[0053] The multidimensional data acquisition structure of a single sensor node is defined by the following formula: (1) In formula (1), Indicates sensor At any moment The complete dataset collected, Indicates water level data. Represents traffic data, Indicates water quality data, This represents settlement data. Indicates the collection timestamp. It represents geospatial coordinate information. The control logic of formula (1) is the full-element management logic of "multi-dimensional data synchronous acquisition + structured encapsulation" for a single sensor node. The core is to realize the data acquisition and control of "full parameter coverage and spatiotemporal traceability" in drainage pipeline monitoring.

[0054] In drainage network monitoring systems, sensors are distributed at various nodes, such as key connection points and pumping stations in urban sewers. These sensors upload real-time data on water level (e.g., current water depth of 0.5 meters), flow rate (e.g., 10 cubic meters per second), water quality indicators (e.g., pH value of 7.2 and dissolved oxygen content of 5 mg / L), and sedimentation (e.g., ground subsidence of 2 mm). After receiving this data, the system extracts the acquisition timestamp for each data point, such as 2023-10-01 14:30:00, and the geospatial coordinates, such as latitude and longitude (116.4, 39.9), thereby generating a raw, multi-source heterogeneous data stream. This raw, multi-source heterogeneous data stream is multi-source because it originates from different types of sensors; it is heterogeneous because the data format may include numerical, string, or binary forms. In this way, the system can capture the dynamic state of the network, providing a basis for subsequent analysis and ensuring data integrity and real-time performance.

[0055] Step S120: Perform time-series alignment and interpolation completion based on the original multi-source heterogeneous data stream to obtain a standardized multidimensional feature vector sequence.

[0056] The time-aligned multidimensional feature vector is obtained using the following formula: (2) In formula (2), Indicates time The time-aligned multidimensional feature vector Indicates the total number of multi-source data streams. Indicates the first Weight coefficients of each data source, Indicates the first The original feature vectors of each data source, Indicates the first The time offset of each data source relative to the reference time axis. The control logic of formula (2) is the time alignment control logic of "time offset correction + weighted fusion" for multi-source data streams. The core is to solve the problem of time asynchrony of different data sources and generate a fusion feature vector under a unified time axis.

[0057] To perform time-series alignment on these raw data streams, the first step is to standardize the time scale. For example, all data should be aligned to one sampling point per minute. If a timestamp is missing, it should be filled in using linear interpolation. For instance, if water level data is missing at 14:30, it can be interpolated to 0.5 meters based on the 0.4 meters at 14:29 and 0.6 meters at 14:31. Simultaneously, standardization is performed, converting data with different dimensions into dimensionless vectors. For example, water level is normalized to the [0, 1] interval, and flow rate is standardized using z-score. This results in a standardized multidimensional feature vector sequence. For example, a standardized multidimensional feature vector sequence [0.5, 0.3, 7.2, 2.0] corresponds to water level, flow rate, water quality pH, and sedimentation. This standardized multidimensional feature vector sequence not only solves the problem of data inconsistency but also improves the accuracy of subsequent calculations. In business operations, it can effectively monitor flood risks and prevent urban flooding caused by pipeline blockage.

[0058] Step S130: Map the standardized multidimensional feature vector sequence to construct a local correlation matrix, and calculate the cross-correlation function value of the local correlation matrix to obtain the weighted correlation feature map.

[0059] The local correlation matrix is ​​constructed by calculating the inner product between standardized multidimensional eigenvectors using the following formula: (3) In formula (3), Represents the first in the local correlation matrix Line number Column elements, The dimension of the feature vector. Indicates the first The first standardized eigenvector of the th eigenvector One portion, Indicates the first The first standardized eigenvector of the th eigenvector Each component. The control logic of formula (3) is the association analysis logic of "similarity measurement + local association matrix construction" between standardized feature vectors. The core is to measure the similarity of different feature vectors through inner product calculation and generate a local matrix that reflects the strength of data association.

[0060] The following formula is used to calculate the cross-correlation function value of the local correlation matrix: (4) In formula (4), Represents the cross-correlation function value. Indicates the time delay parameter. This represents the total length of the sequence of the correlation matrix. Indicates the first Elements of the local correlation matrix at time 1, Indicates delay The local correlation matrix elements after time step. The control logic of formula (4) is the "time delay correlation analysis + cross-correlation quantization" logic of the local correlation matrix sequence. The core is to measure the similarity of the local correlation matrix under different time delays and to explore the temporal change pattern of data correlation.

[0061] The final weighted correlation feature map is constructed by weighting and combining the feature correlations of different layers using the following formula: (5) In formula (5), Indicates the position in the weighted association feature map and The weight values ​​between them Indicates the number of layers in the feature map. Indicates the first Layer weight coefficients, Indicates the first Layer feature map at location and The correlation strength, Represents the exponentially decaying term. Indicates the first The attenuation parameter of the layer. The control logic of formula (5) is the construction logic of the association feature map of the multi-layer feature map "weighted fusion + spatiotemporal attenuation constraint". The core is to integrate the feature correlation of different levels, and at the same time control the influence of spatial distance on the association strength through exponential attenuation, so as to generate a weighted association map that has both multi-layer information and spatial rationality.

[0062] Continuing with these standardized multidimensional feature vector sequences, the process of mapping and constructing a local correlation matrix involves projecting the vector sequences into a matrix form. For example, data from adjacent nodes are selected to construct a 5x5 matrix, where rows and columns represent different feature dimensions, such as the correlation between water level and flow rate. The matrix element values ​​are obtained by calculating the covariance. Subsequently, cross-correlation function values ​​are calculated. For example, applying the cross-correlation formula to the water level and flow rate sequences in the local correlation matrix yields a correlation coefficient of 0.8, indicating a strong positive correlation. This generates a weighted correlation feature map, which is visualized as a heatmap, with color depth representing correlation strength. This weighted correlation feature map captures the local dependencies between pipeline network nodes and can help identify abnormal patterns in actual business operations.

[0063] Step S140: Perform feature fusion and structured encapsulation based on the weighted correlation feature map to form a multi-source integrated data set.

[0064] The following formula describes the process of forming a unified dataset from multiple data sources through structured encapsulation: (6) In formula (6), This represents a multi-source integrated dataset. Indicates the number of data sources. Indicates the first A structured encapsulation function for each data source. Indicates the first The original data from each data source, Indicates the first Feature transformation matrix of each data source Indicates the first The attribute identifier of each data source. The control logic of formula (6) is the standardized management logic of "structured encapsulation + unified integration" of multi-source data. The core is to transform heterogeneous data sources into a unified data set with consistent format and traceable attributes.

[0065] Feature fusion is performed based on a weighted correlation feature map. First, relevant features are fused using a weighted average method; for example, the correlation weights of water level and settlement (0.7 and 0.3) are fused into a comprehensive risk index. Then, the results are structured and encapsulated, such as into a multi-source integrated dataset in JSON format, including fields like {"timestamp": "2023-10-01 14:30", "coordinates": [116.4, 39.9], "fused feature": [0.6, 8.0]}. This multi-source integrated dataset is easy to store and query. In drainage network management, it can support the decision-making system to provide real-time early warnings of potential collapse points, thereby optimizing resource allocation and reducing maintenance costs.

[0066] Preferably, the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines provided in this embodiment includes step S200 as follows: Step S210: Receive the multi-source integrated data set loaded to the edge computing node, map the multi-source integrated data set to the logical topology of the drainage pipe network, and parse to obtain the local state data stream.

[0067] The following formula is used to define the logical topology of the drainage network: (7) In formula (7), This represents the logical topology of the drainage pipe network. Represents the set of pipeline nodes. Represents the set of edges of the pipeline network. Indicates the first One pipeline node, Indicates the connection node and nodes At the edge of the pipeline, This represents the input multi-source data set. The data mapping function is represented by formula (7). The control logic of formula (7) is the construction logic of "multi-source data-driven node-edge topology mapping" for drainage pipe network. The core is to transform multi-source data into the logical topology structure of the pipe network (the association between nodes and edges) to realize the digital and logical expression of the physical structure of the pipe network.

[0068] The local state data flow is derived using the following formula: (8) In formula (8), Indicates the first The state data stream of a local region and Indicates the start and end times of data parsing. This indicates the total number of data parameters involved in the parsing. Indicates the first The influence weight of each parameter Indicates the first The parameters in time Real-time values, The Dirac function is used to locate the spatial position of a data stream. Indicates the first Spatial coordinates of a local region Indicates the first The spatial coordinates of each parameter. The control logic of formula (8) is "local region state data aggregation logic under spatiotemporal dual constraints". The core is to filter effective data within a specific spatiotemporal range from global multi-parameter data and generate a state data stream focusing on the local region.

[0069] Receiving a multi-source integrated data set loaded onto edge computing nodes first involves transmitting this data from the cloud or central server to distributed edge devices, such as computing nodes installed near drainage network pumping stations. These edge computing nodes possess low-latency processing capabilities and can receive multi-source data sets containing information such as water level, flow rate, water quality, and sedimentation in real time. Specifically, the mapping process maps the data set to the logical topology of the drainage network, which resembles a network graph where nodes represent key points in the network, such as manholes or branch outlets, and edges represent pipe connections. Through parsing, for example, matching timestamps and coordinates in a data set, local state data streams are generated, such as a sequence of flow rate changes for a specific pipe section, reflecting the real-time operational status of that area. This mapping ensures the correlation between the data and the actual network layout, providing spatial context for subsequent analysis. For example, in an urban drainage system, suppose a multi-source data set includes readings from multiple sensors, such as depth values ​​reported by water level sensors and volumetric rates from flow meters. When mapping this data to the logical topology of the drainage network, a graph database is used to represent the logical topology, where each node is assigned a unique identifier.

[0070] The process of parsing the local state data stream involves filtering data from specific areas, such as extracting a continuous data sequence from the upstream pumping station to the downstream discharge outlet, including minute-by-minute water level fluctuation curves, to form a local dynamic flow view. This method helps capture the segmented behavior of the pipeline network and avoids global data clutter.

[0071] Step S220: Perform physical mechanism analysis on the local state data stream and extract the water level-discharge relationship curve and water quality sedimentation characteristic sequence.

[0072] Physical mechanism analysis of local state data streams focuses on understanding the physical laws governing the drainage process, such as Bernoulli's principle of water flow dynamics, which describes the relationship between pressure, velocity, and height of the fluid in a pipe. Through analysis, water level-flow rate curves are extracted, such as plotting a two-dimensional curve of water level versus flow rate, showing the non-linear increase in flow rate as the water level rises. Simultaneously, extracting water quality sedimentation characteristic sequences involves monitoring the temporal changes in pollutant concentration and pipe sedimentation, such as recording a pattern where pH decreases over time accompanied by increased sedimentation. This extraction is based on statistical methods, such as sliding window averaging, to smooth data noise and ensure the reliability of the water level-flow rate curves and water quality sedimentation characteristic sequences. For example, for the local data stream of a main drainage pipe, physical mechanism analysis includes simulating a flow model, where the water level-flow rate curve is obtained by fitting historical data. For instance, the water level-flow rate curve shows that the flow rate increases from 5 cubic meters per second to 20 cubic meters per second as the water level rises from 0.2 meters to 1.5 meters, reflecting the pipe's capacity limit. The water quality sedimentation characteristic sequence is in vector form, such as [pH 6.8, sedimentation 3mm; pH 6.5, sedimentation 5mm]. The analysis process involves correlation testing to reveal how water acidification accelerates sedimentation, thus laying the foundation for anomaly detection.

[0073] Step S230: Calculate the data consistency verification coefficient based on the water level-flow rate relationship curve and the water quality sedimentation characteristic sequence, and determine the mutation feature vector based on the data consistency verification coefficient.

[0074] The data consistency check coefficient is obtained using the following formula: (9) In formula (9), Represents the data consistency verification coefficient. This represents the total number of observed data points. Indicates the first The measured flow rate at a given moment. This represents the flow rate predicted based on the water level-flow rate curve. Indicates the first The water level value at that moment. Indicates the first Water sedimentation value at a given time, This represents the water quality settling benchmark value. The control logic of formula (9) is "flow-water level correlation + water quality settling benchmark dual-dimensional data consistency verification logic". The core is to quantify the reliability of drainage network monitoring data by comparing the "deviation between measured data and predicted / benchmark data".

[0075] The data consistency verification coefficient is calculated based on the water level-flow rate relationship curve and the water quality sedimentation characteristic sequence. This coefficient can be considered a quantitative indicator used to assess the logical consistency between data. For example, the coefficient value is obtained by calculating the ratio of the curve's slope to the sequence's variance; a value of 0.9 indicates high consistency. Furthermore, abrupt change feature vectors are determined based on this coefficient. For instance, if the coefficient suddenly drops to 0.4, vector components such as [water level deviation 0.3, sedimentation change 4mm] are extracted to capture discontinuities in the data. In the context of pipeline monitoring, the calculation process involves using a cosine similarity formula to quantify the matching degree between the water level-flow rate relationship curve and the water quality sedimentation characteristic sequence, thus obtaining the data consistency verification coefficient.

[0076] Step S240: If the mutation feature vector exceeds the preset abnormal deviation threshold, lock the abnormal data source and generate an anomaly identification result, wherein the anomaly identification result includes specific fault attributes.

[0077] The deviation value of the mutation feature vector is obtained by the following formula: (10) In formula (10), The deviation value representing the mutation feature vector. This represents the currently detected mutation feature vector. Represents the reference baseline eigenvector. The standard deviation of the benchmark data is represented. The control logic of formula (10) is "the standard deviation quantification logic between the mutation feature vector and the benchmark". The core is to accurately identify the degree of abnormal mutation of the feature vector through "vector difference + benchmark fluctuation normalization".

[0078] The following formula is used to define the locking conditions for abnormal data sources: (11) In formula (11), This indicates that the data source is locked. This represents the preset abnormal deviation threshold, when the deviation value of the mutation feature vector... Exceeding the preset abnormal deviation threshold When the lock state is 1, it indicates that the abnormal data source is locked; otherwise, it is 0, indicating a normal state. The control logic of formula (11) is "abnormal data source binary judgment and locking logic based on mutation deviation value". The core is to realize the automatic identification and status marking of abnormal data sources through preset threshold.

[0079] The following formula is used to generate anomaly identification results containing specific fault attributes based on the detection results: (12) In formula (12), This represents the generated specific fault attribute vector. This indicates the total number of fault attributes. Indicates the first The weighting coefficients of each fault attribute, Indicates the first Feature matrix of each fault attribute The detected abnormal pattern vector is represented. The control logic of formula (12) is the fault attribute vector generation logic of "weighted feature fusion of multiple fault attributes + abnormal pattern mapping". The core is to transform the detected abnormal pattern into a structured result containing attributes such as specific fault type and severity.

[0080] Subsequently, if the mutation feature vector exceeds a threshold such as 0.5, the abnormal data source is identified, for example, by locating a specific sensor ID, and a result containing fault attributes such as "pipeline blockage" or "sensor failure" is generated. This supports rapid response to maintenance needs in business operations. If the mutation feature vector exceeds a preset abnormal deviation threshold, the system will automatically identify the abnormal data source, for example, by tracing the data path to identify the problem node, and generate anomaly identification results. This result is output in a structured format, containing specific fault attributes such as "water pollution source" or "structural crack," facilitating intervention by maintenance personnel.

[0081] Furthermore, the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines provided in this embodiment includes step S300 as follows: Step S310: Obtain the abnormal data and pipeline operating environment data corresponding to the abnormal identification results that exceed the preset safety range, wherein the pipeline operating environment data includes soil type and groundwater level.

[0082] When anomaly identification results show that certain parameters exceed preset safety ranges, the system immediately acquires corresponding anomaly data and pipeline operating environment data. The pipeline operating environment data primarily includes key information such as soil type and groundwater level. This data originates from real-time sensor data collection and geological exploration databases. For example, anomaly data might include sudden high water level readings and abnormal flow rate drops, while environmental data records that the surrounding soil of that pipeline section is clayey and the groundwater level is high. This type of information helps in understanding the impact of external factors on the pipeline.

[0083] Step S320: Construct a local hydraulic structure coupling matrix based on abnormal data and pipeline operating environment data, wherein the local hydraulic structure coupling matrix is ​​used to drive the finite element analysis engine.

[0084] The local hydraulic structure coupling matrix is ​​constructed using the following formula: (13) In formula (13), Represents the local hydraulic structure coupling matrix. This indicates the total number of abnormal data monitoring points. Indicates the first The weighting coefficients of each outlier data point Indicates the first Geometric transformation matrix of each monitoring point Indicates the first Material property matrix corresponding to each monitoring point express The transpose matrix. The control logic of formula (13) is the local hydraulic structure association construction logic of "weighted coupling of geometric-material properties of multiple monitoring points + matrix expression". The core is to integrate the geometric and material properties of different monitoring points to generate a coupling matrix that reflects the interaction between hydraulics and structure in the local area.

[0085] The process of constructing a local hydraulic structure coupling matrix involves integrating abnormal data with pipeline operating environment data to form a matrix that reflects the interaction between water flow and pipeline structure. This local hydraulic structure coupling matrix is ​​essentially a multi-dimensional array, where rows correspond to different cross-sectional locations of the pipeline, and columns represent various influencing factors such as fluid velocity, pipe wall thickness, soil lateral pressure, and groundwater permeability. By coupling hydraulic parameters with structural parameters, this local hydraulic structure coupling matrix provides the input basis for subsequent finite element analysis, ensuring that both fluid dynamics and solid mechanics effects are considered during the simulation. For example, when an anomaly occurs in a section of an old urban drainage pipeline, the acquired abnormal data includes a sudden rise in water level from 1.2 meters to 2.8 meters accompanied by a sharp decrease in flow, while pipeline operating environment data shows that the soil in this area is highly plastic clay, and the groundwater level is only 0.6 meters below the bottom of the pipe.

[0086] When constructing the local hydraulic structure coupling matrix, the pipeline is first divided into multiple local elements. Each element is assigned hydraulic load values ​​such as fluid pressure and velocity, while structural loads such as the soil lateral pressure coefficient of 0.65 and the buoyancy effect caused by groundwater are incorporated. This generates a numerical matrix describing the local interactions, which drives the finite element analysis engine to perform accurate calculations. The finite element analysis engine is a numerical simulation tool based on the finite element method. It discretizes the pipeline and its surrounding environment into a large number of tiny elements and simulates the response process under complex loads by solving a system of partial differential equations. In this scenario, after receiving the local hydraulic structure coupling matrix, the finite element analysis engine first constructs the overall stiffness matrix, then applies boundary conditions such as fixed-end constraints and external soil constraints, and finally iteratively solves to obtain the stress state at each point on the pipe wall.

[0087] Step S330: Calculate the pipe wall stress distribution tensor and fluid pressure fluctuation vector based on the local hydraulic structure coupling matrix, and extract the simulation damage feature set from them.

[0088] The following formula describes the fundamental relationship for calculating pipe wall stress distribution using the local hydraulic structure coupling matrix: (14) In formula (14), Represents the stress distribution tensor of the pipe wall. This represents the coupling stiffness matrix of the local hydraulic structure. Represents the fluid pressure field vector. This represents the local hydraulic structure coupling damping matrix. This represents the partial derivative of fluid pressure with respect to time. The control logic of formula (14) is "dynamic calculation logic of pipe wall stress under hydraulic-structural coupling action". The core is to quantify the stress distribution on the pipe wall through "static action of fluid pressure + dynamic action of pressure change".

[0089] The following formula is used to calculate the pressure fluctuation characteristics of a fluid system: (15) In formula (15), Represents the fluid pressure fluctuation vector. Represents the fluid mass matrix. Represents the pressure node vector. Represents the fluid damping matrix. Represents the fluid stiffness matrix. This represents the second partial derivative of pressure with respect to time. This represents the first-order partial derivative of pressure with respect to time. The control logic of formula (15) is the "dynamic response calculation logic of pressure fluctuation in fluid system". The core is to quantify the dynamic fluctuation characteristics of fluid pressure over time through the dynamic model of "mass-damping-stiffness".

[0090] The following formula is used to extract key damage features from stress and flow parameters for subsequent analysis: (16) In formula (16), Represents the simulated damage feature set vector. This represents the feature extraction weight matrix. Represents the effective stress vector. Indicates the yield stress threshold. Represents the cyclic displacement vector. Represents the turbulent velocity vector. The simulation time period is represented. The control logic of formula (16) is "damage feature extraction logic of multi-dimensional coupling of stress-flow parameters". The core is to screen and integrate the features that are strongly related to pipeline damage from the key parameters of pipe wall stress and fluid flow.

[0091] When calculating the pipe wall stress distribution tensor and fluid pressure fluctuation vector based on the local hydraulic structure coupling matrix, the finite element analysis engine calculates the principal stress direction and amplitude element by element. For example, it finds a tensile stress concentration area on the inner side of the pipe wall, while the fluid pressure fluctuation vector shows periodic pulsation characteristics, which reflects the water hammer effect caused by possible blockage.

[0092] These calculation results extract a set of simulated damage features, including the maximum equivalent stress value, stress concentration factor, and fatigue damage accumulation index, thus forming a comprehensive feature set describing potential damage. For example, the calculation results show that the principal value of the stress distribution tensor at the bottom of a certain pipe section is tensile stress reaching 85 MPa, exceeding the allowable value of the material, while the fluid pressure fluctuation vector amplitude reaches 0.4 MPa. The extracted damage feature set includes the location of stress over-limit points, crack propagation tendency coefficient of 2.3, and settlement-induced bending damage index. These features collectively point to impaired structural integrity.

[0093] Step S340: Determine the defect level based on the simulated damage feature set.

[0094] The following formula is used to determine the defect level: (17) In formula (17), Indicates the defect level. The overall score representing the damage feature set. This represents the baseline threshold for classifying levels. This indicates the severity level increase factor for a critical defect. The critical value indicating a serious defect. The step function is represented. The control logic of formula (17) is the defect level classification logic of "basic score quantification + severe defect step enhancement". The core is to combine the basic severity of conventional defects with the risk amplification of severe defects to achieve accurate level classification.

[0095] When determining the defect level based on the simulated damage feature set, the system adopts a graded evaluation strategy, comparing the feature values ​​with preset thresholds. For example, when the maximum stress exceeds 80 MPa and the damage index is greater than 2, it is judged as a level three defect, requiring immediate shutdown and maintenance. Lower values ​​correspond to level one or two, requiring only enhanced monitoring. This determination process ensures accurate classification of pipeline hazards and supports operation and maintenance decisions.

[0096] Preferably, the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines provided in this embodiment includes step S400 as follows: Step S410: Obtain the damage depth data corresponding to the defect level, and match the base resin viscosity value based on the damage depth data.

[0097] The following formula describes the nonlinear mapping relationship between defect level and damage depth, and the corresponding damage depth is calculated from the defect level value: (18) In formula (18), Represents damage depth data, Indicates the defect level The calculated results Indicates the depth coefficient. Indicates the level index, Indicates the base depth offset. The defect level is represented by the control logic of formula (18), which is a "nonlinear quantization mapping logic from defect level to damage depth". The core of this logic is to transform the abstract defect level into the specific physical damage depth through a power function relationship.

[0098] The following formula is used to match the most suitable base resin viscosity value based on damage depth data: (19) In formula (19), This indicates the viscosity value of the base resin used for matching. This represents the total number in the viscosity database. Indicates the first The weighting coefficient for viscosity, Indicates the first A basic resin viscosity value. Indicates the depth of damage input. Indicates the first The depth threshold corresponding to each viscosity. The matching function is represented by formula (19). The control logic of formula (19) is "weighted matching logic between damage depth and resin viscosity", the core of which is to select resin viscosity that matches the current damage depth from the viscosity database.

[0099] The system first retrieves corresponding damage depth data from the defect level database. For example, a level 3 defect typically has an average damage depth of 25% to 40% of the pipe wall thickness, while a level 2 defect has a depth of 10% to 25%. This damage depth data is derived from statistical summaries of historical maintenance records and non-destructive testing reports, ensuring consistency with actual pipeline aging conditions. Specifically, when the defect level is determined to be level 2, the system extracts data where the damage depth is 0.35 times the pipe wall thickness. This reflects that corrosion or cracks have penetrated to the structural layer but have not yet completely penetrated. For example, in a scenario involving the repair of a section of cast iron water supply and drainage pipeline in an urban area, the defect level is level 2, and the damage depth data indicates that the average depth of the corrosion pits on the inner wall reaches 4.2 mm, while the total pipe wall thickness is 15 mm. Based on this depth, the system matches the base resin viscosity value from the material library. Generally, the greater the depth, the lower the viscosity of the resin needs to be selected to ensure sufficient wetting. The matching result is a specific value for the base resin viscosity in the range of 800 to 1200 mPa·s, such as an unsaturated polyester resin with a viscosity of 1000 mPa·s. This matching process takes into account the resin's ability to penetrate the damaged surface, avoiding incomplete wetting caused by high viscosity.

[0100] Step S420: Simulate the rheology of the mixed fluid based on the viscosity value of the base resin to determine the design thickness of the inner liner.

[0101] The design thickness of the liner is determined using the following formula: (20) In formula (20), This indicates the optimal design thickness of the inner lining layer. Indicates the maximum stress value. Indicates the pipe diameter. This indicates the elastic modulus of the lining material. The value represents the allowable strain. The control logic of formula (20) is "the lining thickness safety design logic based on material mechanics constraints". The core is to determine the minimum lining thickness that meets the strength requirements by coupling the maximum stress of the pipe wall, the pipe size and the material properties.

[0102] Simulated rheology of mixed fluids refers to evaluating the flow behavior of a selected base resin mixed with a curing agent and accelerator in a specific ratio during rollover construction using a rheological model. This rheological model is primarily based on non-Newtonian fluid properties, analyzing viscosity changes and thixotropy at shear rates. In one embodiment, with a base resin viscosity of 1000 mPa·s and a mixing ratio of 100:2:1, the simulation shows that the viscosity of the mixed fluid decreases rapidly at low shear rates, facilitating pumping, while quickly recovering structural strength at rest. Based on the rheological curves, the design thickness of the inner liner was determined to be 6 mm to ensure sufficient coverage and provide structural reinforcement at a damage depth of 4.2 mm.

[0103] Step S430: Based on the design thickness of the inner liner and the rheological properties of the mixed fluid, solve the overturning pressure threshold and curing heating rate, and derive the residual stress distribution state.

[0104] The critical pressure required for the liner layer to overturn is calculated using the following formula: (twenty one) In formula (21), Indicates the flip pressure threshold. Indicates the yield strength of the inner lining layer. Indicates the design thickness of the inner lining layer. Indicates the pipe radius. Indicates the dynamic viscosity of the mixed fluid. The shear rate is represented by the formula (21). The control logic of the formula (21) is the "mechanical-fluid coupling critical pressure calculation logic of the liner overturning". The core is to combine the material strength of the liner with the viscosity of the fluid to quantify the minimum pressure threshold required for the liner to overturn (the key process of the liner fitting the pipeline in trenchless repair).

[0105] The radial distribution of residual stress inside the material after cooling is described by the following formula: (twenty two) In formula (22), Indicates residual stress. Indicates radial position, Indicates the elastic modulus. Indicates the coefficient of thermal expansion. Indicates temperature change, Represents Poisson's ratio. Indicates the inner radius. Indicates the shrinkage coefficient. The initial strain is represented by the control logic of formula (22), which is "quantification logic of radial distribution of residual stress under thermo-mechanical coupling". The core is to combine the thermal expansion characteristics of the material with geometric constraints to calculate the radial distribution of residual stress inside the liner after cooling.

[0106] Next, based on the designed thickness of the liner and the rheological properties of the mixed fluid, the system solves for the overturning pressure threshold and the curing heating rate. The overturning pressure threshold is calculated by considering the deformation resistance of the liner hose under air or water pressure. It typically needs to exceed atmospheric pressure by 0.15 to 0.25 MPa to overcome friction and elbow resistance, while not exceeding 0.3 MPa to avoid pipe bursting. Specifically, in the aforementioned water supply and drainage pipe case, considering the 6 mm thickness and the low viscosity characteristics simulated by rheological analysis, the overturning pressure threshold is calculated to be 0.2 MPa, ensuring the hose smoothly overturns to the damaged section. The curing heating rate is controlled based on the resin's exothermic peak. For example, when using hot water circulation curing, the heating rate is set to increase the temperature by 5 to 8°C per hour to prevent excessive shrinkage due to rapid curing. In one implementation, deriving the residual stress distribution involves thermo-mechanical coupling analysis, simulating the volume shrinkage of the resin from liquid to solid and the internal stress caused by the temperature gradient. The system treats the liner as a composite material layer and calculates the radial and circumferential residual tensile stresses generated after curing due to a shrinkage rate of approximately 5%.

[0107] Step S440: If the residual stress distribution satisfies the interface peeling limit, then generate the optimized repair parameter scheme.

[0108] The optimized repair parameter scheme is generated using the following formula: (twenty three) In formula (23), This indicates the optimized repair parameter values. Indicates the basic repair parameters. Indicates the maximum residual stress. Indicates the parameter adjustment index. Indicates the current degree of damage. This represents the damage limit value. The control logic of formula (23) is "repair parameter optimization logic with dual constraints of residual stress and damage degree". The core is to dynamically adjust the basic repair parameters by combining the safety redundancy of residual stress and the remaining space of damage.

[0109] If the residual stress distribution shows that the maximum shear stress at the interface is lower than the bond strength limit between the resin and the pipe wall, such as below 2.5 MPa, then the interface peeling limit is met. Finally, an optimized repair parameter scheme is generated, including resin ratio, overturning pressure of 0.2 MPa, heating rate of 6℃ / hour, and lining thickness of 6 mm, which supports on-site trenchless repair operations.

[0110] Furthermore, the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines provided in this embodiment includes step S500 as follows: Step S510: Obtain the spatial coordinates of the construction pipe section according to the optimized repair parameter scheme, and load the three-dimensional pipeline topology structure in the geographic information visualization platform based on the spatial coordinates of the construction pipe section.

[0111] In practical urban underground pipeline network repair, obtaining the spatial coordinates of the construction pipeline segment is a crucial step. Specifically, the spatial coordinates of the construction pipeline segment are typically obtained through a combination of pipeline detection equipment and a geographic information system (GIS). The data includes the pipeline segment's latitude and longitude, elevation, and pipeline direction. This coordinate data can be accurate to the centimeter level, ensuring the accuracy of subsequent 3D modeling. In one implementation, assuming a project repairing an old urban drainage pipeline where the construction pipeline segment is located beneath a main urban road, the detection equipment uses acoustic reflection technology to locate the spatial coordinates of the segment's start and end points: 116.3 degrees East longitude, 39.9 degrees North latitude, and -5 meters elevation, and 116.4 degrees East longitude, 39.9 degrees North latitude, and -5.2 meters elevation, respectively. This data will be uploaded to a geographic information visualization platform, providing the foundation for subsequent loading of the 3D pipeline network topology.

[0112] Next, based on the spatial coordinates of the aforementioned construction pipe section, the geographic information visualization platform loads a 3D pipeline network topology. This process involves presenting the geometry, connections, and surrounding environmental elements of the pipelines in the form of a 3D model. Geographic information visualization platforms typically employ vector modeling technology to overlay pipeline network data with urban terrain layers, creating an intuitive visualization effect. For example, in the aforementioned drainage pipeline repair project, the platform loaded a 3D model of the pipe section, displaying the relative positions of the pipeline to surrounding underground cables and gas pipelines, while also labeling information such as the pipe section's material as concrete, diameter of 1.2 meters, and length of 200 meters. This visualization method allows the construction team to intuitively understand the pipeline network layout.

[0113] Step S520: Receive real-time curing temperature values ​​and lining pressure values, and render the real-time curing temperature values ​​and lining pressure values ​​onto the three-dimensional pipeline topology to generate a visual monitoring layer.

[0114] Subsequently, the system receives real-time curing temperature and lining pressure readings, rendering them onto the 3D pipeline topology to generate a visual monitoring layer. Specifically, the curing temperature is collected in real-time by sensors placed on the pipeline lining, while the lining pressure is monitored by pressure sensors. The data is overlaid on the 3D model as a dynamic heatmap or numerical labels. For example, in this drainage pipeline repair, sensors show that the curing temperature of a certain section of the lining is 65 degrees Celsius and the pressure is 0.18 MPa. These data are rendered as red and blue blocks at the corresponding locations in the 3D model, allowing monitoring personnel to easily monitor the construction status in real time.

[0115] Step S530: Extract the abnormal pressure fluctuation area from the visualization monitoring layer, and calculate the static pressure increment by combining the real-time liquid level height of the upstream pump station's sump.

[0116] The area of ​​abnormal pressure fluctuation is extracted using the following formula: (twenty four) In formula (24), A comprehensive indicator representing abnormal pressure fluctuations within a region. Indicates the total number of monitoring points. Indicates the first Real-time pressure values ​​at each monitoring point Indicates the reference pressure value. Indicates the threshold for anomaly detection. The unit step function is 1 when the pressure deviation exceeds the threshold and 0 otherwise. The control logic of formula (24) is "comprehensive quantitative logic of abnormal areas of pressure deviation at multiple monitoring points". The core is to screen the monitoring points where the pressure deviation exceeds the threshold and aggregate them to generate regional pressure anomaly indicators.

[0117] The following formula is used to calculate the hydrostatic pressure increment: (25) In formula (25), This represents the calculated hydrostatic pressure value. Indicates the density of the liquid. Represents gravitational acceleration. This indicates the current liquid level in the sump. Indicates the reference liquid level height. The atmospheric pressure is represented. The control logic of formula (25) is "the hydrostatic pressure increment calculation logic driven by liquid level change". The core is to calculate the hydrostatic pressure in the collection well by the difference in liquid level height, and then add the atmospheric pressure to obtain the final pressure value.

[0118] The system extracts areas of abnormal pressure fluctuations from the visualized monitoring layer and calculates the hydrostatic pressure increment by combining this with the real-time liquid level height of the upstream pump station's sump. This process requires analyzing the impact of liquid level changes on pipeline pressure using a fluid dynamics model. For example, in the case above, the monitoring layer shows pressure fluctuations in the middle of the pipe section, with the value suddenly increasing from 0.18 MPa to 0.22 MPa. Simultaneously, the liquid level height in the upstream sump rises from 2 meters to 2.5 meters. The system's analysis yields a hydrostatic pressure increment of 0.05 MPa. This increment reflects the direct impact of liquid level changes on the pressure within the pipeline.

[0119] Step S540: If the sum of the hydrostatic pressure increment and the lining pressure value exceeds the process safety limit, then generate a control command for pump station linkage that includes the target frequency setting.

[0120] The following formula is used to define the safety judgment conditions: (26) In formula (26), Indicates the total pressure increment. Indicates the increase in hydrostatic pressure. This indicates the pressure bearing capacity of the lining. This represents the process safety limit value, when the total pressure increment... Exceeding process safety limits The control command for the pump station linkage is triggered at the time. The control logic of formula (26) is the "safety threshold judgment logic for total pressure increment", the core of which is to trigger the corresponding control command by comparing the total pressure with the process safety limit.

[0121] Finally, if the sum of the hydrostatic pressure increment and the lining pressure exceeds the process safety limit, the system will generate a pump station linkage control command containing the target frequency setting. Specifically, the process safety limit is usually set as the maximum pressure that the lining material can withstand, for example, 0.25 MPa. In this case, the sum is 0.27 MPa, exceeding the safety limit. The system automatically generates a command to adjust the operating frequency of the upstream pump station from 60 Hz to 50 Hz to reduce water flow pressure and ensure construction safety.

[0122] Preferably, the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines provided in this embodiment includes step S600 as follows: Step S610: Obtain the control commands of the pump station linkage and the monitoring data of real-time curing temperature and lining pressure, and generate a time-series associated dataset by aligning the control commands and monitoring data according to the timestamp.

[0123] The structural composition of a time-series correlation dataset is defined by the following formula: (27) In formula (27), Indicates time The complete dataset Indicates time Pump station linkage control commands, Indicates time Real-time curing temperature monitoring value, Indicates time The pressure monitoring data of the inner lining, and These represent the start and end times of data collection, respectively. This indicates a specific time point index. The control logic of formula (27) is "the structured integration logic of multi-source monitoring data in the time-series dimension". The core is to aggregate key operation instructions and monitoring data according to time nodes to form a complete time-series associated dataset.

[0124] To obtain the control commands for pump station linkage and the real-time monitoring data of curing temperature and lining pressure, it is first necessary to extract the control commands from the pump station dispatching system. These control commands typically include adjustments to the pump station's operating frequency and flow control parameters, while the monitoring data comes from temperature and pressure sensors on the pipeline lining. Specifically, this data is recorded in timestamp form, for example, temperature and pressure values ​​are collected every minute. The process of aligning the control commands and monitoring data based on timestamps to generate a time-series correlated dataset involves a data synchronization mechanism. This matches the issuance time of the pump station linkage control commands with the corresponding temperature and pressure readings, forming a structured dataset where each data point includes a timestamp, command content, temperature value, and pressure value. For example, in an urban drainage network repair project, a pump station control command is issued at 10:00 AM, adjusting the frequency to 45 Hz. At the same time, the curing temperature is 62 degrees Celsius, and the lining pressure is 0.15 MPa. The dataset generated after timestamp alignment displays a continuous sequence from 10:00 AM to 10:30 AM, revealing the immediate impact of the command on the monitoring data. This alignment method ensures the temporal consistency of the data, providing a reliable foundation for subsequent analysis.

[0125] Step S620: Input the time-series correlation dataset into the pipeline hydraulic and material solidification coupling model to simulate the dynamic response curves of fluid impact and solidification rate.

[0126] Next, the time-series correlated dataset is input into the pipeline hydraulic and material curing coupled model. This model is an integrated simulation framework that combines fluid dynamics equations and material curing kinetics principles to simulate the interaction between fluid behavior within the pipeline and the curing process of the lining material. Specifically, the pipeline hydraulic and material curing coupled model first loads the time-series correlated dataset, and then calculates the changes in fluid impact force and curing rate through numerical simulation. For example, it uses the finite volume method to discretize the hydraulic equations and couples the heat conduction equations to describe the effect of temperature on the curing rate, thereby outputting dynamic response curves. These dynamic response curves, with time as the horizontal axis, show the fluctuations in impact force and rate. For example, in the aforementioned drainage network project, after inputting the time-series correlation dataset, the network hydraulic and material curing coupling model simulated the dynamic response curves of fluid impact and curing rate from 10:00 to 11:00. The peak fluid impact occurred at 10:15, reaching 0.20 MPa, while the curing rate decreased from an initial 0.5 mm / min to 0.3 mm / min. The dynamic response curves of fluid impact and curing rate reflect the dynamic response of weakened impact but slower curing after the frequency of the control command was reduced. The principle of this network hydraulic and material curing coupling model lies in correlating hydraulic parameters such as flow velocity and pressure with material parameters such as viscosity and hardening time, generating the dynamic response curves of fluid impact and curing rate through iterative solutions.

[0127] Step S630: Extract the risk period from the dynamic response curve, and calculate the parameters that meet the structural strength constraints for the risk period to generate an optimized configuration set for impact-resistant repair operation.

[0128] The following formula is used to identify risk periods by calculating the maximum risk value among all response parameters: (28) In formula (28), Indicates time The risk assessment value, Indicates the first The response parameters at time... The dynamic value, Indicates the risk threshold. Indicates the maximum allowed response value. The second derivative of the response parameter reflects the rate of change. Indicates time All response parameters. The control logic of formula (28) is "dynamic change of multiple response parameters + threshold constraint risk period identification logic", the core of which is to filter out the period with the highest risk by the "rate of change" and "degree of deviation" of the response parameters.

[0129] The optimal runtime configuration set is generated through multi-objective optimization using the following formula: (29) In formula (29), This indicates the optimal shock-resistant repair configuration. Indicates the feasible configuration space. Indicates configuration Repair energy consumption Indicates configuration Response time Indicates configuration Performance loss, , , These represent the weighting coefficients for energy consumption, time, and performance, respectively. The control logic of formula (29) is "multi-objective weighted optimization shock-resistant repair configuration selection logic". The core is to balance the three objectives of energy consumption, response time, and performance loss among feasible configurations and select the comprehensive optimal repair configuration.

[0130] The process involves extracting risk periods from the dynamic response curve and calculating parameters that meet structural strength constraints for these periods to generate an optimized operational configuration set for impact-resistant repair. Risk periods refer to the intervals in the curve where the impact force exceeds a threshold or the curing rate falls below a safe value. Specifically, the extraction process involves curve analysis algorithms, such as threshold detection methods to identify high-risk segments. Optimization parameters are then calculated for these segments, such as adjusting the liner thickness or curing agent ratio using a constraint optimization solver, to ensure that structural strength constraints, such as maximum stress, do not exceed the material's yield strength. This results in an optimized operational configuration set for impact-resistant repair. For example, in this project, the dynamic response curve shows a risk period from 10:10 to 10:20, with an impact force of 0.22 MPa. The calculation indicates that the liner thickness needs to be increased to 8 mm and a 5% reinforcing agent added to meet the strength constraint of 0.25 MPa. The generated optimized operational configuration set for impact-resistant repair includes multiple parameter combinations for selection.

[0131] Step S640: Based on the impact-resistant repair operation optimization configuration set, reconstruct the construction time window and flow load distribution table, and output the final pipeline repair and operation control scheme that integrates hydraulic scheduling constraints and material curing requirements.

[0132] Based on the optimized configuration set for impact-resistant repair operations, the construction time window and flow load allocation table are reconstructed. This process integrates the configuration set through scheduling algorithms, redefining the construction period and flow distribution to output a final pipeline repair and operation control scheme that incorporates hydraulic scheduling constraints and material curing requirements. Specifically, the reconstruction involves adjusting the time window, such as avoiding peak flow periods during high-risk periods, and specifying the pump station load ratio in the construction time window and flow load allocation table, for example, a 30% load during off-peak hours, to ensure that the curing process is not disturbed. For example, in this project, the reconstruction window based on the optimized configuration set for impact-resistant repair operations is from 2 PM to 4 PM to avoid the morning peak. The construction time window and flow load allocation table are set to reduce the upstream pump station load to 40%. The final pipeline repair and operation control scheme outputs a comprehensive document covering repair steps, hydraulic constraints such as a maximum flow rate of 2 cubic meters per second, and curing requirements such as a minimum temperature of 60 degrees Celsius, ensuring overall coordination.

[0133] Please see Figure 2This embodiment provides a multi-parameter sensing trenchless repair dynamic control system for drainage pipelines, used to execute the aforementioned multi-parameter sensing trenchless repair dynamic control method for drainage pipelines. It includes a multi-source integrated data set formation module 10, an anomaly identification result generation module 20, a defect level determination module 30, a repair parameter scheme generation module 40, a control command generation module 50, and a pipeline repair and operation control scheme formation module 60. The multi-source integrated data set formation module 10 collects data related to water level, flow rate, water quality, and structural settlement within the drainage pipeline using various sensor devices to form a multi-source integrated data set. The anomaly identification result generation module 20 uses edge computing technology to perform real-time cross-validation of various data based on the multi-source integrated data set to identify anomalies. The system generates anomaly identification results; a defect level determination module 30 is used to analyze the abnormal data using pipeline state simulation technology and determine the defect level in combination with the pipeline operating environment if the anomaly identification results exceed the preset safety range; a repair parameter scheme generation module 40 is used to determine the repair process requirements based on the defect level, dynamically adjust the pipeline lining repair process parameters using intelligent algorithms, and generate an optimized repair parameter scheme; a control command generation module 50 is used to monitor the optimized repair parameter scheme in real time throughout the entire process using a geographic information visualization platform, and generate control commands in conjunction with the pump station scheduling system; and a pipeline repair and operation control scheme formation module 60 is used to integrate the control commands and monitoring data to form the final pipeline repair and operation control scheme, ensuring repair effectiveness and system stability.

[0134] The multi-parameter sensing trenchless repair dynamic control method and system for drainage pipelines provided in this embodiment, compared with existing technologies, addresses the business scenario problem of multi-dimensional anomaly identification and repair optimization in drainage pipeline operation, including water level, flow rate, water quality, and structural settlement. It achieves closed-loop management from anomaly identification to repair control through a fusion technology path that integrates multi-source comprehensive data acquisition, real-time cross-validation of anomalies using edge computing, pipeline status simulation analysis of defect levels, intelligent algorithm dynamic adjustment of repair parameters, and full-process monitoring via a geographic information visualization platform. This embodiment first forms a comprehensive dataset through multi-source data acquisition, quickly identifies anomalies using edge computing technology, and assesses defect severity using simulation technology. Then, it optimizes repair process parameters based on defect levels and generates control commands through linkage between the visualization platform and the pump station scheduling system. Finally, it integrates these into a pipeline repair and operation control scheme, ensuring repair effectiveness and system stability. The core innovation of this embodiment lies in the deep integration of multi-source data and intelligent algorithms, significantly improving the accuracy and efficiency of pipeline anomaly handling and repair, providing technical assurance for the safe and stable operation of urban drainage systems.

[0135] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A method for dynamic control of trenchless repair of drainage pipelines using multi-parameter sensing, characterized in that, Includes the following steps: S100: Collects data on water level, flow rate, water quality, and structural settlement inside drainage pipes using various sensor devices to form a multi-source integrated data set; S200. Based on the multi-source integrated data set, edge computing technology is used to perform real-time cross-validation on various types of data to identify abnormal situations and generate anomaly identification results. S300. If the anomaly identification result exceeds the preset safety range, the abnormal data is analyzed using pipeline state simulation technology, and the defect level is determined in combination with the pipeline operating environment. S400. Determine the repair process requirements based on the defect level, use intelligent algorithms to dynamically adjust the pipeline lining repair process parameters, and generate an optimized repair parameter scheme. S500 uses a geographic information visualization platform to monitor the optimized repair parameter scheme in real time throughout the entire process and links the pump station scheduling system to generate control instructions. S600. Based on the control commands and monitoring data, a final pipeline repair and operation control scheme is formed to ensure the repair effect and system stability.

2. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 1, characterized in that, Step S100 includes: S110. Receive water level data, flow data, water quality data and sedimentation data uploaded by sensors, extract the collection timestamp and geospatial coordinate information to generate a raw multi-source heterogeneous data stream, wherein the sensors are distributed at each node of the drainage pipe network. S120. Perform time-series alignment and interpolation completion on the original multi-source heterogeneous data stream to obtain a standardized multi-dimensional feature vector sequence; S130. Map the standardized multidimensional feature vector sequence to construct a local correlation matrix, and calculate the cross-correlation function value of the local correlation matrix to obtain a weighted correlation feature map; S140. Based on the weighted correlation feature map, feature fusion and structured encapsulation are performed to form a multi-source integrated data set.

3. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 1, characterized in that, Step S200 includes: S210: Receive the multi-source integrated data set loaded onto the edge computing node, map the multi-source integrated data set to the logical topology of the drainage network, and parse to obtain the local state data stream; S220. Perform physical mechanism analysis on the local state data stream to extract the water level-flow relationship curve and water quality sedimentation characteristic sequence; S230. Calculate the data consistency verification coefficient based on the water level-flow rate relationship curve and the water quality sedimentation characteristic sequence, and determine the mutation feature vector based on the data consistency verification coefficient. S240. If the mutation feature vector exceeds the preset abnormal deviation threshold, the abnormal data source is locked and an abnormal identification result is generated, wherein the abnormal identification result includes specific fault attributes.

4. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 1, characterized in that, Step S300 includes: S310. Obtain abnormal data and pipeline operating environment data corresponding to the abnormal identification results that exceed the preset safety range, wherein the pipeline operating environment data includes soil type and groundwater level; S320. Construct a local hydraulic structure coupling matrix based on the abnormal data and the pipeline operating environment data, wherein the local hydraulic structure coupling matrix is ​​used to drive the finite element analysis engine. S330. Calculate the pipe wall stress distribution tensor and fluid pressure fluctuation vector based on the local hydraulic structure coupling matrix, and extract the simulation damage feature set from them; S340. Determine the defect level based on the simulated damage feature set.

5. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 1, characterized in that, Step S400 includes: S410. Obtain the damage depth data corresponding to the defect level, and match the base resin viscosity value according to the damage depth data. S420. Simulate the rheology of the mixed fluid based on the viscosity value of the base resin to determine the design thickness of the inner liner layer. S430. Based on the designed thickness of the inner liner and the rheology of the mixed fluid, solve the overturning pressure threshold and the curing heating rate, and derive the residual stress distribution state; S440. If the residual stress distribution state satisfies the interface peeling limit, then an optimized repair parameter scheme is generated.

6. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 1, characterized in that, Step S500 includes: S510. Obtain the spatial coordinates of the construction pipe section according to the optimized repair parameter scheme, and load the three-dimensional pipe network topology structure in the geographic information visualization platform based on the spatial coordinates of the construction pipe section. S520: Receive real-time curing temperature values ​​and lining pressure values, and render the real-time curing temperature values ​​and lining pressure values ​​onto the three-dimensional pipeline topology to generate a visual monitoring layer. S530. Extract the abnormal pressure fluctuation area in the visualization monitoring layer, and calculate the static pressure increment by combining the real-time liquid level height of the upstream pump station's sump. S540. If the sum of the hydrostatic pressure increment and the pressure bearing value of the lining exceeds the process safety limit, a control command for pump station linkage containing the target frequency setting is generated.

7. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 1, characterized in that, Step S600 includes: S610. Obtain the control commands for pump station linkage and the monitoring data of real-time curing temperature and lining pressure, and generate a time-series associated dataset by aligning the control commands and the monitoring data according to the timestamp. S620. Input the time-series correlation dataset into the pipeline hydraulic and material solidification coupling model to simulate the dynamic response curves of fluid impact and solidification rate. S630. Extract the risk period from the dynamic response curve, and calculate the parameters that satisfy the structural strength constraints for the risk period to generate an optimized configuration set for impact-resistant repair operation. S640. Based on the aforementioned impact-resistant repair operation optimization configuration set, reconstruct the construction time window and flow load distribution table, and output the final pipeline repair and operation control scheme that integrates hydraulic scheduling constraints and material curing requirements.

8. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 7, characterized in that, In step S610, the structural composition of the time-series correlation dataset is defined by the following formula: in, Indicates time The complete dataset Indicates time Pump station linkage control commands, Indicates time Real-time curing temperature monitoring value, Indicates time The pressure monitoring data of the inner lining, and These represent the start and end times of data collection, respectively. This indicates a specific time point index.

9. The method for dynamic control of trenchless repair of drainage pipelines with multi-parameter sensing according to claim 8, characterized in that, In step S630, the risk period is identified by calculating the maximum risk value among all response parameters using the following formula: in, Indicates time The risk assessment value, Indicates the first The response parameters at time... The dynamic value, Indicates the risk threshold. Indicates the maximum allowed response value. The second derivative of the response parameter reflects the rate of change. Indicates time All response parameters; The optimal runtime configuration set is generated through multi-objective optimization using the following formula: in, This indicates the optimal shock-resistant repair configuration. Indicates the feasible configuration space. Indicates configuration Repair energy consumption Indicates configuration Response time Indicates configuration Performance loss, , , These represent the weighting coefficients for energy consumption, time, and performance, respectively.

10. A multi-parameter sensing trenchless repair dynamic control system for drainage pipelines, used to execute the multi-parameter sensing trenchless repair dynamic control method for drainage pipelines as described in any one of claims 1 to 9, characterized in that, include: The multi-source integrated data set formation module (10) is used to collect data on water level, flow rate, water quality and structural settlement inside the drainage pipe through various sensor devices to form a multi-source integrated data set; An anomaly identification result generation module (20) is used to perform real-time cross-validation of various types of data based on the multi-source integrated data set and edge computing technology to identify anomalies and generate anomaly identification results. The defect level determination module (30) is used to analyze the abnormal data using pipeline state simulation technology and determine the defect level in combination with the pipeline operating environment if the abnormal identification result exceeds the preset safety range. The repair parameter scheme generation module (40) is used to determine the repair process requirements based on the defect level, dynamically adjust the pipeline lining repair process parameters using intelligent algorithms, and generate an optimized repair parameter scheme. The control instruction generation module (50) is used to monitor the optimized repair parameter scheme in real time throughout the entire process with the help of the geographic information visualization platform, and to generate control instructions in conjunction with the pump station scheduling system. The pipeline repair and operation control scheme formation module (60) is used to integrate the control instructions and monitoring data to form the final pipeline repair and operation control scheme, so as to ensure the repair effect and system stability.

Images