Submarine pipeline soil liquefaction slippage real-time monitoring system based on sensor network

Abstract

The invention relates to the technical field of ocean engineering safety monitoring, in particular to a submarine pipeline soil liquefaction slippage real-time monitoring system based on a sensor network, which comprises a sensor network deployment module, a dynamic baseline analysis module, a multi-source data space-time calibration module, a coupling risk assessment module and a grading early warning module. Wherein the sensor network deployment module is used for collecting multi-parameter data; the dynamic baseline analysis module is used for calculating the dynamic baseline range of each parameter; the multi-source data space-time calibration module is used for eliminating space-time deviation between sensor nodes and outputting space-time calibration data; the coupling risk assessment module is used for calculating a soil liquefaction slippage risk coefficient; and the grading early warning module is used for triggering early warning. According to the invention, through combination of multi-parameter perception, space-time calibration and risk fusion assessment, precise monitoring and graded early warning of submarine pipeline soil liquefaction slippage are realized, and monitoring continuity and early warning timeliness are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of offshore engineering safety monitoring, and particularly to a real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network. Background Art

[0002] In the exploitation of offshore oil and gas resources and the cross-sea transportation project, as a key infrastructure, the operation stability of submarine pipelines is directly related to the safety and sustainability of marine energy transportation. However, in a complex marine geological environment, affected by multiple factors such as seismic disturbance, tidal scour, and sediment layer structure change, the submarine soil is prone to liquefaction and slip phenomena, resulting in strain concentration, dislocation, and even overall instability of the pipeline, forming a serious engineering disaster risk. To ensure the structural safety of submarine pipelines, it is urgent to establish a real-time monitoring system with high precision, continuity, and multi-parameter perception ability to achieve early identification and hierarchical response to potential liquefaction and slip events.

[0003] Existing monitoring technologies mostly adopt a single sensing type or mainly focus on data point monitoring, lacking a comprehensive analysis mechanism for vibration, pressure, displacement, and strain, and being easily interfered by the submarine environment in actual deployment, resulting in data distortion, delay, or loss, and unable to construct a continuous and reliable early warning chain. At the same time, most current methods have not formed an effective spatio-temporal fusion of multi-source data and risk quantification model, making it difficult to meet the dynamic assessment and response requirements during the slip process. Therefore, it is urgent to propose a real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network to improve the monitoring accuracy and early warning timeliness of the whole process of submarine pipeline soil liquefaction and slip. Summary of the Invention

[0004] Based on the above purpose, the present invention provides a real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network.

[0005] The real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network includes a sensor network deployment module, a dynamic baseline analysis module, a spatio-temporal calibration module for multi-source data, a coupled risk assessment module, and a hierarchical early warning module; wherein: Sensor network deployment module: A plurality of groups of sensor nodes are arranged circumferentially along the submarine pipeline to collect soil vibration acceleration data, pore water pressure data, soil body displacement data, and pipeline strain data, and output multi-parameter data; Dynamic baseline analysis module: It is used to receive the multi-parameter data output by the sensor network deployment module, calculate the dynamic baseline range of each parameter based on a sliding time window, and generate real-time dynamic reference data; Spatio-temporal calibration module for multi-source data: It is used to receive the real-time dynamic reference data output by the dynamic baseline analysis module, eliminate the spatio-temporal deviation between sensor nodes through timestamp alignment and spatial interpolation algorithms, and output spatio-temporally calibrated data; Coupling risk assessment module: used to receive the time-space calibration data output by the time-space calibration module, fuse the vibration acceleration, pore water pressure, soil displacement and pipeline strain data according to the weight, and generate the soil liquefaction slip risk coefficient; Gradual warning module: used to receive the soil liquefaction and sliding risk coefficient output by the coupled risk assessment module, trigger the first-level warning, second-level warning or third-level warning according to the preset threshold, and output the warning signal to the monitoring terminal.

[0006] Optionally, the sensor network deployment module includes a circumferential deployment unit, a multi-parameter acquisition unit, a data preprocessing unit and a time synchronization unit; wherein: Circumferential layout unit: used to set a group of sensor nodes every 10 meters along the axial direction of the seabed pipeline. Each group of nodes includes a vibration acceleration sensor, a pore water pressure sensor, a soil displacement sensor and a pipeline strain sensor, which are installed at 0°, 90°, 180° and 270° directions of the pipeline circumference respectively; Multi-parameter acquisition unit: used to collect soil vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data through each group of sensor nodes; Data preprocessing unit: used to perform time domain filtering, dimension normalization and outlier removal on the collected soil vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data to generate multi-parameter data; Time synchronization unit: used to receive the NTP protocol clock signal sent by the surface repeater, perform millisecond-level time synchronization on each sensor node, and ensure that the timestamps of multiple sets of data are consistent.

[0007] Optionally, the dynamic baseline analysis module includes a parameter grouping processing unit, a sliding window construction unit, an abnormal filtering calculation unit and a dynamic baseline generation unit; wherein: Parameter grouping processing unit: used to receive the multi-parameter data output by the sensor network deployment module, and classify the vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data according to the data source; Sliding window construction unit: used to construct a sliding time window of fixed time length for each type of parameter data. The window length is set to T seconds, the step length is Δt seconds, and the continuous historical data sequence in the current window is extracted at each moment; Abnormal filtering calculation unit: used to perform steady-state volatility analysis on historical data in each sliding window, remove data points whose amplitude exceeds the set standard deviation multiple, and use the median filtering method to generate the stable interval of the parameter in the current window; Dynamic baseline generation unit: After anomaly filtering is completed, it defines the upper and lower bounds of the stable interval of each type of parameter within the sliding window as the dynamic baseline range of the corresponding parameter, and takes the mean value of this range as the real-time dynamic baseline data at the current moment for output.

[0008] Optionally, the anomaly filtering calculation unit includes: Volatility evaluation sub-unit: It is used to evaluate the volatility of the received historical data sequence within each sliding window, and calculate the mean value M and standard deviation S of the sequence; Anomaly point elimination sub-unit: It is used to eliminate the anomaly data points that meet the conditions according to the set deviation threshold multiple A, and generate the data set after elimination; Median filtering sub-unit: It is used to perform median filtering on the data set after elimination, calculate the median value P of the sorted sequence as the reference value of the current parameter steady state level, and extract the upper and lower quartiles at the same time , , , and determine the stable interval .

[0009] Optionally, the multi-source data spatio-temporal calibration module includes a time alignment unit, a spatial position registration unit, and an interpolation calculation unit; among them: Time alignment unit: It is used to receive the real-time dynamic baseline data of each sensor node output by the dynamic baseline analysis module, extract the corresponding time stamps according to the system main clock signal, and uniformly reconstruct the data time axis based on the unified reference time, and eliminate the data frames with time drift to ensure that all node data is comparable at the same time point; Spatial position registration unit: It is used to construct a sensor spatial position mapping table according to the preset geographical coordinates or relative position labels when each sensor node is arranged circumferentially on the pipeline, and perform spatial indexing on the data according to the distribution relationship of the nodes in the three-dimensional coordinate system; Interpolation calculation unit: On the basis of completing time alignment and spatial position registration, for the missing or abnormal data nodes, it performs numerical interpolation reconstruction according to the dynamic baseline values of its spatially adjacent nodes, and at the same time fine-tunes the data of adjacent nodes with the same time series change trend to generate spatio-temporal calibration data.

[0010] Optionally, the spatial position registration unit includes: Node information extraction sub-unit: It is used to extract the unique identifier of each sensor node and its initial position information recorded during deployment from the sensor network deployment module, including the pipeline section number, circumferential angle, and installation depth; 3D coordinate modeling sub-unit: used to calculate the rectangular coordinates (X, Y, Z) of each node in space according to the extracted pipeline segment numbers, circumferential angles, and installation depths, in combination with the 3D trajectory model of the pipeline centerline, and store them in the spatial position mapping table of the corresponding sensor; Spatial index coding sub-unit: used to perform spatial hierarchical coding on all nodes based on the node distribution in the 3D coordinate system, and construct a fast spatial query index structure using the hashing method.

[0011] Optionally, the interpolation calculation unit includes: Proximity node retrieval sub-unit: used to retrieve the set of neighboring nodes of the target missing node in the 3D coordinate space from the spatial position mapping table, perform spherical neighborhood matching using a fixed search radius R, and output all nodes whose spatial distance from the target node is not greater than R as the interpolation reference set, denoted as node set N; Spatial interpolation reconstruction sub-unit: used to perform weighted averaging on the current moment dynamic reference values of each node in node set N according to their Euclidean distances from the target node to generate the interpolation estimated value of the target node ; Temporal trend harmonization sub-unit: used to perform time trend fine-tuning on the estimated value after spatial interpolation First, calculate the mean change difference between the previous moment reference value sequence of the neighboring nodes and the current sequence , and then weight the difference and act on to generate the finally output spatio-temporal calibration data , and its calculation formula is: , where is the temporal trend harmonization coefficient, representing the harmonization weight factor. For example,

[0012] Optionally, the coupling risk assessment module includes a normalization unit, a weight fusion unit, and a risk generation unit; where: Parameter standardization unit: used to receive the vibration acceleration data A, pore water pressure data P, soil displacement data D, and pipeline strain data E output by the spatio-temporal calibration module, and standardize them to the interval [0, 1] respectively using the maximum-minimum normalization method to obtain the standardized parameters , , , , to eliminate the scale differences between different physical quantities; Weight allocation unit: used to set the fusion weights of each type of standardized parameter according to historical slip event analysis and regional geological weight factors, and set the weight set as , corresponding to the influence degrees of vibration acceleration, pore water pressure, soil displacement and pipeline strain respectively, and satisfying the constraint conditions: ; Risk coefficient calculation unit: Based on the standardized parameters and the weight set, calculate the comprehensive soil liquefaction and landslide risk coefficient R, and its calculation formula is: .

[0013] Optionally, the grading early warning module includes a threshold judgment unit, an early warning message generation unit and a signal transmission unit; among them: Threshold judgment unit: used to receive the soil liquefaction and landslide risk coefficient output by the coupling risk assessment module, and compare it with three groups of pre-set risk coefficient thresholds; trigger a level-three early warning when, trigger a level-two early warning when, trigger a level-one early warning when, where respectively correspond to the risk thresholds of local landslide, regional liquefaction and pipeline instability; Early warning message generation unit: construct a structured early warning message based on the triggered early warning level; Signal transmission unit: used to transmit the structured early warning message to the monitoring terminal through a wired or wireless communication network.

[0014] Optionally, the early warning message generation unit includes: Early warning level mapping sub-unit: used to determine the corresponding early warning level identification code according to the early warning level triggered by the risk level result output by the threshold judgment unit, where the level-one early warning is mapped to W1, representing local landslide; the level-two early warning is mapped to W2, representing regional liquefaction; the level-three early warning is mapped to W3, representing pipeline instability; Message field filling unit: used to construct structured content including the following fields for each triggered early warning event: 1. Early warning level identification; 2. Trigger time; 3. Coordinates of the early warning center; 4. Corresponding risk coefficient value; 5. Suggested disposal instructions; Structured format encoding sub-unit: used to encode the constructed field content according to a preset format to generate a unified message format.

[0015] Advantages of the present invention: In the present invention, through the layout of multi-type sensor nodes and data collaborative perception, key parameters such as vibration acceleration, pore water pressure, soil displacement and pipeline strain can be synchronously collected, and by using the dynamic baseline analysis and multi-source data spatio-temporal calibration method, high-stability processing of the monitoring data is realized, significantly improving the accuracy and continuity of landslide risk identification, and solving the problems of parameter drift and abnormal point interference in the complex seabed environment.

[0016] In the present invention, by constructing a coupled risk assessment mechanism, multiple parameters are fused and calculated according to weights to obtain the liquefaction and sliding risk coefficient, and hierarchical early warning is executed in combination with a preset threshold value, and automatically pushed to the monitoring terminal through a structured message, effectively realizing rapid response to different degrees of sliding severity and output of disposal suggestions, and improving the safety guarantee ability and the intelligent level of early warning during the whole life cycle operation of the submarine pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0018] Figure 1 Schematic diagram of the real-time monitoring system for soil liquefaction and sliding of the present invention embodiment; Figure 2 Schematic diagram of the multi-source data spatio-temporal calibration module of the present invention embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The present invention will be described in detail below with reference to the drawings and specific embodiments. At the same time, it should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; and the drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0020] As Figure 1 - Figure 2 shown, the real-time monitoring system for soil liquefaction and sliding of the submarine pipeline based on the sensor network includes a sensor network deployment module, a dynamic baseline analysis module, a multi-source data spatio-temporal calibration module, a coupled risk assessment module, and a hierarchical early warning module; among them: Sensor network deployment module: A plurality of groups of sensor nodes are arranged circumferentially along the submarine pipeline, used to collect soil vibration acceleration data, pore water pressure data, soil displacement data, and pipeline strain data, and output multi-parameter data; Dynamic baseline analysis module: Used to receive the multi-parameter data output by the sensor network deployment module, calculate the dynamic baseline range of each parameter based on a sliding time window, and generate real-time dynamic reference data; Multi-source data spatio-temporal calibration module: Used to receive the real-time dynamic reference data output by the dynamic baseline analysis module, eliminate the spatio-temporal deviation between sensor nodes through timestamp alignment and spatial interpolation algorithms, and output spatio-temporal calibration data; Coupling Risk Assessment Module: It is used to receive the spatio-temporal calibration data output by the spatio-temporal calibration module, fuse the vibration acceleration, pore water pressure, soil displacement, and pipeline strain data according to weights, and generate a soil liquefaction and slip risk coefficient; Hierarchical Early Warning Module: It is used to receive the soil liquefaction and slip risk coefficient output by the coupling risk assessment module, trigger a first-level warning (local slip), a second-level warning (regional liquefaction), or a third-level warning (pipeline instability) according to a preset threshold, and output a warning signal to the monitoring terminal.

[0021] The sensor network deployment module includes a circumferential layout unit, a multi-parameter acquisition unit, a data preprocessing unit, and a time synchronization unit; among them: Circumferential Layout Unit: It is used to set a group of sensor nodes every 10 meters along the axial direction of the submarine pipeline. Each group of nodes includes a vibration acceleration sensor, a pore water pressure sensor, a soil displacement sensor, and a pipeline strain sensor, which are respectively installed at the 0°, 90°, 180°, and 270° azimuths of the pipeline circumference; the sensor nodes are connected to the outer wall of the pipeline through articulated fixing rings. An elastic buffer cushion layer is provided on the inner side of the fixing ring, and the outer side is inserted into the seabed soil through anchor rods to ensure that the sensors deform synchronously with the pipeline and the surrounding soil; Multi-parameter Acquisition Unit: It is used to collect soil vibration acceleration data, pore water pressure data, soil displacement data, and pipeline strain data through each group of sensor nodes; The multi-parameter acquisition unit includes: Vibration Acceleration Sensing Sub-unit: It uses a MEMS triaxial accelerometer, which is installed on the contact surface between the outer wall of the pipeline and the soil, with a measurement range of ±5g and a sampling frequency of 100Hz, and is used to collect vibration acceleration data at the pipeline-soil interface; Pore Water Pressure Sensing Sub-unit: It uses a piezoresistive sensor with a permeable stone protective film, which is embedded at the end of the anchor rod 20 cm away from the outer wall of the pipeline, with a sampling frequency of 1Hz, and measures the pore water pressure data during soil liquefaction; Soil Displacement Sensing Sub-unit: It uses a laser rangefinder sensor, which is oriented towards the soil outside the pipeline at an angle of 45°, with a measurement range of 0-30 cm, and obtains the circumferential soil displacement data of the pipeline in real time; Pipeline Strain Sensing Sub-unit: It uses a fiber Bragg grating sensor, which is welded along the axial direction of the pipeline on the outer surface, and a measurement point is arranged every 1 meter to monitor the pipeline strain data; Data Preprocessing Unit: It is used to perform time-domain filtering, dimensional normalization, and outlier rejection on the collected soil vibration acceleration data, pore water pressure data, soil displacement data, and pipeline strain data, and generate multi-parameter data; Time Synchronization Unit: It is used to receive the NTP (Network Time) protocol clock signal sent by the surface repeater, perform millisecond-level time synchronization on each sensor node, and ensure that the time stamps of multiple groups of data are consistent.

[0022] The dynamic baseline analysis module includes a parameter grouping processing unit, a sliding window construction unit, an abnormal filtering calculation unit and a dynamic baseline generation unit; wherein: Parameter grouping processing unit: used to receive the multi-parameter data output by the sensor network deployment module, and classify the vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data according to the data source, to ensure that each type of data maintains an independent processing path in subsequent analysis; Sliding window construction unit: used to construct a sliding time window of fixed time length for each type of parameter data. The window length is set to T seconds, the step length is Δt seconds, and the continuous historical data sequence in the current window is extracted at each moment; Abnormal filtering calculation unit: used to perform steady-state volatility analysis on historical data in each sliding window, remove data points whose amplitude exceeds the set standard deviation multiple, and use the median filtering method to generate the stable interval of the parameter in the current window; Dynamic benchmark generation unit: after completing anomaly filtering, it is used to define the upper and lower limits of the stable interval of each type of parameter in the sliding window as the dynamic baseline range of the corresponding parameter, and output the mean of the range as the real-time dynamic benchmark data at the current moment; through the above module structure and processing mechanism, it is possible to stably extract short-term stable reference values ​​of various geological and structural parameters in the face of complex interference environments on the seabed, effectively avoid misjudgments caused by single-point mutation data, and provide a highly robust dynamic benchmark basis for subsequent risk identification and early warning.

[0023] The abnormal filtering calculation unit includes: Volatility evaluation subunit: used to evaluate the volatility of the received historical data sequence in each sliding window, and calculate the mean M and standard deviation S of the sequence. The calculation formula is: ,in, represents the i-th historical sampling value, and n represents the total number of data points in the sliding window; Outlier elimination subunit: used to eliminate outliers that meet the conditions according to the set deviation threshold multiple A. Abnormal data points are generated to generate the data set after elimination; Median filter subunit: used to perform median filtering on the data set after elimination, calculate the median value P of the sorted sequence as the reference value of the current parameter steady-state level, and extract the upper and lower quartiles at the same time , , determine the stability interval ; By introducing standard deviation elimination and median filtering, the influence of extreme points such as seismic wave disturbance and local water inrush on data stability can be effectively eliminated, the reliability of the dynamic benchmark of parameters in the sliding window can be improved, and a more robust reference data support can be provided for subsequent coupling risk assessment.

[0024] The multi-source data spatio-temporal calibration module includes a time alignment unit, a spatial position registration unit, and an interpolation calculation unit; among them: Time alignment unit: It is used to receive the real-time dynamic reference data of each sensor node output by the dynamic baseline analysis module, extract the corresponding timestamps according to the system main clock signal, uniformly reconstruct the data time axis based on the unified reference time, and eliminate the data frames with time drift, ensuring that all node data is comparable at the same time point; Spatial position registration unit: It is used to construct a sensor spatial position mapping table according to the preset geographical coordinates or relative position tags when each sensor node is arranged circumferentially on the pipeline, and perform spatial indexing on the data according to the distribution relationship of the nodes in the three-dimensional coordinate system, ensuring that the subsequent interpolation operations are based on the real sensor geometric distribution; Interpolation calculation unit: On the basis of completing time alignment and spatial position registration, for missing or abnormal data nodes, it performs numerical interpolation reconstruction based on the dynamic reference values of its spatially adjacent nodes, and at the same time fine-tunes the data of adjacent nodes with consistent time series change trends to generate continuous, complete and spatio-temporally calibrated data; by setting a three-stage processing mechanism of time alignment, spatial registration and interpolation reconstruction, it can effectively solve the data inconsistency problem caused by synchronization errors and insufficient deployment density in the subsea sensor network, and improve the spatial cooperation ability and time accuracy consistency of multi-source parameters in subsequent fusion calculations.

[0025] The spatial position registration unit includes: Node information extraction sub-unit: It is used to extract the unique identifier of each sensor node and its initial position information recorded during deployment from the sensor network deployment module, including the pipeline section number, circumferential angle, and installation depth, for subsequent coordinate modeling; Three-dimensional coordinate modeling sub-unit: It is used to calculate the rectangular coordinates (X, Y, Z) of each node in space according to the extracted pipeline section number, circumferential angle, and installation depth, combined with the three-dimensional trajectory model of the pipeline center line, and store them in the spatial position mapping table of the corresponding sensor; The specific coordinate calculation formula is as follows: ; ; ; where , , represents the central coordinates of the starting point of the current pipeline section, which is preset by the pipeline layout path; L represents the longitudinal distance (along the center line direction) of the pipeline section where the current node is located; represents the inclination angle of the extension direction of the current section on the horizontal plane, if it is a horizontal section then = 0; R represents the radius distance at which the sensor node is installed, i.e., the pipe radius, which is set as a constant; represents the circumferential angle of the sensor node relative to the center of the pipe cross-section, in degrees; D represents the installation depth of the sensor node vertically downward (unit: meters); X, Y, Z: represent the finally converted three-dimensional space rectangular coordinates.

[0026] Spatial index coding subunit: used to perform spatial hierarchical coding on all nodes according to the node distribution in the three-dimensional coordinate system, and construct a fast spatial query index structure using the hash method to support quickly retrieving the spatial point set in the adjacent area of any node during interpolation operations; by normalizing the sensor installation information into the form of three-dimensional coordinates and introducing a spatial index mechanism, the fast search and accurate expression of the node spatial relationship can be realized, thus supporting high-precision spatial interpolation and coupling calculations, and providing spatial continuity guarantee for subsequent slip risk assessment.

[0027] Table 1 Example of sensor spatial position mapping structure Node number Pipeline segment number Circumferential angle Installation depth X coordinate Y coordinate Z coordinate N001 S01 0 2 100 50 -2 N002 S01 90 2 100 52 -2 N003 S01 180 2 100 50 -4 N004 S01 270 2 100 48 -2 N005 S02 0 2 102 50 -2 In the above Table 1, the node number is the unique number assigned to each sensor node; the pipe segment number represents the pipe segment number where the sensor is located; the circumferential angle represents the angle of the node relative to the center of the pipe cross-section; the installation depth represents the depth at which the node is installed below the seabed; the X, Y, Z coordinates are the actual spatial positions after three-dimensional coordinate modeling transformation according to the segment position, circumferential angle and installation depth, for subsequent interpolation calculations.

[0028] The interpolation calculation unit includes: Adjacent node retrieval subunit: used to retrieve the set of adjacent nodes of the target missing node in the three-dimensional coordinate space from the spatial position mapping table, perform spherical neighborhood matching using a fixed search radius R, and output all nodes whose spatial distance from the target node is not greater than R as the interpolation reference set, denoted as the node set N; Spatial interpolation reconstruction subunit: used to perform weighted average on the current moment dynamic reference values of each node in the node set N according to their Euclidean distance from the target node to generate the interpolation estimated value of the target node , and its calculation formula is: , where, represents the interpolation estimated value of the missing node, represents the dynamic reference value of the i-th adjacent node, represents its Euclidean distance from the target node, and k is the number of adjacent nodes; Temporal trend harmonization subunit: used to perform time trend fine-tuning on the estimated value after spatial interpolation. First, calculate the previous moment reference value sequence of the adjacent nodes and the current sequence The mean change difference , and then the difference is weighted and applied to to generate the final output of spatio-temporal calibration data , and its calculation formula is: , where is the time series trend harmonic coefficient, representing the harmonic weight factor; the above-mentioned sub-unit can effectively compensate for the data loss problem caused by node failure or signal interference by first realizing interpolation reconstruction based on the distance weighted average of spatially adjacent nodes and then performing trend fine-tuning in combination with the time series trend, ensuring the continuity and consistency of the calibration data in both the time and space dimensions, thereby guaranteeing the accuracy and stability of the subsequent coupling evaluation results.

[0029] The coupling risk assessment module includes a normalization unit, a weight fusion unit, and a risk generation unit; among them: Parameter standardization unit: used to receive the vibration acceleration data A, pore water pressure data P, soil displacement data D, and pipeline strain data E output by the spatio-temporal calibration module, and standardize them to the interval [0, 1] respectively using the maximum-minimum normalization method to obtain the standardized parameters , , , , so as to eliminate the scale difference between different physical quantities; Weight allocation unit: used to set the fusion weight of each type of standardized parameter according to the historical slip event analysis and the regional geological weight factor, and set the weight set as , corresponding to the influence degrees of vibration acceleration, pore water pressure, soil displacement, and pipeline strain respectively, and satisfying the constraint condition: ; The weight setting method in the weight allocation unit is as follows: Step 1: First collect the groups of known slip events in the target sea area and similar geological units; each event records its corresponding standardized parameter and the actual slip strength index ; and then construct the historical response matrix H and the observation result vector , and their expressions are respectively: ; ; Step 2: Use the linear least squares method to inversely solve the optimal weight vector to make the predicted risk coefficient as close as possible to the observed value ; then solve the objective function: ; the constraint condition is: ; and then obtain a group of uniquely determined fusion weights , for subsequent real-time monitoring scenarios; Risk coefficient calculation unit: Based on the standardized parameters and weight set, calculate the comprehensive soil liquefaction and slip risk coefficient R, and its calculation formula is: ; where R represents the liquefaction and slip risk level score of the current node or area, and the value range is [0, 1]. The larger the value, the higher the potential risk. By constructing a risk quantification process that integrates standardization and weighting, multi-source heterogeneous parameters can be uniformly incorporated into a unified evaluation framework, improving the objectivity and real-time nature of soil liquefaction and slip risk judgment, and having good regional adaptability, providing a quantitative input basis for downstream hierarchical early warning.

[0030] The hierarchical early warning module includes a threshold judgment unit, an early warning message generation unit, and a signal transmission unit; among them: Threshold judgment unit: Used to receive the soil liquefaction and slip risk coefficient output by the coupled risk assessment module and compare it with three sets of pre-set risk coefficient thresholds; trigger a level-three early warning when, trigger a level-two early warning when, trigger a level-one early warning when, where respectively correspond to the risk thresholds of local slip, regional liquefaction, and pipeline instability; Early warning message generation unit: Construct a structured early warning message based on the triggered early warning level. The message includes information such as an early warning level identifier, a generation timestamp, and the location of the early warning area, so that the monitoring terminal can accurately parse it; Signal transmission unit: Used to transmit the structured early warning message to the monitoring terminal through a wired or wireless communication network to ensure the real-time nature of the early warning information; through multi-level threshold determination combined with structured message and transmission, accurate differentiation and instant notification of slip risks of different severity levels are achieved, improving the response speed and early warning accuracy of the system to submarine pipeline soil liquefaction and slip events.

[0031] The early warning message generation unit includes: Early warning level mapping sub-unit: Used to determine the corresponding early warning level identification code according to the early warning level triggered by the risk level result output by the threshold judgment unit. Among them, a level-one early warning is mapped to W1, representing local slip; a level-two early warning is mapped to W2, representing regional liquefaction; a level-three early warning is mapped to W3, representing pipeline instability; Message field filling unit: Used to construct structured content containing the following fields for each triggered early warning event: 1. Early warning level identifier (W1 / W2 / W3); 2. Trigger time (system standard timestamp); 3. Coordinates of the early warning center (represented by the node number and its three-dimensional position of the trigger point); 4. Corresponding risk coefficient value R; 5. Recommended disposal instructions (attached with an operation instruction number according to the early warning level); Structured format encoding subunit: used to encode the completed field content in a preset format to generate a unified message format; Specific examples are as follows: [LEVEL=W2];[TIME=2025-04-25T10:33:10];[NODE=N037];[POS=(102.3,48.7,-3.2)];[R=0.68];[ACTION=A02], where the field order is fixed and the separator uniformly uses a half-width semicolon, facilitating subsequent parsing and execution; by standardizing the mapping and encoding of the warning level and the structured content fields, the unified expression and rapid transmission of warning information are realized, which not only improves the message parsing efficiency, but also ensures the accurate correspondence of response measures under different warning levels, enhancing the response standardization and automation level of the system to sudden slip events.

[0032] The present invention covers any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of the present invention. To enable the public to have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, and those skilled in the art can fully understand the present invention without these detailed descriptions. In addition, well-known methods, processes, procedures, components, and circuits are not described in detail to avoid unnecessary confusion to the essence of the present invention.

[0033] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A real-time monitoring system for soil liquefaction and slippage of submarine pipelines based on a sensor network, characterized in that: It includes sensor network deployment module, dynamic baseline analysis module, multi-source data spatiotemporal calibration module, coupling risk assessment module and graded warning module; among which: Sensor network deployment module: multiple groups of sensor nodes are deployed around the seabed pipeline to collect soil vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data, and output multi-parameter data; Dynamic baseline analysis module: used to receive multi-parameter data output by the sensor network deployment module, calculate the dynamic baseline range of each parameter based on the sliding time window, and generate real-time dynamic baseline data; Multi-source data spatiotemporal calibration module: used to receive the real-time dynamic baseline data output by the dynamic baseline analysis module, eliminate the spatiotemporal deviation between sensor nodes through timestamp alignment and spatial interpolation algorithms, and output spatiotemporal calibration data; Coupling risk assessment module: used to receive the time-space calibration data output by the time-space calibration module, fuse the vibration acceleration, pore water pressure, soil displacement and pipeline strain data according to the weight, and generate the soil liquefaction slip risk coefficient; Gradual warning module: used to receive the soil liquefaction and sliding risk coefficient output by the coupled risk assessment module, trigger the first-level warning, second-level warning or third-level warning according to the preset threshold, and output the warning signal to the monitoring terminal.

2. The sensor network-based real-time monitoring system for submarine pipeline soil liquefaction and slippage according to claim 1 is characterized in that: The sensor network deployment module includes a circumferential deployment unit, a multi-parameter acquisition unit, a data preprocessing unit and a time synchronization unit; wherein: Circumferential layout unit: used to set a group of sensor nodes every 10 meters along the axial direction of the seabed pipeline. Each group of nodes includes a vibration acceleration sensor, a pore water pressure sensor, a soil displacement sensor and a pipeline strain sensor, which are installed at 0°, 90°, 180° and 270° directions of the pipeline circumference respectively; Multi-parameter acquisition unit: used to collect soil vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data through each group of sensor nodes; Data preprocessing unit: used to perform time domain filtering, dimension normalization and outlier removal on the collected soil vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data to generate multi-parameter data; Time synchronization unit: used to receive the NTP protocol clock signal sent by the surface repeater, perform millisecond-level time synchronization on each sensor node, and ensure that the timestamps of multiple sets of data are consistent.

3. The real-time monitoring system for soil liquefaction and slippage of submarine pipelines based on a sensor network according to claim 1 is characterized in that: The dynamic baseline analysis module includes a parameter grouping processing unit, a sliding window construction unit, an abnormal filtering calculation unit and a dynamic baseline generation unit; wherein: Parameter grouping processing unit: used to receive the multi-parameter data output by the sensor network deployment module, and classify the vibration acceleration data, pore water pressure data, soil displacement data and pipeline strain data according to the data source; Sliding window construction unit: used to construct a sliding time window of fixed time length for each type of parameter data. The window length is set to T seconds, the step length is Δt seconds, and the continuous historical data sequence in the current window is extracted at each moment; Abnormal filtering calculation unit: used to perform steady-state volatility analysis on historical data in each sliding window, remove data points whose amplitude exceeds the set standard deviation multiple, and use the median filtering method to generate the stable interval of the parameter in the current window; Dynamic benchmark generation unit: used to define the upper and lower bounds of the stability interval of each type of parameter in the sliding window as the dynamic baseline range of the corresponding parameter after completing the abnormal filtering, and output the mean of the range as the real-time dynamic benchmark data at the current moment.

4. The sensor network-based real-time monitoring system for soil liquefaction and slippage of submarine pipelines according to claim 3 is characterized in that: The abnormal filtering calculation unit comprises: Volatility evaluation subunit: used to evaluate the volatility of the received historical data sequence in each sliding window and calculate the mean M and standard deviation S of the sequence; Outlier elimination subunit: used to eliminate outliers that meet the conditions according to the set deviation threshold multiple A. Abnormal data points are generated to generate the data set after elimination; Median filter subunit: used to perform median filtering on the data set after elimination, calculate the median value P of the sorted sequence as the reference value of the current parameter steady-state level, and extract the upper and lower quartiles at the same time , , determine the stability interval .

5. The sensor network-based real-time monitoring system for soil liquefaction and slippage of submarine pipelines according to claim 1 is characterized in that: The multi-source data spatiotemporal calibration module includes a time alignment unit, a spatial position registration unit and an interpolation calculation unit; wherein: Time alignment unit: used to receive the real-time dynamic baseline data of each sensor node output by the dynamic baseline analysis module, extract the corresponding timestamp according to the system master clock signal, uniformly reconstruct the data time axis based on the unified reference time, remove the data frames with time drift, and ensure that all node data are comparable at the same time point; Spatial position registration unit: used to construct a sensor spatial position mapping table according to the preset geographic coordinates or relative position labels of each sensor node when it is arranged around the pipeline, and to spatially index the data according to the distribution relationship of the nodes in the three-dimensional coordinate system; Interpolation calculation unit: It is used to perform numerical interpolation reconstruction on missing or abnormal data nodes based on the dynamic reference values ​​of their spatially adjacent nodes on the basis of completing time alignment and spatial position registration. At the same time, it fine-tunes the adjacent node data with consistent time series change trends to generate time-space calibration data.

6. The sensor network-based real-time monitoring system for soil liquefaction and slippage of submarine pipelines according to claim 5 is characterized in that: The spatial position registration unit comprises: Node information extraction subunit: used to extract the unique identifier of each sensor node and its initial position information recorded during deployment from the sensor network deployment module, including the pipeline segment number, circumferential angle and installation depth; 3D coordinate modeling subunit: used to calculate the rectangular coordinates (X, Y, Z) of each node in space according to the extracted pipeline segment number, circumferential angle and installation depth, combined with the 3D trajectory model of the pipeline centerline, and store it in the spatial position mapping table of the corresponding sensor; Spatial index encoding subunit: used to perform spatial hierarchical encoding of all nodes according to the node distribution in the three-dimensional coordinate system, and to construct a fast spatial query index structure using a hash method.

7. The real-time monitoring system for soil liquefaction and slippage of submarine pipelines based on a sensor network according to claim 6 is characterized in that: The interpolation calculation unit comprises: Neighboring node retrieval subunit: used to retrieve the neighboring node set of the target missing node in the three-dimensional coordinate space from the spatial position mapping table, use a fixed search radius R to perform spherical neighborhood matching, and output all nodes whose spatial distance to the target node is not greater than R as the interpolation reference set, recorded as the node set N; Spatial interpolation reconstruction subunit: used to perform weighted averaging of the current dynamic reference value of each node in the node set N according to its Euclidean distance to the target node, and generate the interpolation estimate of the target node ; Time series trend reconciliation subunit: used to estimate the value after spatial interpolation To fine-tune the time trend, first calculate the previous moment baseline value sequence of the neighboring nodes With the current sequence The mean change difference , and then the difference Weighted effect , generating the final output spatiotemporal calibration data , and its calculation formula is: ,in, is the time series trend harmonic coefficient, which represents the harmonic weight factor.

8. The sensor network-based real-time monitoring system for soil liquefaction and slippage of submarine pipelines according to claim 1 is characterized in that: The coupling risk assessment module includes a normalization unit, a weight fusion unit and a risk generation unit; wherein: Parameter standardization unit: used to receive the vibration acceleration data A, pore water pressure data P, soil displacement data D and pipeline strain data E output by the time-space calibration module, and standardize them to the interval [0,1] using the maximum and minimum value normalization method to obtain the standardized parameters , , , , to eliminate the scale differences between different physical quantities; Weight allocation unit: used to set the fusion weight of each type of standardized parameter according to the historical slip event analysis and regional geological weight factor. The weight set is , which correspond to the influence of vibration acceleration, pore water pressure, soil displacement and pipeline strain, and meet the constraints: ; Risk coefficient calculation unit: Based on the standardized parameters and weight set, the comprehensive soil liquefaction slip risk coefficient R is calculated. The calculation formula is: .

9. The real-time monitoring system for soil liquefaction and slippage of submarine pipelines based on a sensor network according to claim 1 is characterized in that: The hierarchical warning module includes a threshold judgment unit, a warning message generation unit and a signal transmission unit; wherein: Threshold judgment unit: used to receive the soil liquefaction slip risk coefficient output by the coupled risk assessment module and compare it with three sets of pre-set risk coefficient thresholds; when the third-level warning is triggered, when the second-level warning is triggered, when the first-level warning is triggered, which correspond to the risk thresholds of local slip, regional liquefaction and pipeline instability respectively; Early warning message generation unit: constructs a structured early warning message based on the triggered early warning level; Signal transmission unit: used to transmit structured early warning messages to the monitoring terminal through wired or wireless communication networks.

10. The real-time monitoring system for soil liquefaction and slippage of submarine pipelines based on a sensor network according to claim 9, characterized in that: The early warning message generating unit comprises: Warning level mapping subunit: used to determine the corresponding warning level identification code according to the warning level triggered by the risk level result output by the threshold judgment unit, where the first-level warning is mapped to W1, representing local slip; the second-level warning is mapped to W2, representing regional liquefaction; the third-level warning is mapped to W3, representing pipeline instability; Message field filling unit: used to construct structured content containing the following fields for each triggered warning event: 1) Warning level identification; 2) Trigger time; 3) Coordinates of the early warning center; 4) Corresponding risk factor value; 5) Proposed disposal instructions; Structured format encoding subunit: used to encode the constructed field content according to the preset format to generate a unified message format.

Images