Real-time Monitoring System for Soil Liquefaction and Slip of Submarine Pipelines Based on Sensor Networks

Abstract

The present invention relates to the technical field of offshore engineering safety monitoring, and particularly relates to a real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network, which 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; wherein: The sensor network deployment module: is used to collect multi-parameter data; The dynamic baseline analysis module: calculates the dynamic baseline range of each parameter; The multi-source data spatio-temporal calibration module: eliminates the spatio-temporal deviation between sensor nodes and outputs spatio-temporally calibrated data; The coupled risk assessment module: calculates the soil liquefaction and slip risk coefficient; The hierarchical early warning module: is used to trigger an early warning. With the present invention, through the combination of multi-parameter perception, spatio-temporal calibration, and risk fusion assessment, accurate monitoring and hierarchical early warning of submarine pipeline soil liquefaction and slip are achieved, improving the continuity of monitoring and the timeliness of early warning.

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 disturbances, tidal scour, and changes in sediment layer structure, 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 sensing capabilities 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 monitor data points, lacking a comprehensive analysis mechanism for vibration, pressure, displacement, and strain, and are easily interfered by the submarine environment during 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, and it is 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:

[0006] The 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;

[0007] The 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;

[0008] Multi-source data spatio-temporal calibration module: 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;

[0009] Coupled risk assessment module: It is used to receive the spatio-temporally calibrated 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 landslide risk coefficient;

[0010] Hierarchical warning module: It is used to receive the soil liquefaction and landslide risk coefficient output by the coupled risk assessment module, trigger a first-level warning, a second-level warning or a third-level warning according to a preset threshold, and output a warning signal to the monitoring terminal.

[0011] Optionally, 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:

[0012] 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°, 270° azimuths of the pipeline circumference;

[0013] 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;

[0014] Data preprocessing unit: It is used to perform time-domain filtering, dimension 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;

[0015] Time synchronization unit: It is 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 groups of data are consistent.

[0016] Optionally, the dynamic baseline analysis module includes a parameter grouping processing unit, a sliding window construction unit, an anomaly filtering calculation unit and a dynamic reference generation unit; among them:

[0017] Parameter grouping processing unit: It is 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;

[0018] Sliding window construction unit: used to construct a sliding time window with a fixed time length for each type of parameter data. The window length is set to T seconds, and the step size is Δt seconds. At each moment, a continuous historical data sequence within the current window is extracted;

[0019] Abnormal filtering calculation unit: used to perform steady-state volatility analysis on historical data within each sliding window, eliminate data points whose amplitudes exceed a set multiple of the standard deviation, and use the median filtering method to generate a stable interval of the parameter within the current window;

[0020] Dynamic baseline generation unit: used to, after completing abnormal filtering, define 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 output the mean value of this range as the real-time dynamic baseline data at the current moment.

[0021] Optionally, the abnormal filtering calculation unit includes:

[0022] Volatility evaluation sub-unit: used to perform volatility evaluation on the received historical data sequence within each sliding window, and calculate the mean value M and standard deviation S of the sequence;

[0023] Abnormal point elimination sub-unit: used to eliminate abnormal data points that meet the condition according to the set deviation threshold multiple A, and generate a data set after elimination;

[0024] Median filtering sub-unit: 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 steady-state level of the current parameter, and at the same time extract the upper and lower quartiles , , and determine the stable interval .

[0025] 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:

[0026] 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 timestamps according to the system master clock signal, and uniformly reconstruct a data time axis based on a unified reference time, and eliminate data frames with time drift to ensure that data from all nodes is comparable at the same time point;

[0027] Spatial position registration unit: 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 around the pipeline, and perform spatial indexing on the data according to the distribution relationship of the nodes in the three-dimensional coordinate system;

[0028] Interpolation calculation unit: Based on the completion of time alignment and spatial position registration, for missing or abnormal data nodes, numerical interpolation reconstruction is performed according to the dynamic reference values of their spatially adjacent nodes, and at the same time, the data of adjacent nodes with consistent temporal change trends are fine-tuned to generate spatio-temporal calibration data.

[0029] Optionally, the spatial position registration unit includes:

[0030] Node information extraction sub-unit: 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 pipeline segment number, circumferential angle, and installation depth;

[0031] Three-dimensional coordinate modeling sub-unit: 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 three-dimensional trajectory model of the pipeline centerline, and store them in the spatial position mapping table of the corresponding sensor;

[0032] Spatial index coding sub-unit: 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.

[0033] Optionally, the interpolation calculation unit includes:

[0034] Adjacent node retrieval sub-unit: 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 distance from the target node in space is not greater than R as the interpolation reference set, denoted as node set N;

[0035] 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 distance from the target node to generate the interpolation estimated value of the target node ;

[0036] 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 adjacent 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.

[0037] Optionally, the coupling risk assessment module includes a normalization unit, a weight fusion unit, and a risk generation unit; specifically:

[0038] Parameter standardization unit: It is 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 normalize them to the interval [0, 1] respectively by using the maximum-minimum normalization method to obtain the standardized parameters , , , , so as to eliminate the scale differences between different physical quantities;

[0039] Weight allocation unit: It is used to set the fusion weights of each type of standardized parameter according to the historical slip event analysis and the regional geological weight factor. The set weights are , corresponding to the influence degrees of vibration acceleration, pore water pressure, soil displacement, and pipeline strain respectively, and satisfying the constraint condition: ;

[0040] Risk coefficient calculation unit: Based on the standardized parameters and the weight set, calculate the comprehensive soil liquefaction and slip risk coefficient R, and its calculation formula is: .

[0041] Optionally, the hierarchical early warning module includes a threshold judgment unit, an early warning message generation unit, and a signal transmission unit; specifically:

[0042] Threshold judgment unit: It is used to receive the soil liquefaction and slip risk coefficient output by the coupling 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;

[0043] Early warning message generation unit: Construct a structured early warning message based on the triggered early warning level;

[0044] Signal transmission unit: It is used to transmit the structured early warning message to the monitoring terminal through a wired or wireless communication network.

[0045] Optionally, the early warning message generation unit includes:

[0046] Early warning level mapping sub-unit: It is 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, the level-one early warning is mapped to W1, representing local slip; the level-two early warning is mapped to W2, representing regional liquefaction; the level-three early warning is mapped to W3, representing pipeline instability;

[0047] Message field filling unit: For each triggered warning event, it is used to construct structured content including the following fields:

[0048] 1. Warning level identifier;

[0049] 2. Trigger time;

[0050] 3. Warning center coordinates;

[0051] 4. Corresponding risk coefficient value;

[0052] 5. Suggested handling instructions;

[0053] Structured format encoding subunit: It is used to encode the constructed field content according to a preset format to generate a unified message format.

[0054] Advantages of the present invention:

[0055] 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 monitoring data is achieved, significantly improving the accuracy and continuity of slip risk identification, and solving the problems of parameter drift and abnormal point interference in the complex seabed environment.

[0056] In the present invention, by constructing a coupled risk assessment mechanism, multiple parameters are fused and calculated according to weights to obtain the liquefaction slip risk coefficient, and combined with a preset threshold to perform hierarchical early warning, and automatically push the structured message to the monitoring terminal, effectively realizing rapid response to different slip severities and output of handling suggestions, and improving the safety guarantee ability and early warning intelligence level during the whole life cycle operation of the submarine pipeline. Description of the drawings

[0057] 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.

[0058] Figure 1 It is a schematic diagram of the real-time monitoring system for soil liquefaction and slip in the embodiment of the present invention;

[0059] Figure 2 It is a schematic diagram of the multi-source data spatio-temporal calibration module in the embodiment of the present invention. Detailed implementation manners

[0060] The present invention will be described in detail below with reference to the accompanying 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; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0061] As Figure 1 - Figure 2 shown, 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 multi-source data spatio-temporal calibration module, a coupling risk assessment module, and a hierarchical early warning module; where:

[0062] The sensor network deployment module: Multiple groups of sensor nodes are arranged circumferentially along the submarine pipeline, used to collect soil vibration acceleration data, pore water pressure data, soil body displacement data, and pipeline strain data, and output multi-parameter data.

[0063] The 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.

[0064] The 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-temporally calibrated data.

[0065] The coupling risk assessment module: Used to receive the spatio-temporally calibrated data output by the spatio-temporal calibration module, fuse the vibration acceleration, pore water pressure, soil body displacement, and pipeline strain data according to weights, and generate a soil liquefaction and slip risk coefficient.

[0066] The hierarchical early warning module: Used to receive the soil liquefaction and slip risk coefficient output by the coupling risk assessment module, trigger a first-level early warning (local slip), a second-level early warning (regional liquefaction), or a third-level early warning (pipeline instability) according to a preset threshold, and output an early warning signal to the monitoring terminal.

[0067] The sensor network deployment module includes a circumferential layout unit, a multi-parameter acquisition unit, a data preprocessing unit, and a time synchronization unit; where:

[0068] Circumferential Arrangement Unit: It is used to set a group of sensor nodes at intervals of 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 azimuths of 0°, 90°, 180°, and 270° in the circumferential direction of the pipeline. The sensor nodes are connected to the outer wall of the pipeline through a hinged fixing ring. 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 an anchor rod to ensure that the sensors deform synchronously with the pipeline and the surrounding soil.

[0069] 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.

[0070] The multi-parameter acquisition unit includes:

[0071] 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. The measuring range is ±5g, and the sampling frequency is 100Hz. It is used to collect vibration acceleration data at the pipeline-soil interface.

[0072] 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. The sampling frequency is 1Hz, and it measures the pore water pressure data during the soil liquefaction process.

[0073] 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°. The measuring range is 0 - 30 cm, and it obtains the circumferential soil displacement data of the pipeline in real time.

[0074] 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 measuring point is arranged every 1 meter to monitor the pipeline strain data.

[0075] Data Preprocessing Unit: It is used to perform time-domain filtering, dimension normalization, and outlier rejection on the collected soil vibration acceleration data, pore water pressure data, soil displacement data, and pipeline strain data to generate multi-parameter data.

[0076] 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 timestamps of multiple groups of data are consistent.

[0077] The dynamic baseline analysis module includes a parameter grouping processing unit, a sliding window construction unit, an anomaly filtering calculation unit, and a dynamic baseline generation unit; among them:

[0078] 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;

[0079] 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;

[0080] 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;

[0081] 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.

[0082] The abnormal filtering calculation unit includes:

[0083] 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;

[0084] 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;

[0085] 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.

[0086] 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:

[0087] The time alignment unit: 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 master clock signal, uniformly reconstruct the data time axis based on the unified reference time, and eliminate the data frames with time drift, ensuring that the data of all nodes is comparable at the same time point;

[0088] The spatial position registration unit: 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;

[0089] The interpolation calculation unit: is used to, on the basis of completing time alignment and spatial position registration, perform numerical interpolation reconstruction for missing or abnormal data nodes according to the dynamic reference values of their spatially adjacent nodes, and at the same time fine-tune 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, the problem of data inconsistency caused by synchronization errors and insufficient deployment density in the subsea sensor network can be effectively solved, and the spatial cooperation ability and time accuracy consistency of multi-source parameters in subsequent fusion calculations can be improved.

[0090] The spatial position registration unit includes:

[0091] The node information extraction subunit: 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;

[0092] The three-dimensional coordinate modeling subunit: 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, in combination with the three-dimensional trajectory model of the pipeline center line, and store them in the spatial position mapping table corresponding to the sensor;

[0093] The specific coordinate calculation formula is as follows:

[0094] ;

[0095] ;

[0096] ; where , , represents the central coordinates of the starting point of the current pipeline segment, which are preset by the pipeline layout path; L represents the longitudinal distance (along the centerline direction) of the pipeline segment where the current node is located; represents the inclination angle of the extension direction of the current segment on the horizontal plane. If it is a horizontal segment, then = 0; R represents the radius distance for installing the sensor node, that is, the pipeline radius, which is set as a constant; represents the circumferential angle of the sensor node relative to the center of the pipeline cross-section, with the unit of degree; D represents the installation depth of the sensor node vertically downward (unit: meter); X, Y, Z: represent the finally converted three-dimensional space rectangular coordinates.

[0097] Spatial index coding subunit: used to hierarchically code 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 the interpolation operation; by standardizing the sensor installation information into the form of three-dimensional coordinates and introducing the 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 calculation, and providing spatial continuity guarantee for the subsequent slip risk assessment.

[0098] Table 1 Example of sensor spatial position mapping structure

[0099] 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

[0100] In the above Table 1, the node number is the unique number assigned to each sensor node; the pipeline segment number represents the pipeline segment number where the sensor is located; the circumferential angle represents the angle of the node relative to the center of the pipeline cross-section; the installation depth represents the depth of the node 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, which are used for subsequent interpolation calculation.

[0101] The interpolation calculation unit includes:

[0102] 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 node set N;

[0103] Spatial interpolation reconstruction subunit: used to perform weighted average on the current moment dynamic reference values of each node in 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 neighboring node, represents its Euclidean distance from the target node, and k is the number of neighboring nodes;

[0104] Temporal trend harmonization subunit: Used to perform time trend fine-tuning on the estimated value after spatial interpolation First, calculate the sequence of reference values at the previous moment of neighboring nodes and the current sequence The mean change difference , and then apply the difference by weighting to , generating the final output spatio-temporal calibration data , and its calculation formula is: , where is the temporal trend harmonization coefficient, representing the harmonization weight factor; The above subunit first realizes interpolation reconstruction through distance weighted average based on spatial neighboring nodes, and then combines time series trends for trend fine-tuning, which can effectively make up for data missing problems caused by node failures or signal interference, ensuring the continuity and consistency of calibration data in both time and space dimensions, thereby guaranteeing the accuracy and stability of subsequent coupling evaluation results.

[0105] The coupling risk assessment module includes a normalization unit, a weight fusion unit, and a risk generation unit; Among them:

[0106] 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;

[0107] 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 condition: ;

[0108] The method steps for weight setting in the weight allocation unit are as follows:

[0109] 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 ; Then construct the historical response matrix H and the observation result vector , and their expressions are respectively:

[0110] ;

[0111] ;

[0112] Step 2: Adopt the linear least squares method to inversely solve the optimal weight vector , so that the predicted risk coefficient approximates the observed value as much as possible ; then solve the objective function: ; the constraint condition is: ; and then obtain a uniquely determined set of fusion weights , which is used for subsequent real-time monitoring scenarios;

[0113] Risk coefficient calculation unit: Based on the standardized parameters and the weight set, calculate the comprehensive soil liquefaction and sliding risk coefficient R, and its calculation formula is: ; where, R represents the liquefaction and sliding 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 of standardization and weighted fusion, multi-source heterogeneous parameters can be uniformly incorporated into a unified evaluation framework, improving the objectivity and real-time nature of soil liquefaction and sliding risk judgment, and having good regional adaptability, providing a quantitative input basis for downstream hierarchical early warning.

[0114] The hierarchical early warning module includes a threshold judgment unit, an early warning message generation unit and a signal transmission unit; among them:

[0115] Threshold judgment unit: Used to receive the soil liquefaction and sliding risk coefficient output by the coupled risk assessment module, and compare it with three groups of pre-set risk coefficient thresholds; trigger a level 3 early warning when, trigger a level 2 early warning when, and trigger a level 1 early warning when, where respectively correspond to the risk thresholds of local sliding, regional liquefaction and pipeline instability;

[0116] Early warning message generation unit: Construct a structured early warning message based on the triggered early warning level. The message includes information such as the early warning level identifier, generation timestamp and early warning area location, etc., so that the monitoring terminal can accurately parse it;

[0117] 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 immediate notification of sliding risks of different severity levels are achieved, improving the response speed and early warning accuracy of the system to submarine pipeline soil liquefaction and sliding events.

[0118] The early warning message generation unit includes:

[0119] Early warning level mapping subunit: 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 first-level early warning is mapped to W1, representing local slip; the second-level early warning is mapped to W2, representing regional liquefaction; the third-level early warning is mapped to W3, representing pipeline instability;

[0120] Report text field filling unit: used to construct structured content containing the following fields for each triggered early warning event:

[0121] 1. Early warning level identification (W1 / W2 / W3);

[0122] 2. Trigger time (system standard timestamp);

[0123] 3. Coordinates of the early warning center (represented by the node number and its three-dimensional position of the trigger point);

[0124] 4. Corresponding risk coefficient value R;

[0125] 5. Suggested handling instructions (attached operation instruction number according to the early warning level);

[0126] Structured format encoding subunit: used to encode the constructed field content according to a preset format to generate a unified message format;

[0127] The specific example is as follows:

[0128] [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, which is convenient for subsequent parsing and execution; by standardizing the mapping and encoding of the early warning level and the structured content fields, the unified expression and rapid transmission of early warning information are realized, which not only improves the message parsing efficiency, but also ensures the accurate correspondence of response measures under different early warning levels, enhancing the response standardization and automation level of the system to sudden slip events.

[0129] The present invention covers any substitutes, modifications, equivalent methods and solutions made on the essence and scope of the present invention. In order 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 the description of these details. In addition, in order to avoid unnecessary confusion to the essence of the present invention, well-known methods, processes, flows, components and circuits are not described in detail.

[0130] 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 submarine pipeline soil liquefaction and slip based on a sensor network, characterized in that, It includes a sensor network deployment module, a dynamic baseline analysis module, a multi-source data spatio-temporal calibration module, a coupling risk assessment module, and a hierarchical warning module; among which: Sensor network deployment module: Multiple 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. The dynamic baseline analysis module includes a parameter grouping processing unit, a sliding window construction unit, an anomaly filtering calculation unit, and a dynamic reference generation unit; among which: Parameter grouping processing unit: It is 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 body displacement data, and pipeline strain data respectively according to the data source; Sliding window construction unit: It is used to construct a sliding time window with a fixed time length for each type of parameter data. The window length is set to T seconds, and the step size is Δt seconds. At each moment, the continuous historical data sequence within the current window is extracted; Anomaly filtering calculation unit: It is used to perform steady-state volatility analysis on the historical data within each sliding window, eliminate the data points whose amplitudes exceed the set standard deviation multiple, and use the median filtering method to generate the stable interval of the parameter within the current window; Dynamic reference generation unit: It is used to, after completing anomaly filtering, define 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 output the mean value of this range as the real-time dynamic reference data at the current moment; Multi-source data spatio-temporal calibration module: 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: It is used to receive the spatio-temporally calibrated data output by the spatio-temporal calibration module, fuse the vibration acceleration, pore water pressure, soil body displacement, and pipeline strain data according to weights, and generate a soil liquefaction and landslide risk coefficient; Hierarchical warning module: It is used to receive the soil liquefaction and landslide risk coefficient output by the coupling risk assessment module, trigger a first-level warning, a second-level warning, or a third-level warning according to a preset threshold, and output a warning signal to the monitoring terminal; The multi-source data spatio-temporal calibration module includes a time alignment unit, a spatial position registration unit, and an interpolation calculation unit; among which: 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 master clock signal, and uniformly reconstruct the data time axis based on a unified reference time, eliminating the data frames with time drift, so as to ensure the comparability of all node data 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: Based on the completion of time alignment and spatial position registration, for missing or abnormal data nodes, numerical interpolation reconstruction is performed according to the dynamic reference values of their spatially adjacent nodes, and at the same time, the data of adjacent nodes with consistent temporal variation trends are fine-tuned to generate spatio-temporal calibration data; The spatial position registration unit includes: Node information extraction sub-unit: 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 pipeline section number, circumferential angle, and installation depth; Three-dimensional coordinate modeling sub-unit: 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, in combination with the three-dimensional 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 according to the node distribution in the three-dimensional coordinate system, and construct a fast spatial query index structure by using the hash method; The interpolation calculation unit includes: Adjacent node retrieval sub-unit: 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 with a fixed search radius R, and output all nodes whose distance from the target node in space is not greater than R as the interpolation reference set, denoted as node set N; Spatial interpolation reconstruction subunit: used to perform weighted averaging on the current moment dynamic reference values of each node in the node set N according to their Euclidean distances from the target node to generate an interpolation estimated value of the target node ; Temporal Trend Reconciliation Subunit: used for the estimated value after spatial interpolation to perform fine-tuning of the time trend. First, calculate the reference value sequence at the previous moment of adjacent nodes and the current sequence to obtain the mean change difference . Then, weight the difference and apply it to to generate the final output spatio-temporal calibration data . Its calculation formula is: , where is the temporal trend reconciliation coefficient, representing the reconciliation weight factor; The coupling risk assessment module includes a normalization unit, a weight fusion unit, and a risk generation unit; among them: Parameter normalization unit: It is 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 normalize them to the interval [0, 1] respectively by using the maximum-minimum normalization method to obtain the normalized parameters , , , , so as to eliminate the scale differences between different physical quantities; Weight distribution unit: used to set the fusion weight of each type of standardized parameter according to the analysis of historical slip events 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: ; 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: .

2. The real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network according to claim 1, characterized in that 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: 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; 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 groups of data are consistent.

3. The real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network according to claim 1, wherein The anomaly filtering calculation unit includes: Volatility assessment sub-unit: Used to perform volatility assessment on the received historical data sequence in each sliding window, and calculate the mean M and standard deviation S of the sequence; Abnormal point rejection sub-unit: used to reject abnormal data points that meet the conditions according to the set deviation threshold multiple A, and generate a data set after rejection; ​ Median Zebra Unit: It is used to perform median filtering on the data set after rejection, calculate the median value P of the sorted sequence as the reference value of the current parameter steady state level, and simultaneously extract the upper and lower quartiles , , determine the stable interval .

4. The real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network according to claim 1, characterized in that 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: It is used to receive the soil liquefaction and landslide risk coefficient output by the coupling risk assessment module and compare it with three sets of pre-set risk coefficient thresholds; trigger a level-three warning when, trigger a level-two warning when, and trigger a level-one warning when, where respectively correspond to the risk thresholds of local landslide, regional liquefaction, and pipeline instability; Warning message generation unit: Construct a structured warning message based on the triggered warning level; Signal transmission unit: It is used to transmit the structured warning message to the monitoring terminal through a wired or wireless communication network.

5. The real-time monitoring system for submarine pipeline soil liquefaction and slip based on a sensor network according to claim 4, characterized in that, The warning message generation unit includes: Warning level mapping sub-unit: It is 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 level-one warning is mapped to W1, representing local landslide; the level-two warning is mapped to W2, representing regional liquefaction; the level-three warning is mapped to W3, representing pipeline instability; Message field filling unit: It is 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 warning center; 4. Corresponding risk coefficient value; 5. Suggested disposal instructions; Structured format encoding sub-unit: It is used to encode the constructed field content according to a preset format to generate a unified message format.

Images