A loRa-based submarine cable state remote wireless telemetry system and method

Abstract

The application discloses a kind of based on LoRa's sea cable state remote wireless telemetry system and method, belong to cable laying state monitoring technical field, system includes being laid on several state monitoring nodes of sea cable, relay node and shipborne monitoring center.Adaptive different compensation strategies are taken according to the proportion of missing packet loss monitored by analysis computer: when mild packet loss, trend filling is executed using effective node change rate and spatial correlation;When moderate packet loss, confidence weight that attenuates dynamically with interruption time is introduced in reconstruction model;When severe packet loss, spatial physical correlation constraint matrix feature sub-block is extracted to perform order reduction calculation and combine spatial interpolation method to recover data.The application solves the problem of data loss caused by unstable wireless communication in extreme sea conditions through hierarchical compensation mechanism, ensuring that the full-line state distribution of sea cable can be reconstructed stably and accurately and the warning can be output under different packet loss conditions, significantly improving the robustness and construction safety of the sea cable monitoring system.

Description

Technical Field

[0001] This invention relates to the technical field of cable laying status monitoring, and in particular to a LoRa-based remote wireless telemetry system and method for submarine cable status. Background Technology

[0002] With the rapid development of offshore wind power, cross-sea power transmission, submarine communication and other projects, the scale of submarine cable laying is constantly expanding and the laying environment is becoming increasingly complex. During the laying process, submarine cables are subjected to complex mechanical loads, including gravity, buoyancy, ocean currents, dynamic loads caused by ship movement, and seabed contact forces. Excessive stress can lead to damage to the cable insulation layer, breakage of the armor layer, and even laying accidents.

[0003] In existing technologies, monitoring the status of submarine cable laying mainly employs a wired sensor solution, which involves pre-embedding sensors inside the submarine cable and transmitting data to a shipborne acquisition system via signal cables. However, this solution has significant technical drawbacks: submarine cables can be several kilometers long, making the cabling process extremely complex, and the signal cables are susceptible to mechanical damage during laying, leading to monitoring failure; furthermore, long-distance wired transmission suffers from severe signal attenuation, increasing system complexity and cost; additionally, once the sensor node locations are determined, they are difficult to adjust, resulting in poor deployment flexibility.

[0004] In recent years, the development of wireless sensor network technology has provided new ideas for this, especially the emergence of LoRa wireless spread spectrum technology, which has made long-distance, low-power wireless telemetry possible. However, when applying LoRa technology for submarine cable status monitoring in actual marine environments, wireless communication is prone to packet loss or temporary offline status of monitoring nodes due to extreme sea conditions (such as wave surging, ship hull obstruction, multipath effects, etc.), resulting in incomplete local status data collected in real time. Furthermore, this incomplete data causes the solution process based on the spatial continuity physical correlation equations to diverge or even collapse, leading to significant state reconstruction errors.

[0005] Therefore, overcoming the data loss problem caused by LoRa communication instability under extreme sea conditions and building a highly robust monitoring system that can adaptively perform data trend filling, dynamic confidence weight adjustment, and reduced-order matrix calculation according to different severity levels of communication packet loss, so as to reconstruct the status distribution of the entire submarine cable stably and accurately and provide reliable early warning even when the data is incomplete, is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] To address the technical problem that existing wireless communication-based submarine cable monitoring systems are prone to LoRa communication packet loss or node offline issues under extreme sea conditions, resulting in incomplete local state data and causing the solution of the global state reconstruction model based on spatial physical correlation to diverge or fail, thus making it impossible to stably and accurately reconstruct the state of the entire submarine cable and provide reliable early warning under harsh communication conditions with missing data, this invention provides a LoRa-based remote wireless telemetry system and method for submarine cable status.

[0007] According to a first aspect of the present invention, a LoRa-based remote wireless telemetry system for submarine cable status is proposed, comprising:

[0008] Several status monitoring nodes are deployed at intervals on the submarine cable to be laid, configured to collect local status data of the submarine cable and transmit it via a wireless communication module;

[0009] At least one relay node is deployed on the wireless transmission path and configured to receive local status data and route it to the shipborne monitoring center.

[0010] The shipborne monitoring center includes a gateway and an analysis computer connected to the gateway. The analysis computer is configured to monitor the communication link status in real time to identify the set of valid nodes that actually receive data and the set of missing nodes that have experienced packet loss. It extracts the spatial physical correlation constraint matrix of the submarine cable based on all status monitoring nodes, calculates the missing proportion of missing nodes, and compares the missing proportion with a preset first packet loss threshold and a second packet loss threshold, wherein the first packet loss threshold is less than the second packet loss threshold. If the missing proportion is below the first packet loss threshold, it uses the current state change rate and spatial correlation of the local state data corresponding to the valid node set, combined with the historical local state data of the missing nodes, to perform trend filling on the local state data corresponding to the missing nodes to construct a complete equivalent state vector. It then uses the spatial physical correlation constraint matrix of the submarine cable to construct a state reconstruction model for global solution. If the missing proportion is greater than the first packet loss threshold but less than the second packet loss threshold, it introduces a confidence weight that dynamically decays with the communication interruption time into the local state data after missing node filling in the state reconstruction model.

[0011] If the missing proportion is greater than the second packet loss threshold, feature sub-blocks corresponding to the set of valid nodes in the spatial physical association constraint matrix are extracted for order reduction calculation, and the local state data corresponding to the missing nodes is restored by distance-based spatial interpolation method, so as to combine and obtain the complete global node state feature vector.

[0012] Based on the acquired global node state feature vectors, the state distribution of the entire submarine cable is reconstructed, and hierarchical early warnings are output.

[0013] In some specific embodiments, the analysis computer extracts the spatial physical correlation constraint matrix of the submarine cable based on all state monitoring nodes. Specifically, the configuration involves discretizing the submarine cable using the finite segment method. There are 1 unit, and each unit corresponds to a status monitoring node. The 1st unit is calculated based on the spatial continuity and physical correlation. The axial force and bending moment of each element are expressed as follows:

[0014] ;

[0015] ;

[0016] in, , Indicates the first Axial force of each unit Indicates the first The bending moment of each element, This represents the elastic modulus of the submarine cable material. This represents the cross-sectional area of ​​the submarine cable. The moment of inertia of the submarine cable section is represented by the moment of inertia of the submarine cable section. Indicates the unit length. and They represent the first The and the first Axial displacement of each status monitoring node and They represent the first The and the first The bending angle of each condition monitoring node; extracting the axial force calculation formula. With the bending moment calculation formula As unit characteristic coefficients, they are assembled according to the corresponding state monitoring nodes to form a spatial physical correlation constraint matrix.

[0017] In some specific embodiments, the analysis computer utilizes the current state change rate and spatial correlation of the local state data corresponding to the effective node set, combined with the historical local state data of the missing nodes, to perform trend imputation on the local state data corresponding to the missing nodes, in order to construct a complete equivalent state vector. The specific configuration is as follows:

[0018] If historical data exists for a missing node, the data is filled in by combining the state change trend of the valid node set with spatial correlation, as shown below:

[0019] ;

[0020] in, This indicates the existence of nodes with missing historical data. Real-time local state data corresponding to the current moment. Indicates missing nodes Historical local state data corresponding to the previous moment, Indicates the sampling interval time. The length of the time window used for smoothing trend estimation. Represents the set of valid nodes. Represents the set of valid nodes Valid nodes in Indicates a valid node The corresponding rate of change of the current state of the local state data. Indicates missing nodes With valid nodes Spatial correlation weighting coefficients between them;

[0021] If a missing node has no historical data, the data is filled in using the global average trend, as shown below:

[0022] ;

[0023] in, Indicates a missing node with no historical data. Real-time local state data corresponding to the current moment. This represents the global average of the historical local state data corresponding to all valid nodes at the previous time step. Represents the set of valid nodes. Represents the set of valid nodes Valid nodes in Indicates a valid node The corresponding rate of change of the current state of the local state data. Indicates the sampling interval time;

[0024] And valid nodes The current state change rate is based on the effective nodes. The real-time local state data corresponding to the current moment and the historical local state data corresponding to the previous moment are calculated and obtained, and are represented as follows: ; Indicates a valid node Real-time local state data corresponding to the current moment. Indicates a valid node Historical local state data corresponding to the previous moment

[0025] Integrate nodes with missing historical data The current real-time local state data and / or missing nodes with no historical data at the current moment. The local state data corresponding to the current moment is combined with the real-time local state data of each valid node in the set of valid nodes to construct a complete equivalent state vector.

[0026] In some specific embodiments, the analysis computer calculates missing nodes according to the following formula. With valid nodes The spatial correlation weighting coefficient between them is expressed as:

[0027] ;

[0028] in, Indicates missing nodes With valid nodes Spatial correlation weighting coefficient between them Indicates missing nodes With valid nodes The covariance of the corresponding historical local state data, and These represent missing nodes. With valid nodes The variance of the corresponding historical local state data.

[0029] In some specific embodiments, the analysis computer constructs a state reconstruction model by combining the spatial physical correlation constraint matrix of the submarine cable, and introduces a confidence weight that dynamically decays with the communication interruption time into the local state data after filling in the missing nodes in the state reconstruction model. The specific configuration is as follows:

[0030] When the missing rate is below the first packet loss threshold, a state reconstruction model is constructed by combining the spatial physical correlation constraint matrix of the submarine cable, which is expressed as:

[0031] ;

[0032] And when the missing percentage is greater than the first packet loss threshold but less than the second packet loss threshold, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after the missing nodes are filled in the state reconstruction model, which is expressed as:

[0033] ;

[0034] in, and Let represent the basic reconstruction objective function and the reconstruction objective function with confidence weights, respectively. Represents the set of valid nodes. Represents the set of missing nodes; Represents the set of valid nodes Valid nodes in Represents the set of missing nodes Missing nodes in; Represents the effective nodes in the complete equivalent state vector. The corresponding real-time local state data at the current moment. This represents the valid nodes obtained based on the global node state feature vector to be determined. The model calculates the state function value; Represents missing nodes in the complete equivalent state vector The local state data after filling in the gaps This represents the missing node obtained based on the feature vector of the global node state to be determined. The model calculates the state function value; Indicates a valid node The initial weighting coefficients, Indicates missing nodes Introduced confidence weights; This represents the first-order smoothing regularization parameter. This indicates the total length of the submarine cable to be laid. This indicates the axial displacement of the submarine cable. This represents the second-order smoothing regularization parameter. Indicates the bend angle of the submarine cable. The coordinates represent the arc length of the submarine cable.

[0035] In some specific embodiments, the analysis computer calculates the filled missing nodes according to the following formula. The confidence weights introduced dynamically decay with the duration of communication interruption:

[0036] ;

[0037] in, Indicates missing nodes The introduced confidence weights, Indicates missing nodes The initial weighting coefficients, This represents an exponential function with the natural constant as its base. Indicates missing nodes The duration of continuous packet loss during communication. This represents the preset time decay constant.

[0038] In some specific embodiments, the analysis computer constructs a state reconstruction model based on the spatial physical correlation constraint matrix of the submarine cable for global solution. Specifically, the configuration involves discretizing the state reconstruction model using the finite element method to obtain a system of linear equations concerning the spatial characteristics of the nodes, expressed as:

[0039] ;

[0040] in, Represents the space physical correlation constraint matrix. This represents the global node state feature vector to be determined. Represents the complete equivalent state vector;

[0041] The adaptive regularized conjugate gradient method is used to solve the linear equation system to obtain the global node state feature vector, including: initializing and setting the initial state feature vector. Initial residual vector Initial search direction vector and initial regularization parameters For the first The next iteration, in which Perform the following iterative updates until the convergence condition is met:

[0042] Calculate the iteration step size , is represented as:

[0043] ;

[0044] Update global node state feature vector , is represented as:

[0045] ;

[0046] Update residual vector , is represented as:

[0047] ;

[0048] Calculate the conjugate coefficient , is represented as:

[0049] ;

[0050] Update the search direction vector , is represented as:

[0051] ;

[0052] The regularization parameter is dynamically adjusted based on the current rate of change of the residual. , is represented as:

[0053] ;

[0054] in, Indicates the number of iterations. Indicates the first The step size of the next iteration. and They represent the first Second and third The global node state feature vector of the next iteration. and They represent the first Second and third The residual vector of the next iteration and They represent the first Second and third The search direction vector for the next iteration. Indicates the first The conjugate coefficients calculated in the next iteration. and They represent the first Second and third The regularization parameter for the next iteration Represents the identity matrix. This represents the transpose of a vector. Denotes the Frobenius norm of a matrix. and They represent the first Second and third The next iteration produces the norm of the residual vector; the convergence condition is... Or reach the maximum number of iterations, where, This indicates the preset convergence accuracy.

[0055] In some specific embodiments, the analysis computer extracts feature sub-blocks corresponding to the set of valid nodes from the spatial physical correlation constraint matrix for order reduction calculation, and uses a distance-based spatial interpolation method to recover the local state data corresponding to the missing nodes, so as to combine and obtain the complete global node state feature vector. The specific configuration is as follows:

[0056] Based on the set of valid nodes, extract the feature sub-blocks of the spatial-physical association constraint matrix corresponding to the set of valid nodes. And the state sub-vectors corresponding to the effective node set extracted from the complete equivalent state vector. And construct a reduced-order system of equations to solve it, expressed as:

[0057] ;

[0058] in, This represents the state feature vector of the set of valid nodes obtained by the reduced-order solution, and for missing nodes in the set of missing nodes... The axial displacement and bending angle are calculated using the adjacent node interpolation method, and are expressed as follows:

[0059] ;

[0060] ;

[0061] in, Represents the set of valid nodes. Indicates missing nodes The set of valid neighboring nodes, Represents the set of valid neighborhood nodes Valid nodes in Indicates missing nodes With neighboring valid nodes Interpolation weights based on distance between them. Indicates missing nodes axial displacement, Indicates the effective nodes in the neighborhood. axial displacement, Indicates missing nodes The bends and turns, Indicates the effective nodes in the neighborhood. The bending angle; missing nodes obtained based on calculations. axial displacement With bending angle Construct the local state data corresponding to the missing nodes through interpolation estimation;

[0062] State feature subvectors of the effective node set obtained by integrating the order reduction solution The complete global node state feature vector is obtained by combining the local state data corresponding to the missing nodes estimated by interpolation.

[0063] In some specific embodiments, the analysis computer reconstructs the overall state distribution of the submarine cable based on the acquired global node state feature vector. Specifically, this is configured as follows: extracting the axial displacement and bending angle of each node from the acquired global node state feature vector; and reconstructing the axial stress, bending stress, and overall state distribution at any location on the submarine cable based on the extracted axial displacement and bending angle of each node, as shown below:

[0064] ;

[0065] ;

[0066] ;

[0067] in, Indicates the submarine cable in arc length coordinates Place Axial stress at time t, This represents the elastic modulus of the submarine cable material. This represents the axial strain obtained based on the extracted axial displacements of each node. Indicates the submarine cable in arc length coordinates Place Bending stress at any moment, This represents the bending moment obtained based on the extracted bending angles of each node. This represents the distance from the calculation point to the neutral axis of the submarine cable section. The moment of inertia of the submarine cable section is represented by the moment of inertia of the submarine cable section. Indicates the submarine cable in arc length coordinates Place The overall state distribution at time t.

[0068] According to a second aspect of the present invention, a LoRa-based remote wireless telemetry method for submarine cable status applied to the above-mentioned system is proposed, comprising:

[0069] Each status monitoring node collects local status data of the submarine cable and transmits it via a wireless communication module;

[0070] The relay node receives local status data and forwards it to the shipborne monitoring center.

[0071] The shipboard monitoring center's analysis computer monitors the communication link status in real time to identify the set of valid nodes that actually receive data and the set of missing nodes that experience packet loss. It extracts the spatial-physical correlation constraint matrix of the submarine cable based on all status monitoring nodes, calculates the missing node ratio, and compares this ratio with preset first and second packet loss thresholds, where the first threshold is less than the second threshold. If the missing ratio is below the first threshold, it uses the current state change rate and spatial correlation of the local state data corresponding to the valid node set, combined with the historical local state data of the missing nodes, to perform trend filling on the local state data corresponding to the missing nodes, constructing a complete equivalent state vector. A state reconstruction model is then built using the submarine cable's spatial-physical correlation constraint matrix for global solution. If the missing ratio is greater than the first threshold but less than the second threshold, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after missing node filling in the state reconstruction model.

[0072] If the missing proportion is greater than the second packet loss threshold, feature sub-blocks corresponding to the set of valid nodes in the spatial physical association constraint matrix are extracted for order reduction calculation, and the local state data corresponding to the missing nodes is restored by distance-based spatial interpolation method, so as to combine and obtain the complete global node state feature vector.

[0073] Based on the acquired global node state feature vectors, the state distribution of the entire submarine cable is reconstructed, and hierarchical early warnings are output.

[0074] Compared with the prior art, the present invention has the following significant advantages:

[0075] 1. This invention innovatively proposes a hierarchical compensation mechanism based on missing threshold comparison by analyzing the real-time monitoring of communication link status and statistically analyzing the missing proportion. The system does not rely on an ideal communication environment and can adaptively switch processing strategies under different degrees of communication interruption, completely solving the technical bottleneck of traditional wireless monitoring systems where data breakpoints prevent the solution of global physical equations, thus ensuring the stable operation of the monitoring system under extreme sea conditions.

[0076] 2. When the missing data ratio is below the threshold for mild or moderate packet loss, this invention abandons simple numerical interpolation and instead utilizes the data change rate of valid nodes, spatial correlation, and historical data of missing nodes to perform physical trend filling to construct a complete equivalent state vector. Particularly under moderate packet loss conditions, a confidence weight that dynamically decays with the communication interruption time is introduced into the state reconstruction model. This mechanism gradually reduces the contribution (weight) of nodes that have been offline for a long time to the global inversion result, effectively avoiding erroneous data contamination of the solution process caused by long-term filling, minimizing reconstruction errors, and ensuring high accuracy in the global solution.

[0077] 3. This invention performs order reduction calculations by extracting feature sub-blocks corresponding to valid nodes in the spatial physical correlation constraint matrix, removing the degrees of freedom corresponding to missing nodes from the equation system, and optimizing the matrix condition number from a mathematical perspective, ensuring absolute convergence of the algorithm under extremely harsh communication conditions. Simultaneously, by combining distance-based spatial interpolation to recover missing node data, it ensures the correct output of the state distribution trend of key areas of the submarine cable.

[0078] 4. This invention, through the intermittent deployment of status monitoring nodes and relay nodes, utilizes LoRa technology to send local status data to the shipboard monitoring center. This completely eliminates the extremely complex signal cable follow-up wiring process in traditional long-distance submarine cable monitoring, fundamentally eliminating the risks of mechanical damage and severe signal attenuation inherent in wired transmission. Ultimately, based on dynamically reconstructed global node status feature vectors, the entire submarine cable status distribution is restored, and tiered early warnings are output in real time. This provides intuitive and real-time safety guidance for submarine cable laying operations, effectively preventing laying accidents such as cable damage, and greatly improving construction efficiency and safety. Attached Figure Description

[0079] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and, together with the description, serve to explain the principles of the invention. Other embodiments and many anticipated advantages of the embodiments will be readily recognized as they become better understood through reference to the following detailed description. Other features, objects, and advantages of this application will become more apparent from reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0080] Figure 1 This is a framework diagram of a LoRa-based remote wireless telemetry system for submarine cable status, according to an embodiment of the present invention.

[0081] Figure 2 This is a structural diagram of a status monitoring node according to an embodiment of the present invention;

[0082] Figure 3This is a flowchart of a LoRa-based remote wireless telemetry system for submarine cable status, according to an embodiment of the present invention.

[0083] Figure 4 This is a flowchart of a LoRa-based remote wireless telemetry method for submarine cable status according to an embodiment of the present invention; Detailed Implementation

[0084] In the description of this invention, it should be noted that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0085] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "installation," "connection," and "fixation," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0086] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0087] refer to Figure 1 This invention provides a LoRa-based remote wireless telemetry system for submarine cable status, comprising: a submarine cable 200 to be laid, a status monitoring node 100 and a relay node 300 installed thereon, and a shipborne monitoring center 400 including a gateway 401 and an analysis computer 402.

[0088] Specifically, the configurations of each hardware module in the system are as follows:

[0089] Several status monitoring nodes 100 are deployed at intervals on the submarine cable 200 to be laid, and are configured to collect local status data of the submarine cable and transmit it via a wireless communication module.

[0090] Preferably, several status monitoring nodes 100 are installed using magnetic or binding methods, and their placement and spacing can be flexibly adjusted according to monitoring needs.

[0091] Preferably, several status monitoring nodes 100 have a built-in high-precision real-time clock chip 105, which periodically broadcasts a reference time through the shipborne monitoring center 400 to achieve clock synchronization of all nodes in the network, with a synchronization accuracy better than 1ms. Each of the several status monitoring nodes 100 includes:

[0092] (1) Status sensor unit 101: adopts fiber optic strain sensor or resistance strain gauge to collect local status data in real time during the laying of submarine cable;

[0093] (2) Signal conditioning circuit 102: including A / D converter 103, used to amplify, filter and convert the weak signal output by state sensor unit 101;

[0094] (3) Power management unit 104: It is powered by a high-energy lithium battery and integrates an energy harvesting module. Optionally, it can use the vibration energy of the submarine cable or solar energy to supplement the power, so as to achieve low power consumption operation.

[0095] (4) Microcontroller unit 106: controls the status sensor unit to perform data acquisition, processing, storage and wireless transmission, and executes the time synchronization protocol between nodes;

[0096] (5) LoRa wireless communication module 107: Using LoRa spread spectrum communication technology, the collected local status data is wirelessly transmitted to relay node 300 or shipborne monitoring center 400 through antenna 108.

[0097] At least one relay node 300 is deployed along the wireless transmission path and configured to receive local status data and route it to the shipboard monitoring center 400, thereby extending network coverage. Each relay node 300 includes: a LoRa wireless communication module supporting multi-channel parallel reception and forwarding; a data buffer unit for temporarily storing forwarded data; a microcontroller unit that executes routing protocols to optimize the data transmission path; and a power supply unit that can be powered by a high-capacity battery or the ship.

[0098] To ensure the reliability of long-distance transmission between nodes, this embodiment adopts the standard LoRa data packet frame structure, which includes: an 8-byte preamble (used for receiver synchronization and gain control), a 2-byte synchronization word (used to distinguish data packets from different networks), a 4-byte address field (containing a 2-byte source node ID and a 2-byte destination node ID), a 1-byte command field (identifying the data packet type, such as data acquisition, time synchronization, etc.), an N-byte data field (carrying the payload data), and a 2-byte check field (using CRC16 check to ensure transmission integrity).

[0099] The shipborne monitoring center 400, deployed on the cable-laying vessel, includes a gateway 401 and an analysis computer 402 communicating with the gateway. Furthermore, the shipborne monitoring center 400 also includes a data server storing real-time and historical data, and a human-machine interface for displaying real-time monitoring data and early warning information. Optionally, the system also includes a cloud server connected to the shipborne monitoring center 400 via satellite communication or a 4G / 5G network, enabling remote aggregation, storage, and analysis of monitoring data from multiple cable-laying vessels.

[0100] refer to Figure 3 The system workflow diagram shows the entire telemetry method executed sequentially from the beginning: S301, System initialization (node ​​self-test, network registration); S302, Shipborne monitoring center broadcasts time synchronization command via 400; S303, Each node receives and calibrates its local clock (synchronization accuracy). S304. Then, the loop execution phase (sampling frequency is configurable) begins, including: S305. The status sensor collects local status data; S306. Signal conditioning and A / D conversion are performed; S307. The microcontroller encapsulates data packets (adding timestamps and node IDs); S308. The relay node 300 receives and forwards data (multi-hop routing); S309. The shipborne LoRa gateway receives data from all nodes; S310. The data server stores the raw data; S311. The analysis computer performs state reconstruction and global solution; S312. Finally, the status distribution of the entire submarine cable is reconstructed and an early warning is output; S313. The data is simultaneously displayed in real time on the human-machine interface.

[0101] Furthermore, regarding the core component S312, the analysis computer is configured to monitor the communication link status in real time to identify the set of valid nodes that actually received data and the set of missing nodes where packet loss occurred. To address the technical challenge of LoRa communication instability under extreme sea conditions, the analysis computer specifically implements the following adaptive hierarchical compensation mechanism:

[0102] Step 1: Extract the space-physical correlation constraint matrix:

[0103] The computer analyzes and extracts the spatial-physical correlation constraint matrix of the submarine cable based on all status monitoring nodes, with the following specific configuration:

[0104] First, considering the submarine cable as a one-dimensional continuum, we establish its dynamic equations for axial tension and bending deformation. Let the length of the submarine cable be... The axial coordinate of the coastal cable is , time The axial displacement at is lateral displacement is The axial strain of the submarine cable and curvature They are respectively:

[0105] ；

[0106] ;

[0107] Axial force of submarine cable and bending moment Given by constitutive relations:

[0108] ;

[0109] ;

[0110] in, The elastic modulus of the submarine cable material. For cross-sectional area, Let be the moment of inertia of the cross section.

[0111] Subsequently, the analysis computer used the finite segment method to discretize the submarine cable into... The system consists of several units, each corresponding to a state monitoring node, thus transforming a continuous mechanical problem into a discrete problem that can be solved numerically. The system calculates the value of the first unit based on the spatial continuity and physical correlation. The axial force and bending moment of each element are expressed as follows:

[0112] ;

[0113] ;

[0114] in, , Indicates the first Axial force of each unit Indicates the first The bending moment of each element, This represents the elastic modulus of the submarine cable material. This represents the cross-sectional area of ​​the submarine cable. The moment of inertia of the submarine cable section is represented by the moment of inertia of the submarine cable section. Indicates the unit length. and They represent the first The and the first Axial displacement of each status monitoring node and They represent the first The and the first The bending angle of each condition monitoring node; extracting the axial force calculation formula. With the bending moment calculation formula As unit characteristic coefficients, they are assembled according to the corresponding state monitoring nodes to form a spatial physical correlation constraint matrix.

[0115] Step 2: Packet loss ratio statistics and data filling for minor packet loss:

[0116] In marine environments, LoRa wireless communication may experience packet loss or temporary node offline due to factors such as waves, ship hull obstruction, and multipath effects, resulting in incomplete real-time status monitoring data. Without processing, the load vectors in subsequent equations will have "gap" or "breakpoints," causing the conjugate gradient method iteration to fail to converge or fail to solve.

[0117] The system provides real-time statistics on the data reception status of each node. Let the current time be... The total number of status monitoring nodes is The actual set of nodes receiving data is The set of nodes with missing data is .

[0118] The missing percentage of statistically missing nodes is analyzed and compared with preset first and second packet loss thresholds, where the first packet loss threshold is less than the second packet loss threshold. If the missing percentage is less than the first packet loss threshold (e.g., ...), the missing percentage is calculated based on the missing percentage of nodes at the first packet loss threshold. When the following occurs, for missing nodes Since the state monitoring value is unknown, this invention abandons the simple interpolation method and instead uses the current state change rate and spatial correlation of the local state data corresponding to the effective node set, combined with the historical local state data of the missing node, to perform trend filling on the local state data corresponding to the missing node in order to construct a complete equivalent state vector.

[0119] The specific configuration is as follows:

[0120] If historical data exists for a missing node, the data is filled in by combining the state change trend of the valid node set with spatial correlation, as shown below:

[0121] ;

[0122] in, This indicates the existence of nodes with missing historical data. Real-time local state data corresponding to the current moment. Indicates missing nodes Historical local state data corresponding to the previous moment, Indicates the sampling interval time. The length of the time window used for smoothing trend estimation. Represents the set of valid nodes. Represents the set of valid nodes Valid nodes in Indicates a valid node The corresponding rate of change of the current state of the local state data. Indicates missing nodes With valid nodes Spatial correlation weighting coefficients between them;

[0123] If a missing node has no historical data, the data is filled in using the global average trend, as shown below:

[0124] ;

[0125] in, Indicates a missing node with no historical data. Real-time local state data corresponding to the current moment. This represents the global average of the historical local state data corresponding to all valid nodes at the previous time step. Represents the set of valid nodes. Represents the set of valid nodes Valid nodes in Indicates a valid node The corresponding rate of change of the current state of the local state data. Indicates the sampling interval time;

[0126] And valid nodes The current state change rate is based on the effective nodes. The real-time local state data corresponding to the current moment and the historical local state data corresponding to the previous moment are calculated and obtained, and are represented as follows: ; Indicates a valid node Real-time local state data corresponding to the current moment. Indicates a valid node Historical local state data corresponding to the previous moment

[0127] Integrate nodes with missing historical data The current real-time local state data and / or missing nodes with no historical data at the current moment. The local state data at the current moment is combined with the real-time local state data of each valid node in the set of valid nodes to construct a complete equivalent state vector. , is represented as:

[0128] ;

[0129] .

[0130] Step 3: Dynamic confidence weight compensation under moderate packet loss conditions:

[0131] The computer analysis unit constructs a state reconstruction model by combining the spatial physical correlation constraint matrix of the submarine cable. In this model, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after filling in missing nodes. The specific configuration is as follows:

[0132] When the missing rate is below the first packet loss threshold, a state reconstruction model is constructed by combining the spatial physical correlation constraint matrix of the submarine cable, which is expressed as:

[0133] ;

[0134] And when the missing percentage is greater than the first packet loss threshold but less than the second packet loss threshold (e.g., moderate packet loss rate), Packet loss rate In the state reconstruction model, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after the missing nodes are filled, which is expressed as:

[0135] ;

[0136] in, and Let represent the basic reconstruction objective function and the reconstruction objective function with confidence weights, respectively. Represents the set of valid nodes. Represents the set of missing nodes; Represents the set of valid nodes Valid nodes in Represents the set of missing nodes Missing nodes in; Represents the effective nodes in the complete equivalent state vector. The corresponding real-time local state data at the current moment. This represents the valid nodes obtained based on the global node state feature vector to be determined. The model calculates the state function value; Represents missing nodes in the complete equivalent state vector The local state data after filling in the gaps This represents the missing node obtained based on the feature vector of the global node state to be determined. The model calculates the state function value; Indicates a valid node The initial weighting coefficients, Indicates missing nodes Introduced confidence weights; This represents the first-order smoothing regularization parameter. This indicates the total length of the submarine cable to be laid. This indicates the axial displacement of the submarine cable. This represents the second-order smoothing regularization parameter. Indicates the bend angle of the submarine cable. The coordinates represent the arc length of the submarine cable.

[0137] Furthermore, the analysis computer calculates the filled missing nodes based on the following formula. The confidence weights introduced dynamically decay with the duration of communication interruption:

[0138] ;

[0139] in, Indicates missing nodes The introduced confidence weights, Indicates missing nodes The initial weighting coefficients, This represents an exponential function with the natural constant as its base. Indicates missing nodes The duration of continuous packet loss during communication. This represents the preset time decay constant. This mechanism gradually reduces the contribution of nodes that have been offline for a long time to the inversion results, avoiding erroneous data from contaminating the solution process.

[0140] Step 4: Dynamic reduction and interpolation recovery of the matrix under severe packet loss conditions:

[0141] If the missing proportion is greater than the second packet loss threshold (e.g., the packet loss rate is too high) In other words, when the number of missing nodes is too large (e.g., exceeding 30% of the total number of nodes) or all nodes in key areas are lost, a simple filling strategy may lead to excessive uncertainty in the solution. Therefore, the analysis computer will extract feature sub-blocks corresponding to the set of valid nodes from the spatial-physical association constraint matrix for order reduction calculations. The analysis computer then uses distance-based spatial interpolation to recover the local state data corresponding to the missing nodes, combining them to obtain a complete global node state feature vector. The specific configuration is as follows:

[0142] Based on the set of valid nodes, extract the feature sub-blocks of the spatial-physical association constraint matrix corresponding to the set of valid nodes. And the state sub-vectors corresponding to the effective node set extracted from the complete equivalent state vector. And construct a reduced-order system of equations to solve it, expressed as:

[0143] ;

[0144] in, This represents the state feature vector of the set of valid nodes obtained by the reduced-order solution, and for missing nodes in the set of missing nodes... The axial displacement and bending angle are calculated using the adjacent node interpolation method, and are expressed as follows:

[0145] ;

[0146] ;

[0147] in, Represents the set of valid nodes. Indicates missing nodes The set of valid neighboring nodes, Represents the set of valid neighborhood nodes Valid nodes in Indicates missing nodes With neighboring valid nodes Interpolation weights based on distance between them. Indicates missing nodes axial displacement, Indicates the effective nodes in the neighborhood. axial displacement, Indicates missing nodes The bends and turns, Indicates the effective nodes in the neighborhood. The bending angle; missing nodes obtained based on calculations. axial displacement With bending angle Construct the local state data corresponding to the missing nodes through interpolation estimation;

[0148] State feature subvectors of the effective node set obtained by integrating the order reduction solution The local state data corresponding to the missing nodes estimated by interpolation are combined to obtain the complete global node state feature vector. The size of the equation system is reduced after the order reduction, and the condition number is usually better than that of the original system, which can ensure the fast convergence of the conjugate gradient method. This ensures that the system can still converge and obtain the stress distribution trend in the key region even under extreme communication conditions.

[0149] Step 5: Globally solve for the adaptive regularized conjugate gradient:

[0150] For the global state reconstruction model in step 3, the analysis computer constructs a state reconstruction model based on the spatial physical correlation constraint matrix of the submarine cable for global solution. Specifically, the finite element method is used to discretize the state reconstruction model, yielding a system of linear equations concerning the spatial characteristics of the nodes, expressed as:

[0151] ;

[0152] in, Represents the space physical correlation constraint matrix. This represents the global node state feature vector to be determined. Represents the complete equivalent state vector;

[0153] The adaptive regularized conjugate gradient method is used to solve the linear equation system to obtain the global node state feature vector, including: initializing and setting the initial state feature vector. Initial residual vector Initial search direction vector and initial regularization parameters For the first The next iteration, in which Perform the following iterative updates until the convergence condition is met:

[0154] Calculate the iteration step size , is represented as:

[0155] ;

[0156] Update global node state feature vector , is represented as:

[0157] ;

[0158] Update residual vector , is represented as:

[0159] ;

[0160] Calculate the conjugate coefficient , is represented as:

[0161] ;

[0162] Update the search direction vector , is represented as:

[0163] ;

[0164] The regularization parameter is dynamically adjusted based on the current rate of change of the residual. , is represented as:

[0165] ;

[0166] in, Indicates the number of iterations. Indicates the first The step size of the next iteration. and They represent the first Second and third The global node state feature vector of the next iteration. and They represent the first Second and third The residual vector of the next iteration and They represent the first Second and third The search direction vector for the next iteration. Indicates the first The conjugate coefficients calculated in the next iteration. and They represent the first Second and third The regularization parameter for the next iteration Represents the identity matrix. This represents the transpose of a vector. Denotes the Frobenius norm of a matrix. and They represent the first Second and third The next iteration produces the norm of the residual vector; the convergence condition is... Or reach the maximum number of iterations, where, This indicates the preset convergence accuracy.

[0167] S6. Reconstruction and graded early warning of the status distribution of the entire submarine cable:

[0168] The analysis computer reconstructs the overall state distribution of the submarine cable based on the acquired global node state feature vectors. Specifically, the configuration involves: extracting the axial displacement and bending angle of each node from the acquired global node state feature vectors; and reconstructing the axial stress, bending stress, and overall state distribution at any location on the submarine cable based on the extracted axial displacement and bending angle of each node, as shown below:

[0169] ;

[0170] ;

[0171] ;

[0172] in, Indicates the submarine cable in arc length coordinates Place Axial stress at time t, This represents the elastic modulus of the submarine cable material. This represents the axial strain obtained based on the extracted axial displacements of each node. Indicates the submarine cable in arc length coordinates Place Bending stress at any moment, This represents the bending moment obtained based on the extracted bending angles of each node. This represents the distance from the calculation point to the neutral axis of the submarine cable section. The moment of inertia of the submarine cable section is represented by the moment of inertia of the submarine cable section. Indicates the submarine cable in arc length coordinates Place The overall state distribution at time t.

[0173] Through the above-mentioned hierarchical compensation mechanism, the present invention can still maintain the stable operation of the inversion algorithm even when packet loss occurs in LoRa communication:

[0174] 1. Packet loss rate At that time, a filling strategy combined with the adaptive regularized conjugate gradient method is adopted to reduce the state inversion error. ;

[0175] 2. Packet loss rate When using confidence weight adjustment, the error is... ;

[0176] 3. Packet loss rate At that time, a matrix dynamic order reduction process is adopted to ensure absolute convergence of the solution and the correct state distribution trend in the key region.

[0177] This mechanism closely integrates purely mathematical solution methods with communication uncertainties in engineering practice, significantly improving the robustness and practicality of the system.

[0178] As an example, to guide on-site construction, a stress concentration factor is defined. ,in, This represents the rated operating stress of the submarine cable. Based on this, a graded early warning system is output:

[0179] like 1.2, triggering a Level 1 warning and prompting attention;

[0180] like 1.5, triggering a Level 2 warning, parameter adjustments are recommended;

[0181] like 2.0, triggering a Level 3 warning will immediately halt the laying process.

[0182] It should be specifically noted that, in specific marine engineering and submarine cable laying applications, the "state" in this invention can specifically refer to the "stress" or "strain" of the submarine cable. That is, the "state monitoring node" specifically manifests as a "stress monitoring node," the "local state data" it collects specifically refers to "local stress data" or "local strain data," and the constructed "equivalent state vector" specifically refers to the "equivalent load vector" used for mechanical solutions. The LoRa-based remote wireless telemetry system and adaptive packet loss compensation algorithm provided by this invention are perfectly applicable to real-time stress state inversion and tension over-limit early warning of submarine cables under extreme sea conditions.

[0183] This invention also proposes a LoRa-based remote wireless telemetry method for submarine cable status applied to the above-mentioned system, comprising the following core steps:

[0184] S100: Each status monitoring node collects local status data of the submarine cable and transmits it via the wireless communication module;

[0185] S200, The relay node receives the local status data and forwards it to the shipborne monitoring center;

[0186] S300, the analysis computer of the shipborne monitoring center monitors the status of the communication link in real time to identify the set of valid nodes that actually receive data and the set of missing nodes that have lost communication packets, extracts the spatial physical association constraint matrix of the submarine cable based on all the status monitoring nodes, calculates the missing ratio of missing nodes, and compares the missing ratio with the preset first packet loss threshold and second packet loss threshold.

[0187] S301. If the missing ratio is below the first packet loss threshold, the current state change rate and spatial correlation of the local state data corresponding to the effective node set are used, combined with the historical local state data of the missing node, to perform trend filling on the local state data corresponding to the missing node, so as to construct a complete equivalent state vector, and combine the spatial physical correlation constraint matrix of the submarine cable to construct a state reconstruction model for global solution.

[0188] S302. If the missing ratio is greater than the first packet loss threshold and less than the second packet loss threshold, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after the missing nodes are filled in the state reconstruction model.

[0189] S303. If the missing proportion is greater than the second packet loss threshold, extract the feature sub-blocks in the spatial physical association constraint matrix corresponding to the set of effective nodes for order reduction calculation, and use the distance-based spatial interpolation method to recover the local state data corresponding to the missing node, so as to combine and obtain the complete global node state feature vector.

[0190] S400: Based on the acquired global node state feature vectors, reconstruct the state distribution of the entire submarine cable and output hierarchical early warnings.

[0191] As an example, the system of this application was applied to the monitoring of submarine cable laying in shallow-sea wind farms (water depth 30-50 meters). 75 status monitoring nodes were deployed at 200-meter intervals along the surface of a 15-kilometer-long, 120mm-diameter submarine cable, with one relay node (300) deployed every 2 kilometers, and a sampling frequency of 10Hz. The system configuration included: fiber optic grating sensors with an accuracy of ±5με, LoRa at 470MHz, and a spreading factor of SF12. The system achieved a data reception success rate of 99.2%, a maximum communication distance of 8.7km, and an average relative error of only 4.7% in stress inversion. A secondary early warning was triggered when traversing steep slopes, preventing mechanical damage.

[0192] As an example two, the system of this application was applied to the monitoring of deep-sea communication optical cables (water depth 2000-4000 meters). Differential strain measurement was used to eliminate the influence of temperature, and optimizations were made in steep slope sections (100-meter intervals) and adaptive sampling (adjusted higher when there are severe fluctuations). Even under the interference of deep-sea multipath effects, combined with the packet loss dynamic reconstruction mechanism of this paper, the reception success rate remained at 98.5%, the stress concentration location identification error was less than 10 meters, and the endurance was greater than 500 days.

[0193] In this specification, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, those skilled in the art can combine and integrate different embodiments or examples and features of different embodiments or examples described in this specification without contradiction. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention. The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made according to the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A LoRa-based remote wireless telemetry system for submarine cable status, characterized in that, include: Several status monitoring nodes are deployed at intervals on the submarine cable to be laid, and are configured to collect local status data of the submarine cable and transmit it via a wireless communication module; At least one relay node is deployed on the wireless transmission path and configured to receive the local status data and route it to the shipborne monitoring center. The shipborne monitoring center includes a gateway and an analysis computer communicatively connected to the gateway. The analysis computer is configured to monitor the communication link status in real time to identify the set of valid nodes that actually receive data and the set of missing nodes that experience packet loss. It extracts the spatial-physical correlation constraint matrix of the submarine cable based on all the status monitoring nodes, calculates the missing proportion of missing nodes, and compares the missing proportion with a preset first packet loss threshold and a second packet loss threshold, wherein the first packet loss threshold is less than the second packet loss threshold. If the missing proportion is below the first packet loss threshold, it uses the current state change rate and spatial correlation of the local state data corresponding to the set of valid nodes, combined with the historical local state data of the missing nodes, to perform trend filling on the local state data corresponding to the missing nodes, to construct a complete equivalent state vector. It then uses the spatial-physical correlation constraint matrix of the submarine cable to construct a state reconstruction model for global solution. If the missing proportion is greater than the first packet loss threshold but less than the second packet loss threshold, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after the missing nodes are filled in the state reconstruction model. If the missing proportion is greater than the second packet loss threshold, feature sub-blocks corresponding to the set of effective nodes in the spatial physical association constraint matrix are extracted for order reduction calculation, and the local state data corresponding to the missing node is restored by distance-based spatial interpolation method, so as to combine and obtain the complete global node state feature vector. Based on the acquired global node state feature vectors, the state distribution of the entire submarine cable is reconstructed, and hierarchical early warnings are output.

2. The system according to claim 1, characterized in that, The analysis computer extracts the spatial-physical correlation constraint matrix of the submarine cable based on all the status monitoring nodes. Specifically, the configuration involves discretizing the submarine cable using the finite segment method. There are 1 unit, and each unit corresponds to one of the aforementioned state monitoring nodes. The 1st state monitoring node is calculated based on the spatial continuity and physical correlation. The axial force and bending moment of each element are expressed as follows: ； ； in, , Indicates the first Axial force of each unit Indicates the first The bending moment of each element, This represents the elastic modulus of the submarine cable material. This represents the cross-sectional area of ​​the submarine cable. The moment of inertia of the submarine cable section is represented by the moment of inertia of the submarine cable section. Indicates the unit length. and They represent the first The and the first Axial displacement of each status monitoring node and They represent the first The and the first The bending angle of each state monitoring node; extract the axial force calculation formula from the formula. With the bending moment calculation formula As unit characteristic coefficients, they are assembled according to the corresponding state monitoring nodes to form the spatial physical correlation constraint matrix.

3. The system according to claim 1, characterized in that, The analysis computer utilizes the current state change rate and spatial correlation of the local state data corresponding to the effective node set, combined with the historical local state data of the missing nodes, to perform trend imputation on the local state data corresponding to the missing nodes, in order to construct a complete equivalent state vector. The specific configuration is as follows: If the missing node has historical data, data filling is performed by combining the state change trend of the valid node set with spatial correlation, as follows: ； in, This indicates the existence of nodes with missing historical data. Real-time local state data corresponding to the current moment. Indicates missing nodes Historical local state data corresponding to the previous moment, Indicates the sampling interval time. The length of the time window used for smoothing trend estimation. Represents the set of valid nodes. Represents the set of valid nodes Valid nodes in Indicates a valid node The corresponding rate of change of the current state of the local state data, Indicates missing nodes With valid nodes Spatial correlation weighting coefficients between them; If the missing node has no historical data, then the data is filled using the global average trend, as shown below: ； in, Indicates a missing node with no historical data. Real-time local state data corresponding to the current moment. This represents the global average of the historical local state data corresponding to all valid nodes at the previous time step. Represents the set of valid nodes. Represents the set of valid nodes Valid nodes in Indicates a valid node The corresponding rate of change of the current state of the local state data, Indicates the sampling interval time; And the effective node The current state change rate is based on the effective nodes. The real-time local state data corresponding to the current moment and the historical local state data corresponding to the previous moment are calculated and obtained, and are represented as follows: ; Indicates a valid node Real-time local state data corresponding to the current moment. Indicates a valid node Historical local state data corresponding to the previous moment; Integrate the missing nodes of the existing historical data The current real-time local state data and / or the missing nodes without historical data. The real-time local state data corresponding to the current moment is combined with the real-time local state data of each valid node in the set of valid nodes to construct the complete equivalent state vector.

4. The system according to claim 3, characterized in that, The analysis computer calculates missing nodes according to the following formula. With valid nodes The spatial correlation weighting coefficient between them is expressed as: ； in, Indicates missing nodes With valid nodes Spatial correlation weighting coefficient between them Indicates missing nodes With valid nodes The covariance of the corresponding historical local state data, and These represent missing nodes. With valid nodes The variance of the corresponding historical local state data.

5. The system according to claim 1, characterized in that, The analysis computer constructs a state reconstruction model by combining the spatial physical correlation constraint matrix of the submarine cable, and introduces a confidence weight that dynamically decays with the communication interruption time into the local state data after the missing nodes are filled in the state reconstruction model. The specific configuration is as follows: When the missing rate is below the first packet loss threshold, a state reconstruction model is constructed based on the spatial physical correlation constraint matrix of the submarine cable, as follows: ； And when the missing proportion is greater than the first packet loss threshold but less than the second packet loss threshold, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after the missing nodes are filled in the state reconstruction model, which is expressed as: ； in, and Let represent the basic reconstruction objective function and the reconstruction objective function with confidence weights, respectively. Represents the set of valid nodes. Represents the set of missing nodes; Represents the set of valid nodes Valid nodes in Represents the set of missing nodes Missing nodes in; Represents the effective nodes in the complete equivalent state vector. The corresponding real-time local state data at the current moment. This represents the valid nodes obtained based on the global node state feature vector to be determined. The model calculates the state function value; This represents the missing nodes in the complete equivalent state vector. The local state data after filling in, This represents the missing node obtained based on the global node state feature vector to be determined. The model calculates the state function value; Indicates a valid node The initial weighting coefficients, Indicates missing nodes The introduced confidence weights; This represents the first-order smoothing regularization parameter. This indicates the total length of the submarine cable to be laid. This indicates the axial displacement of the submarine cable. This represents the second-order smoothing regularization parameter. Indicates the bend angle of the submarine cable. The coordinates represent the arc length of the submarine cable.

6. The system according to claim 5, characterized in that, The analysis computer calculates the filled missing nodes according to the following formula. The introduced confidence weights that dynamically decay with communication interruption time: ； in, Indicates missing nodes The introduced confidence weights, Indicates missing nodes The initial weighting coefficients, This represents an exponential function with the natural constant as its base. Indicates missing nodes The duration of continuous packet loss during communication. This represents the preset time decay constant.

7. The system according to claim 5, characterized in that, The analysis computer constructs a state reconstruction model based on the spatial physical correlation constraint matrix of the submarine cable for global solution. Specifically, the configuration involves discretizing the state reconstruction model using the finite element method to obtain a system of linear equations concerning the spatial characteristics of the nodes, expressed as: ； in, This represents the spatial physical correlation constraint matrix. Let represent the global node state feature vector to be determined. This represents the complete equivalent state vector; The adaptive regularized conjugate gradient method is used to solve the linear equation system to obtain the global node state feature vector, including: initializing and setting the initial state feature vector. Initial residual vector Initial search direction vector and initial regularization parameters For the first The next iteration, in which Perform the following iterative updates until the convergence condition is met: Calculate the iteration step size , is represented as: ； Update the global node state feature vector , is represented as: ； Update residual vector , is represented as: ； Calculate the conjugate coefficient , is represented as: ； Update the search direction vector , is represented as: ； The regularization parameter is dynamically adjusted based on the current rate of change of the residual. , is represented as: ； in, Indicates the number of iterations. Indicates the first The step size of the next iteration. and They represent the first Second and third The global node state feature vector of the next iteration. and They represent the first Second and third The residual vector of the next iteration and They represent the first Second and third The search direction vector for the next iteration. Indicates the first The conjugate coefficients calculated in the next iteration. and They represent the first Second and third The regularization parameter for the next iteration Represents the identity matrix. This represents the transpose of a vector. Denotes the Frobenius norm of a matrix. and They represent the first Second and third The norm of the residual vector is generated in the next iteration; the convergence condition is... Or reach the maximum number of iterations, where, This indicates the preset convergence accuracy.

8. The system according to claim 1, characterized in that, The analysis computer extracts feature sub-blocks corresponding to the set of valid nodes from the spatial physical correlation constraint matrix, performs order reduction calculations, and uses a distance-based spatial interpolation method to recover the local state data corresponding to the missing nodes, so as to combine and obtain a complete global node state feature vector. The specific configuration is as follows: Based on the set of valid nodes, extract the feature sub-blocks of the spatial-physical association constraint matrix corresponding to the set of valid nodes. and the state sub-vectors corresponding to the set of effective nodes extracted from the complete equivalent state vector. And construct a reduced-order system of equations to solve it, expressed as: ； in, This represents the state feature sub-vector of the effective node set obtained by the reduced-order solution, and for the missing nodes in the missing node set... The axial displacement and bending angle are calculated using the adjacent node interpolation method, and are expressed as follows: ； ； in, Represents the set of valid nodes. Indicates missing nodes The set of valid neighboring nodes, Represents the set of valid nodes in the neighborhood. Valid nodes in Indicates missing nodes With neighboring valid nodes Interpolation weights based on distance between them. Indicates missing nodes axial displacement, Indicates the effective nodes in the neighborhood. axial displacement, Indicates missing nodes The bends and turns, Indicates the effective nodes in the neighborhood. The bending angle; based on the calculated missing node. axial displacement With bending angle Construct the local state data corresponding to the missing nodes estimated by the interpolation; The state feature subvectors of the effective node set obtained by integrating the reduced-order solution The local state data corresponding to the missing nodes estimated by interpolation are combined to obtain the complete global node state feature vector.

9. The system according to claim 1, characterized in that, The analysis computer reconstructs the overall state distribution of the submarine cable based on the acquired global node state feature vector. Specifically, it extracts the axial displacement and bending angle of each node from the acquired global node state feature vector; and reconstructs the axial stress, bending stress, and overall state distribution at any location on the submarine cable based on the extracted axial displacement and bending angle of each node, as shown below: ； ； ； in, Indicates the submarine cable in arc length coordinates Place Axial stress at time t, This represents the elastic modulus of the submarine cable material. This represents the axial strain obtained based on the extracted axial displacements of each node. Indicates the submarine cable in arc length coordinates Place Bending stress at any moment, This represents the bending moment obtained based on the extracted bending angles of each node. This represents the distance from the calculation point to the neutral axis of the submarine cable section. The moment of inertia of the submarine cable section is represented by the moment of inertia of the submarine cable section. Indicates the submarine cable in arc length coordinates Place The overall state distribution at time t.

10. A LoRa-based remote wireless telemetry method for submarine cable status applied to the system described in any one of claims 1 to 9, characterized in that, include: Each status monitoring node collects local status data of the submarine cable and transmits it via a wireless communication module; The relay node receives the local status data and forwards it to the shipborne monitoring center. The shipboard monitoring center's analysis computer monitors the communication link status in real time to identify the set of valid nodes that actually receive data and the set of missing nodes that experience packet loss. It extracts the spatial-physical correlation constraint matrix of the submarine cable based on all the monitored nodes, calculates the missing node ratio, and compares this ratio with a preset first packet loss threshold and a second packet loss threshold, where the first packet loss threshold is less than the second packet loss threshold. If the missing ratio is below the first packet loss threshold, it uses the current state change rate and spatial correlation of the local state data corresponding to the valid node set, combined with the historical local state data of the missing nodes, to perform trend filling on the local state data corresponding to the missing nodes, constructing a complete equivalent state vector. A state reconstruction model is then constructed using the spatial-physical correlation constraint matrix of the submarine cable for global solution. If the missing ratio is greater than the first packet loss threshold but less than the second packet loss threshold, a confidence weight that dynamically decays with the communication interruption time is introduced into the local state data after missing node filling in the state reconstruction model. If the missing proportion is greater than the second packet loss threshold, feature sub-blocks corresponding to the set of effective nodes in the spatial physical association constraint matrix are extracted for order reduction calculation, and the local state data corresponding to the missing node is restored by distance-based spatial interpolation method, so as to combine and obtain the complete global node state feature vector. Based on the acquired global node state feature vectors, the state distribution of the entire submarine cable is reconstructed, and hierarchical early warnings are output.

Images