Multi-dimensional dynamic submarine cable motion performance quantitative evaluation method

Abstract

The invention relates to the field of submarine cable motion performance, and discloses a multi-dimensional dynamic submarine cable motion performance quantitative evaluation system, which comprises a sensing layer used for acquiring multi-source data of an environment and a structure response of a submarine cable, and realizing time synchronization through a unified time service module; the edge calculation and model assimilation layer is used for establishing a geometric and boundary model of the submarine cable based on the multi-source data, generating a prediction state, and assimilating the prediction state and the observation data under the constraint of a non-extensible condition, a curvature and stress upper limit and a grounding point complementary condition to obtain a full-length correction state; and an index calculation and comprehensive evaluation layer. According to the method, numerical prediction and multi-source observation are assimilated under a physical constraint condition, so that the expansion of limited observation information to a full-length correction state is realized, the problem that the state is invisible due to sparse observation or local loss in the existing method is solved, and the integrity and reliability of submarine cable state evaluation are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of submarine cable motion performance, in particular to a multi-dimensional dynamic submarine cable motion performance quantitative evaluation method. BACKGROUND

[0002] As an important infrastructure for offshore wind farm grid connection, cross-sea power transmission and communication transmission, submarine cables need to withstand complex marine environmental effects, including wave, tidal current, eddy current and seabed contact coupling of multiple factors during long-term operation. Under deep water or complex sea conditions, the bending, tension, touchdown point stability and fatigue damage of the submarine cable directly affect its service life and operation safety, so it is of great significance to reliably quantify the motion performance of the submarine cable.

[0003] Existing submarine cable performance evaluation methods mostly rely on static design calculation or single data source monitoring methods. For example, one method simulates the stress and shape evolution of the submarine cable through numerical simulation, but due to the uncertainty and dynamics of the environmental boundary conditions in actual operation, the simulation results often deviate from the true state; another method relies on local sensors such as tension meters or fiber optic grating sensors to monitor the stress and strain of some key points of the submarine cable, but due to the sparseness of observation, it cannot fully reflect the dynamic response of the entire length of the submarine cable; some methods try to combine numerical models and observation data, but mostly stay in data comparison or single-dimensional index calculation, lacking a unified physical constraint assimilation mechanism, resulting in insufficient stability and consistency of the results.

[0004] In addition, traditional evaluation often only targets a single performance indicator, such as top tension or fatigue life, making it difficult to consider multiple factors such as touchdown point stability, vortex-induced vibration risk and wear and tear. There is a clear coupling relationship between different performance indicators, and if there is a lack of comprehensive quantitative means, the overall risk may still increase even if the local indicators are qualified. Existing methods generally lack an adaptive mechanism, and cannot automatically adjust the weights of different indicators according to the sea conditions, resulting in overemphasis on instantaneous risk in calm sea conditions and failure to highlight tension and extreme value risk in extreme sea conditions, thereby affecting the accuracy of the evaluation. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a multi-dimensional dynamic submarine cable motion performance quantitative evaluation method, which can integrate multi-source data, reconstruct the full-length state under physical constraints, realize unified calculation and adaptive comprehensive evaluation of multi-dimensional indicators, and form an evaluation and operation strategy closed loop to improve the safety and reliability of submarine cable operation.

[0006] To achieve the above purpose, the present application is implemented by the following technical solutions: a multi-dimensional dynamic submarine cable motion performance quantitative evaluation system, comprising: A perception layer for collecting multi-source data of the environment and structural response of the submarine cable and achieving time synchronization through a unified timing module; An edge computing and model assimilation layer for establishing a geometric and boundary model of the submarine cable based on the multi-source data, calling a dynamic modeling module to generate a predicted state, and assimilating the predicted state with observation data under the constraints of inextensibility, curvature and stress upper limit, and touchdown point and seabed complementarity conditions to obtain a full-length corrected state; An index calculation and comprehensive evaluation layer for calculating the touchdown point stability, mechanical safety margin, fatigue damage rate, vortex-induced vibration risk, extreme value overrun probability, and touchdown point wear risk based on the corrected state, and adaptively adjusting the weights of each index based on the environmental severity index to generate a comprehensive performance index; A decision feedback layer for outputting strategy suggestions for laying speed adjustment, top tension correction, and ship bow angle correction when the comprehensive performance index exceeds a threshold value, and checking the strategy effect based on new data in the next cycle to form a closed loop of evaluation and strategy.

[0007] Preferably, the perception layer includes a distributed optical fiber sensing unit embedded in the submarine cable sheath and an optical fiber grating sensor arranged at a key node to achieve full-length coverage and local high-precision strain and temperature collection.

[0008] Preferably, the multi-source data includes flow velocity field and wave elements provided by a numerical model, profile flow velocity collected by an ADCP, surface wave elements obtained by a wave radar, top tension and attitude information collected by a cable end tension gauge and an IMU, and strain and temperature obtained by a distributed optical fiber sensor.

[0009] Preferably, the geometric and boundary model in the edge computing and model assimilation layer includes the length, diameter, linear density, bending stiffness, and axial stiffness of the submarine cable, and determines the initial position of the touchdown point in combination with the laying plan and seabed profile, while introducing the friction coefficient and seabed normal stiffness as the contact conditions of the touchdown point.

[0010] Preferably, the edge computing and model assimilation layer includes an observation mapping module for converting the strain data of the distributed optical fiber sensor into tension and curvature, converting the wavelength drift data of the optical fiber grating sensor into strain and eliminating the thermal effect through temperature compensation, and converting the angular velocity and acceleration of the IMU into curvature and acceleration for establishing a corresponding relationship with the dynamic predicted state.

[0011] Preferably, the edge computing and model assimilation layer adopts constraints including inextensible approximation, curvature and stress upper limit, and touchdown point and seabed complementarity conditions when assimilating the predicted state with the observation data to ensure that the corrected state remains reasonable physically.

[0012] Preferably, the index calculation and comprehensive evaluation layer calculates the touchdown point stability index, the mechanical safety margin index of tension and curvature, the fatigue damage rate based on stress history, the vortex-induced vibration risk based on reduced velocity, the probability of exceeding the limit based on extreme value distribution and the touchdown point wear risk on the basis of the modified state, and keeps consistent in space and time.

[0013] Preferably, the comprehensive evaluation layer dynamically adjusts the weight of each dimension index based on the environmental severity index, so as to focus on long-term fatigue damage in smooth sea conditions and focus on tension and extreme value risk in extreme sea conditions, thereby adapting to different environmental scenarios.

[0014] Preferably, the decision feedback layer issues the strategy to the laying console or dynamic positioning system after generating the strategy, and verifies the effect of the strategy by newly collected data in the next period, and if it is found that there is a deviation between the actual state and the expectation, the confidence matrix and the weight function are automatically adjusted.

[0015] A multi-dimensional dynamic submarine cable motion performance quantitative evaluation method, comprising the following steps: S1. Data acquisition and unified time base, joint acquisition of multi-source data is realized through numerical model and in-situ sensor, including numerical prior part and in-situ observation part, and time stamp calibration and confidence evaluation are carried out in a unified space-time reference frame; S2. Submarine cable geometry and boundary modeling, a geometry and boundary model of the submarine cable is established based on the multi-source data, the model includes the length, diameter, linear density, bending stiffness and axial stiffness of the submarine cable, the touchdown point boundary is determined in combination with the laying plan and the seabed profile, and the friction coefficient and the seabed normal stiffness are introduced as the contact condition; S3. Physical prior dynamics prediction, after establishing the geometry and boundary model, the dynamics prediction model of the submarine cable is constructed by using Morison equation and modal reduction method; S4. Observation mapping and anomaly suppression, the observation and the predicted state are connected by introducing mechanical conversion, and the observation is pretreated by Huber loss function and time sequence consistency filtering before assimilation; S5. Physical constraint data assimilation, the predicted state and the observation are assimilated to obtain a full-length modified state; S6. Multi-dimensional performance quantification, the touchdown point stability, the mechanical safety margin, the fatigue damage rate, the vortex-induced vibration risk, the probability of exceeding the limit and the touchdown point wear risk are calculated on the basis of the modified state, and the weight of each performance index is adaptively adjusted according to the environmental severity index, so as to generate a comprehensive performance index, and a strategy suggestion is formed according to the comprehensive performance index.

[0016] The application provides a multi-dimensional dynamic submarine cable motion performance quantitative evaluation method. 1. This invention, by assimilating numerical prediction with multi-source observations under physical constraints, extends the limited observation information to the full-length corrected state, solving the problem of invisible state caused by sparse or local missing observations in existing methods, thereby significantly improving the completeness and reliability of submarine cable state assessment.

[0017] 2. This invention calculates multiple performance indicators such as contact point stability, mechanical safety margin, fatigue damage rate, vortex-induced vibration risk, extreme value exceedance probability, and contact point wear in the same state domain. This avoids the fragmentation and inconsistency of results caused by independent calculations of different models in traditional methods, and keeps the evaluation results consistent in time and space.

[0018] 3. This invention, through a weighted adaptive mechanism based on the environmental severity index, can automatically adjust the emphasis of each performance indicator according to the sea state. In a stable environment, it pays more attention to fatigue damage, and in an extreme environment, it highlights tension and extreme value risks, thereby improving the accuracy of the comprehensive evaluation and the adaptability of the scenario.

[0019] 4. This invention constructs a closed-loop process of evaluation-strategy-re-evaluation by combining comprehensive performance index with operation and maintenance strategy generation and effect verification. This enables the system to continuously optimize strategies and achieve self-learning during operation, avoiding the problem of monitoring and operation and maintenance being disconnected in traditional methods, and effectively improving the real-time performance and proactiveness of submarine cable operation and maintenance. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0021] The technical solutions in the embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings.

[0022] Example: Please see the appendix Figure 1 This invention provides a multi-dimensional dynamic submarine cable motion performance quantitative evaluation method, including: S1. Data acquisition and unified time base; First, multi-source data is jointly acquired using numerical models and in-situ sensors, including both numerical priors and in-situ observations. Timestamp calibration and confidence assessment are then performed within a unified spatiotemporal reference framework. This not only provides high-quality data support for subsequent modeling and assimilation but also enhances overall reliability through cross-validation of multi-source data, laying a data foundation for subsequent dynamic fusion. Specifically, the numerical prior part includes the velocity field output by the power flow numerical model. And the effective wave height output by the third-generation spectral wave model , main frequency and main wave direction This information constitutes the boundary input of the sea state; the advantage of numerical models lies in their spatial continuity and temporal predictive capabilities, which can reflect the fluid characteristics of unobserved areas in advance, thus providing a reasonable initial distribution for data assimilation. The in-situ observation part includes the profile velocity obtained by ADCP. Surface wave elements acquired by wave radar, top tension acquired by cable end tension meter and IMU Along with angular velocity acceleration and strain acquired by a distributed fiber optic sensing system In relation to temperature, ADCPs are typically deployed along both sides of the submarine cable path to capture profile shear flow characteristics; distributed optical fibers are bonded along the entire length of the cable, enabling continuous strain monitoring with meter-level resolution; IMUs and tension meters are located at the cable end nodes to provide boundary condition constraint information.

[0023] All data is uniformly fed with a time reference provided by the access module upon connection, and a confidence matrix is ​​generated based on signal-to-noise ratio, packet loss rate, and sensor health. This allows for dynamic weighting in subsequent data fusion. The confidence matrix is ​​not only used for sensor weight allocation but also for anomaly diagnosis and sensor node redundancy scheduling. When the data quality of a certain channel deteriorates, the system automatically increases the weight of other channels to ensure the robustness of the overall monitoring network.

[0024] S2 Submarine Cable Geometry and Boundary Modeling After data preparation, the system establishes a geometric and boundary model based on the submarine cable design parameters and laying conditions. The submarine cable is discretized into a sequence of nodes, with arc length parameters as follows: ,length ,diameter Linear density Bending stiffness and axial stiffness It is known that parameters directly determine the dynamic response of the submarine cable. For example, bending stiffness affects the vibration modes and eddy-induced response threshold, while axial stiffness determines tensile deformation characteristics and tension transmission efficiency. The upper boundary is constrained by position and attitude provided by the IMU, and by tension gauges... This is then used as direct input. The contact point boundary is initially provided by the laying plan and the seabed profile obtained from multibeam bathymetry, and the contact condition is determined by the friction coefficient. With normal stiffness The seabed friction coefficient is typically estimated through sampling laboratory tests or acoustic inversion, while the normal stiffness can be calibrated using finite element simulation combined with actual monitoring data. Unlike traditional modeling, the boundary parameters in this method are not static inputs but can be corrected based on observational feedback in subsequent assimilation stages. This makes the boundary conditions no longer single inputs but rather part of the dynamic evolution of the system.

[0025] S3 Physical Prior Dynamics Prediction After the geometry and boundary model is established, the method constructs the dynamics prediction model of the cable using the Morison equation and the modal reduction method. The cable state vector is defined as The discretized dynamics form is , where the external environmental excitation is Under the action of the fluid, the unit length resistance term is , The added mass term is written as

[0026] where is the relative velocity. The introduction of the Morison equation here is based on the classical mechanism of the interaction between “slender flexible structure—wave current”: the inertia term reflects the added mass effect of the fluid acceleration on the cable, and the resistance term reflects the viscous resistance brought by the velocity shear and vortex shedding. Through modal reduction, the partial differential equation of the continuous system can be converted into a finite-dimensional ordinary differential equation, thereby realizing real-time computability. The model gives the prediction state of the curvature, tension and displacement, which is used as the prior for assimilation; The system adopts a segmented modal splicing strategy, which splits the dynamic response under different working conditions into a finite modal space and then synthesizes it, improving the adaptability and computational efficiency of the model. In field testing, the calculation time of this method is reduced by about 67% compared to traditional finite element simulation, and the displacement prediction error is reduced by about 35%.

[0027] S4 Observation Mapping and Anomaly Suppression In order to establish a connection between the observation and the prediction state, the method introduces a mechanical conversion in the observation mapping link. The axial strain signal of the distributed optical fiber is mapped to the curvature and tension through the cross-sectional relationship:

[0028] where is the distance from the fiber position to the neutral axis. The wavelength shift of the FBG sensor satisfies , which is compensated through the temperature channel. The angular velocity and acceleration measured by the IMU are converted into node curvature and acceleration through rigid body kinematics, and the tension meter directly constrains the top tension state. The essence of observation mapping is to project “local quantities measured by sensors” into “the overall state space of the structure”, thereby realizing one-to-one correspondence between the observation domain and the state domain. This allows data from different sensing channels to complement each other, forming a higher-dimensional state perception capability.

[0029] All observations are unified in the observation equation

[0030] The observations are preprocessed by Huber loss function and temporal consistency filtering to eliminate outliers. Huber function has the advantages of both L1 and L2 norms, preserving quadratic convergence when error is small and suppressing outliers when error is large. Temporal filtering ensures the physical continuity of data and avoids global state drift caused by single-point outliers.

[0031] S5. Physical constraint data assimilation In the data assimilation process, the dynamics prediction and observation data are fused under physical constraints.

[0032] The defined optimization objective is:

[0033] where is the predicted state, is the preprocessed observation, is the physical constraint term, including the inextensible approximation , curvature upper limit , stress upper limit , and contact point complementarity condition , , The constraint conditions reflect the basic physical properties of the cable: the inextensible approximation ensures length conservation, the curvature upper limit prevents structural overbending that can cause fatigue, the stress upper limit constrains the material yield limit, and the contact point complementarity condition ensures that the mechanical behavior of the contact zone conforms to the actual friction law.

[0034] By solving this objective function through the augmented Lagrangian method or the constraint lossless Kalman filter, the full-length corrected state can be obtained. Unlike traditional simple interpolation or linear weighted assimilation, this method directly embeds mechanical constraints into state estimation optimization, naturally converging the solution space to the physically feasible region, and significantly reducing non-physical solutions caused by observation errors.

[0035] The unexpected effect of this step is that even if some observation data is missing, the physical constraints can still ensure that the estimated state is mechanically reasonable, thereby extending the limited observation to full-length visibility and limiting the impact of abnormal observations within the feasible region.

[0036] S6. Multi-dimensional performance quantification On the basis of the corrected state, this method uniformly calculates multi-dimensional performance indicators. Touch point stability is measured by and the length of the suspended segment; mechanical safety margin is measured by the quantile of tension and the root mean square of curvature Calculation; fatigue damage rate is obtained from the stress history using rainflow counting. The Miner's Law is expressed as follows: ; The risk of vortex-induced vibration is reduced by velocity The lateral displacement variance is calculated by intersecting the locked interval; the peak over-threshold probability is estimated using the peak over-threshold (POT) method, which fits the tension peak sequence to a generalized Pareto distribution for a given return period; and the contact point wear risk is obtained by time integration of contact pressure and tangential slip energy density. These indicators constitute a multi-dimensional perspective for structural performance evaluation: safety margin and fatigue rate reflect "tolerance level," vortex-induced risk and extreme value probability reflect "vulnerability under random conditions," and contact point wear reflects "long-term operation and maintenance risks." The synergistic calculation of multiple indicators avoids the one-sidedness of a single indicator and achieves a holistic profile of the entire life cycle performance.

[0037] Since these indicators all originate from the same state domain, they are naturally consistent in time and space. When fatigue damage and vortex-induced vibration hotspots coincide at the same location, the subsequent weight allocation will automatically amplify the importance of that segment.

[0038] To dynamically integrate various indicators under different sea states, this method constructs an environmental severity index.

[0039] Each quantity has been normalized. The system is based on... Adaptively update the weights of each dimension And normalize each indicator to The final comprehensive performance index is defined as follows:

[0040] when When the preset threshold is exceeded, the system will trigger an alarm and generate optimized suggestions for deployment speed, tension settings, and ship attitude based on state estimation. After new observations enter the next cycle, the effectiveness of the strategy will be verified, and... The weight function is updated to form a closed loop of evaluation-policy-re-evaluation; S7. Decision Feedback and Closed-Loop Control Limited observations are expanded to the full-length state under physical constraints, allowing previously invisible segments to be reconstructed; multi-dimensional indicators are calculated uniformly within the same state domain, making them comparable and triggering a hotspot priority mechanism under coupled conditions; environmental severity-driven weight adjustments enable the comprehensive evaluation to adapt to different sea states; boundary conditions are dynamically updated based on feedback from assimilation results, rather than being statically set; finally, the evaluation results directly influence the operation and maintenance strategy through a closed-loop mechanism and are continuously optimized under the correction of new observations.

[0041] Based on the implementation of the above method, a multi-dimensional dynamic submarine cable motion performance quantitative evaluation system is proposed, comprising: The perception layer, in the perception layer, the system is laid out with multiple types of sensing devices, these sensors include both environmental observation devices and structural response monitoring devices. The environmental observation part can use ship-mounted or platform-mounted ADCP arrays to obtain profile flow velocity data at different water depth layers, use X-band or Ku-band wave radars to obtain surface wave elements and two-dimensional wave spectrum, and can also access prior data output by tidal current numerical models and third-generation spectral wave models through satellites, so as to still provide reliable environmental input in the case of insufficient on-site observation.

[0042] The structural response part includes high-precision tension meters and six-axis IMUs installed at the cable end for collecting top tension and attitude information; at the same time, a distributed optical fiber sensing system is laid out at the key nodes of the submarine cable, in which the DAS can obtain the axial strain changes along the entire length of the cable at a spatial resolution of 1-5 meters, and the FBG array can be laid out at a spacing of 50-100 meters to provide high-precision local strain and temperature data. As an embodiment, the DAS optical fiber can be embedded in the submarine cable sheath, and the FBG grating can be fixed at specific nodes by a wrap-around or adhesive method, which ensures full-length coverage and has the ability of high-precision compensation. All sensors are clocked by a unified time module (such as a GNSS receiver or an IEEE1588 PTP module) for clock calibration, so that data from different sources can be aligned at a millisecond level of accuracy, creating conditions for subsequent time series fusion.

[0043] The edge computing and model assimilation layer receives multi-source data from the perception layer and completes format conversion and abnormality rejection through a data access gateway. The data preprocessing unit filters and interpolates the observation signals, such as using Huber loss function to eliminate short-time glitches, using time window consistency test to exclude underwater bubbles or radar artifacts, and then constructing a confidence matrix to assign dynamic weights to each data source. On this basis, the system calls the built-in dynamic modeling module, which calculates the hydrodynamic load based on the Morison equation and establishes a discrete dynamic model of the submarine cable by combining the modal reduction method, in which the state vector includes position, velocity, curvature and tension.

[0044] The vector includes position, velocity, curvature and tension. The model prediction part predicts the prior state at the next time point through the state equation , and the external input is composed of the background flow velocity and the wave-induced orbital velocity At the same time, the perception mapping module maps the sensing signals to the quantities corresponding to the state vector by using structural mechanics relationship. For example, the DAS strain is mapped by the formula The FBG wavelength shift is converted to tension and curvature, and the temperature effect is eliminated by compensation model. The IMU angles and accelerations are transformed to curvature and acceleration by installation matrix. Then, the assimilation module fuses the predicted state and observations by using a physical constraint solver. The objective function contains observation error term, prediction error term and physical constraint term, which includes inextensibility condition, curvature and stress upper limit, and complementary condition between touchdown point and seabed. The optimization solution can use augmented Lagrangian method or constrained unscented Kalman filter, and finally outputs the full-length corrected state. As an embodiment, the calculation unit can be implemented by an industrial edge computer with GPU acceleration function, which can complete the state solution of hundreds of nodes in real time within a time window of 5-10 minutes.

[0045] The index calculation and comprehensive evaluation layer calculates various performance indicators based on the corrected full-length state. The touchdown point stability is obtained by analyzing the drift rate of TDP and the length of suspension, the mechanical safety margin is represented by the quantile of tension and the root mean square value of curvature, and the fatigue damage rate is calculated by rain flow counting and Miner's rule, the formula is , . The extreme value risk uses the threshold gate model to fit the tension peak value to estimate the exceedance probability under the return period; the vortex-induced vibration risk determines the transverse vibration energy by the reduced velocity and the lock-in interval criterion; the touchdown point wear risk is calculated by the integral of contact pressure and tangential slip energy density. All indicators are calculated in the same state domain, so they are naturally aligned in time and space, avoiding the inconsistency caused by splicing different models in traditional methods. More importantly, when fatigue damage and vortex risk occur in the same section, the system will automatically amplify the weight of that section, forming a 'hot spot priority' risk identification mechanism.

[0046] In the comprehensive evaluation link, the system constructs the environmental severity index ESI(τ)=α1H̄e+α2Ū+α3ω̄p+α4Δ̄θ, and dynamically adjusts the weights of each indicator according to the index. When the environment is relatively stable, the system automatically allocates more weight to the fatigue damage indicator; while in extreme sea conditions, the system increases the weight of tension and extreme value risk. Through this scene perception mechanism, the comprehensive performance index MPI(s,t)=∑iwi(ESI(τ))qi(s,t) can be generated in real time, and displayed in the form of arc length heat map and time curve. In an example, MPI can be presented in real time in the form of dynamic chart through industrial tablet or operation and maintenance console, and operation and maintenance personnel can grasp the overall health status of the submarine cable at a glance.

[0047] The decision feedback layer triggers an alarm immediately when the MPI exceeds the preset threshold, and generates operation and maintenance strategies according to the correction state. These strategies include reducing the laying speed, adjusting the top tension setting value, and correcting the heading angle of the dynamic positioning ship. All suggestions are provided in interval form to ensure the feasibility of actual operation. The strategy issuing interface can be directly connected to the laying console or the DP system to realize semi-automatic or automatic strategy execution. More importantly, the assimilation of the next time window verifies the execution effect of the previous round of strategies. If differences are found between the actual state and the expected state, the system will automatically adjust the confidence matrix and the weight function, thus forming a self-learning closed loop of "evaluation-strategy-re-evaluation". As an example, strategy suggestions can be directly pushed to operators through the central console of the submarine cable laying ship, and combined with visual alarm prompts to achieve immediate response.

[0048] The system framework forms a complete implementation scheme that can monitor, dynamically evaluate and self-learn and optimize in real time through the multi-source sensors of the perception layer, the dynamic prediction and constraint assimilation of the edge computing layer, the multi-dimensional quantification and scene adaptive weighting of the index calculation layer, and the closed-loop strategy generation of the decision feedback layer. Compared with the prior art, the system not only can reconstruct the full-length state under sparse observation, but also can automatically amplify risk hotspots when multiple indicators are coupled, and can realize adaptive adjustment of weights under different sea conditions, thereby improving the evaluation accuracy and operation reliability.

[0049] Although embodiments of the present application have been shown and described, it is to be understood that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-dimensional dynamic submarine cable motion performance quantification evaluation system, characterized in that: The method comprises the following steps: a perception layer for collecting multi-source data of the environment and structural response of the submarine cable and realizing time synchronization through a unified timing module; an edge computing and model assimilation layer for establishing a geometric and boundary model of the submarine cable based on the multi-source data and generating a predicted state, and assimilating the predicted state with observation data under the constraints of inextensibility, curvature and stress upper limit and complementary conditions of the touchdown point and seabed to obtain a full-length corrected state; an index calculation and comprehensive evaluation layer for calculating the touchdown point stability, mechanical safety margin, fatigue damage rate, vortex-induced vibration risk, extreme value overrun probability and touchdown point wear risk based on the corrected state, and adaptively adjusting the weights of each index based on the environmental severity index to generate a comprehensive performance index; a decision feedback layer for outputting strategy suggestions of laying speed adjustment, top tension correction and ship heading angle correction when the comprehensive performance index exceeds a threshold value, and checking the strategy effect based on new data in the next cycle to form a closed loop of evaluation and strategy.

2. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system according to claim 1, characterized in that: The perception layer comprises a distributed optical fiber sensing unit embedded in the submarine cable sheath and a fiber Bragg grating sensor arranged at the node to realize full-length coverage and local high-precision strain and temperature collection.

3. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system according to claim 1 or 2, characterized in that: The multi-source data comprises a flow field and wave elements provided by a numerical model, profile flow velocities collected by an ADCP, surface wave elements obtained by a wave radar, top tension and attitude information collected by a cable end tension gauge and an IMU, and strain and temperature obtained by a distributed optical fiber sensor.

4. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system of claim 1, wherein: The geometric and boundary model in the edge computing and model assimilation layer comprises the length, diameter, linear density, bending stiffness and axial stiffness of the submarine cable, and determines the initial position of the touchdown point in combination with the laying plan and seabed profile, and introduces the friction coefficient and seabed normal stiffness as the contact conditions of the touchdown point.

5. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system of claim 1, wherein: The edge computing and model assimilation layer comprises an observation mapping module for converting the strain data of the distributed optical fiber sensor into tension and curvature, converting the wavelength drift data of the fiber Bragg grating sensor into strain and eliminating the thermal effect through temperature compensation, and converting the angular velocity and acceleration of the IMU into curvature and acceleration for establishing a corresponding relationship with the dynamic predicted state.

6. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system of claim 1, wherein: When the edge computing and model assimilation layer assimilates the predicted state with the observation data, the constraints of inextensibility approximation, curvature and stress upper limit and the complementary conditions of the touchdown point and seabed are adopted to ensure that the corrected state is physically reasonable.

7. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system of claim 1, wherein: The index calculation and comprehensive evaluation layer calculates the touchdown point stability index, mechanical margin index represented by tension and curvature, fatigue damage rate based on stress history, vortex-induced vibration risk based on reduced velocity, overrun probability based on extreme value distribution and touchdown point wear risk based on the corrected state, and keeps consistent in space and time.

8. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system of claim 1, wherein: The comprehensive evaluation layer dynamically adjusts the weights of each dimension index based on the environmental severity index, so as to focus on long-term fatigue damage in smooth sea conditions and focus on tension and extreme value risk in extreme sea conditions, thereby adapting to different environmental scenarios.

9. The multi-dimensional dynamic submarine cable motion performance quantification evaluation system of claim 1, wherein: The decision feedback layer will issue the strategy to the laying console or dynamic positioning system after generating it, and verify the effect of the strategy through newly collected data in the next cycle. If a deviation is found between the actual state and the expected state, the confidence matrix and weight function will be automatically adjusted.

10. The method of any one of claims 1-9, wherein: The method comprises the following steps: S1. Data acquisition and unified time base, joint acquisition of multi-source data through numerical models and in-situ sensors, including numerical prior part and in-situ observation part, and time stamp calibration and confidence evaluation in a unified space-time reference frame; S2. Modeling of the geometry and boundary of the submarine cable, the geometry and boundary model of the submarine cable is established based on the multi-source data, the model includes the length, diameter, linear density, bending stiffness and axial stiffness of the submarine cable, the boundary of the touchdown point is determined in combination with the laying plan and the seabed profile, and the friction coefficient and the normal stiffness of the seabed are introduced as the contact condition; S3. Physical prior dynamic prediction, after the establishment of the geometry and boundary model, the dynamic prediction model of the submarine cable is constructed by using the Morison equation and the modal reduction method, and the predicted state is given; S4. Observation mapping and anomaly suppression, the observation and the predicted state are connected by introducing mechanical conversion, and the observation is pretreated by Huber loss function and time series consistency filtering before assimilation; S5. Physical constraint data assimilation, the predicted state and the observation are assimilated to obtain a full-length corrected state; S6. Multi-dimensional performance quantification, based on the corrected state, the touchdown point stability, mechanical safety margin, fatigue damage rate, vortex-induced vibration risk, extreme value overrun probability and touchdown point wear risk are calculated, the weight of each performance index is adaptively adjusted according to the environmental severity index, so as to generate a comprehensive performance index, and a strategy suggestion is formed according to the comprehensive performance index.

Images