Mooring system life prediction method and device, electronic equipment and storage medium

Abstract

The present disclosure provides a mooring system life prediction method and device, electronic equipment and storage medium, wherein the method comprises: constructing a coupling damage physical equation of corrosion and fatigue, the coupling damage physical equation of corrosion and fatigue comprising the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue; determining a physical equation of comprehensive damage degree based on the coupling damage physical equation of corrosion and fatigue; embedding the coupling damage physical equation and the physical equation of comprehensive damage degree into an initial prediction model; training the initial prediction model to obtain a target prediction model; obtaining time series monitoring data of a to-be-tested mooring system; inputting the time series monitoring data of the to-be-tested mooring system into the target prediction model to determine the corresponding comprehensive damage degree of the to-be-tested mooring system; and determining the remaining life of the to-be-tested mooring system based on the comprehensive damage degree.

Description

Technical Field

[0001] This disclosure relates to the field of marine data processing technology, and in particular to a method, apparatus, electronic device and storage medium for predicting the lifespan of a mooring system. Background Technology

[0002] The mooring system of an offshore floating platform is its lifeline for long-term safe berthing. Over its decades-long service life, the mooring chain (usually a high-strength, strapped chain) continuously withstands dynamic fatigue loads caused by harsh marine environments (wind, waves, and currents), as well as corrosion damage caused by prolonged seawater immersion, alternating wet and dry conditions, microbial adhesion, and electrochemical processes. As global marine engineering continues to expand into deep-sea and extreme environments, the service conditions faced by mooring systems are becoming increasingly demanding, and their structural integrity directly impacts the platform's overall safety, operational continuity, and environmental protection.

[0003] Currently, the prediction of the remaining life of mooring systems mainly relies on empirical formulas based on standard specifications, periodic underwater inspections, and simplified numerical simulations. However, these methods have significant limitations: empirical formulas (such as those based on SN curves and Miner's linear cumulative damage rule) cannot accurately characterize the coupled nonlinear effects of corrosion and fatigue; periodic inspections are costly and have long intervals, and can only obtain status information at discrete time points, making it difficult to achieve continuous health monitoring and early warning; while traditional numerical simulations often treat corrosion as uniform cross-sectional loss or decouple fatigue and corrosion processes in calculations, ignoring the physical nature of their interaction. Summary of the Invention

[0004] This disclosure provides a method, apparatus, electronic device, and storage medium for predicting the lifespan of a mooring system, in order to at least solve the above-mentioned technical problems existing in the prior art.

[0005] According to a first aspect of this disclosure, a method for predicting the lifespan of a mooring system is provided, the method comprising:

[0006] A coupled damage physics equation for corrosion and fatigue is constructed, which includes the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue.

[0007] A physical equation for determining the comprehensive damage degree is derived from the coupled damage physical equations of corrosion and fatigue.

[0008] The coupled damage physical equation and the comprehensive damage degree physical equation are embedded into the initial prediction model;

[0009] The initial prediction model is trained to obtain the target prediction model;

[0010] Acquire time-series monitoring data of the mooring system under test;

[0011] The time-series monitoring data of the mooring system under test is input into the target prediction model to determine the comprehensive damage degree of the mooring system under test.

[0012] Based on the comprehensive damage level, the remaining life of the mooring system under test is determined.

[0013] In one possible implementation, the construction of a coupled damage physics equation for corrosion and fatigue, the coupled damage physics equation including the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue, includes:

[0014] Based on the coupling effect of fatigue on corrosion, a physical equation for corrosion damage is constructed, which includes the effect of stress on corrosion rate.

[0015] Based on the coupling effect of corrosion on fatigue, a fatigue damage physical equation is constructed, which includes the influence of corrosion depth on crack propagation rate and the influence of hydrogen atoms on crack propagation rate.

[0016] In one possible implementation, constructing the physical equation of corrosion damage includes: determining the physical equation of corrosion damage based on the following formula:

[0017]

[0018] in, The corrosion rate is dimensionless. The coefficients for the corrosion model are the overall coefficients. Let the dimensionless state variable be the corrosion depth. The saturation index The mechanochemical coupling factor characterizes the effect of stress on the corrosion rate. For normalized load, These are the material-related fitting coefficients;

[0019] The construction of the fatigue damage physical equation includes: determining the fatigue damage physical equation based on the following formula:

[0020]

[0021] in, The crack propagation rate is a dimensionless value. These are the comprehensive coefficients of the fatigue model. For material constants, Let be the dimensionless state variable representing the crack length. The geometric coupling factor characterizes the effect of corrosion depth on crack propagation rate. Geometric dimension ratio, It is a hydrogen embrittlement sensitizer, characterizing the effect of hydrogen atoms on crack propagation rate.

[0022] In one possible implementation, the physical equation for determining the overall damage degree based on the coupled damage physical equation of corrosion and fatigue includes:

[0023] Based on the coupled damage physics equations of corrosion and fatigue, the corrosion depth and crack length at the current moment are determined.

[0024] Based on the current corrosion depth and crack length, a physical equation for the overall damage degree is determined; wherein, the physical equation for the overall damage degree is shown in the following formula:

[0025]

[0026] in, The overall damage level at the current moment. The current corrosion depth. The crack length at the current moment. The weight of the current corrosion depth. The weight is the crack length at the current moment.

[0027] In one possible implementation, training the initial prediction model to obtain the target prediction model includes:

[0028] Obtain a sample training set, which includes time-series monitoring data of the sample mooring system;

[0029] The sample training set is input into the initial prediction model;

[0030] Construct a hybrid loss function for the initial prediction model, the hybrid loss function including a data fitting loss function, a physical consistency loss function, and an initial condition loss function;

[0031] The initial prediction model is trained based on the hybrid loss function to obtain the target prediction model.

[0032] In one embodiment, the data fitting loss function includes the error between the corrosion depth and crack length predicted by the initial prediction model and their respective first target values;

[0033] The physical consistency loss function includes the error between the corrosion rate and crack propagation rate predicted by the initial prediction model and their respective second target values.

[0034] In one possible implementation, determining the remaining lifetime of the mooring system under test based on the overall damage level includes:

[0035] Determine whether the overall damage degree is greater than or equal to 1;

[0036] When the overall damage level is greater than or equal to 1, it indicates that the mooring system under test has failed, and the time corresponding to the current overall damage level is determined as the total service life of the mooring system under test.

[0037] The remaining lifespan of the mooring system under test is determined based on the total service life.

[0038] According to a second aspect of this disclosure, a mooring system lifetime prediction apparatus is provided, the apparatus comprising:

[0039] A construction module is used to construct a coupled damage physics equation for corrosion and fatigue, wherein the coupled damage physics equation includes the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue.

[0040] The first determining module is used to determine the physical equation of the overall damage degree based on the coupled damage physical equation of corrosion and fatigue.

[0041] An embedding module is used to embed the coupled damage physical equation and the comprehensive damage degree physical equation into the initial prediction model;

[0042] The training module is used to train the initial prediction model to obtain the target prediction model;

[0043] The acquisition module is used to acquire time-series monitoring data of the mooring system under test;

[0044] The second determining module is used to input the time-series monitoring data of the mooring system under test into the target prediction model in order to determine the comprehensive damage degree of the mooring system under test.

[0045] The third determining module is used to determine the remaining life of the mooring system under test based on the comprehensive damage level.

[0046] According to a third aspect of this disclosure, an electronic device is provided, comprising:

[0047] At least one processor; and a memory communicatively connected to said at least one processor; wherein,

[0048] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods described in this disclosure.

[0049] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing the computer to perform the methods described in this disclosure.

[0050] The mooring system life prediction method, apparatus, electronic device, and storage medium disclosed herein fundamentally solve the shortcomings of existing technologies that simply superimpose the independent calculations of fatigue and corrosion by constructing physical equations that incorporate the bidirectional coupling effect of fatigue and corrosion. This accurately reconstructs the evolution process of fatigue and corrosion accelerating each other, significantly improving the accuracy of remaining life prediction. Furthermore, by embedding the physical equations of coupled damage and the physical equations of comprehensive damage into the prediction model, the model integrates physical mechanisms and data-driven approaches, avoiding the risk of absurd predictions by purely data-driven models under extreme or unseen conditions, thus improving the accuracy of model predictions.

[0051] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0052] The above and other objects, features, and advantages of this disclosure will become readily apparent from the following detailed description of exemplary embodiments, taken in conjunction with the accompanying drawings. Several embodiments of this disclosure are illustrated in the drawings by way of example and not limitation, in which:

[0053] In the accompanying drawings, the same or corresponding reference numerals indicate the same or corresponding parts.

[0054] Figure 1 A flowchart of the mooring system lifetime prediction method provided in the embodiments of this disclosure;

[0055] Figure 2 This is a complete mooring system model obtained through simulation in the embodiments of this disclosure;

[0056] Figure 3 This is a simulation model of a failed mooring system obtained in the embodiments of this disclosure;

[0057] Figure 4 A schematic diagram of the structure of the mooring system life prediction device provided in the embodiments of this disclosure;

[0058] Figure 5 A schematic diagram of the composition structure of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0059] To make the objectives, features, and advantages of this disclosure more apparent and understandable, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0060] The failure modes of mooring systems mainly manifest as fatigue fracture and corrosion thinning. On the one hand, the six-degree-of-freedom motion of the platform under complex sea conditions causes the mooring chain to be subjected to high-frequency cyclic tension, triggering the initiation and propagation of fatigue cracks. On the other hand, seawater corrosion (including uniform corrosion, pitting corrosion, and stress corrosion cracking) not only directly weakens the load-bearing cross-section of the chain links but also forms stress concentration sources on the material surface, significantly accelerating the initiation of fatigue cracks. These two damage mechanisms are not independent but exhibit a strong time-varying coupling effect: corrosion pits exacerbate local stress concentration, promoting fatigue crack initiation; crack propagation, in turn, alters the local electrochemical environment and strain field, further accelerating the corrosion process. The evolution of this coupled damage is highly nonlinear and stochastic, posing a significant challenge to traditional life assessment methods based on linear superposition and deterministic models.

[0061] Existing technologies for predicting the remaining life of mooring systems have significant limitations:

[0062] 1. Mechanism Separation: Traditional methods typically calculate corrosion damage (considered as uniform cross-sectional loss) and fatigue damage (based on SN curves or fracture mechanics) independently, then verify them using simple linear superposition (such as Miner's rule) during the design phase. This severely neglects the strong coupling effect between corrosion and fatigue. Corrosion pits, acting as stress concentration sources, significantly accelerate the initiation of fatigue cracks; while crack propagation alters the local electrochemical environment, accelerating corrosion. This coupling effect dominates during long-term service, leading to insufficient prediction accuracy in existing methods.

[0063] 2. Load simplification: During the design phase, regular waves or simplified load spectra are often used for fatigue analysis, which fails to fully account for the complex dynamic load history of the mooring system caused by the combined effects of random, non-stationary, and multi-directional coupled wind, waves and currents in the real marine environment, resulting in load input distortion.

[0064] 3. Rigid Predictive Models: While purely physical models (such as extended fracture mechanics models) can describe coupling mechanisms, these models are complex, and their parameters (such as coupling factors and crack propagation parameters) are highly dependent on hard-to-obtain field data and are difficult to update in real time. Purely data-driven machine learning methods, while able to learn from data, lack physical constraints. In extreme conditions not covered by training data or in long-term extrapolation predictions, they are prone to producing physically unreasonable results, resulting in poor reliability and generalization ability.

[0065] 4. Lack of real-time and adaptability: Existing assessments are mostly based on analysis during the design phase or periodic limited testing, and cannot combine real-time motion monitoring data and environmental measurement data of the platform to perform dynamic and adaptive remaining life updates.

[0066] Based on this, embodiments of this disclosure provide a method for predicting the lifetime of a mooring system. Figure 1A flowchart of the mooring system lifetime prediction method provided in the embodiments of this disclosure is shown below. Figure 1 As shown, the method includes:

[0067] Step 101: Construct the coupled damage physics equation of corrosion and fatigue. The coupled damage physics equation includes the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue.

[0068] Specifically, step 101 aims to establish a quantitative description of the mutually accelerating evolution of corrosion depth and crack length during the service life of the mooring system.

[0069] Without considering the coupling effect between fatigue and corrosion, we establish pure corrosion evolution equations and pure fatigue evolution equations.

[0070] First, based on Faraday's law, a pure corrosion evolution equation is established.

[0071] Specifically, the corrosion depth growth rate of a metal surface is directly proportional to the corrosion current density, as shown in the following formula (1):

[0072] (1)

[0073] in, Where M is the corrosion depth, n is the molar mass of the metal, n is the number of electrons participating in the reaction, and F is the Faraday constant. For metal density, This represents the corrosion current density.

[0074] Let the material corrosion constant Then we obtain the pure corrosion evolution equation:

[0075] (2)

[0076] Next, based on the Paris formula, a pure fatigue evolution equation was established.

[0077] Specifically, crack propagation rate With stress intensity factor amplitude Therefore, the pure fatigue evolution equation is as follows:

[0078] (3)

[0079] in, , This is the geometric correction factor. Nominal stress amplitude, This represents the half-length of the crack, or the crack length. Let C and m be the load frequency, and C and m be material constants. For simplicity, the load frequency term can be incorporated into the expression. middle.

[0080] In one embodiment, a coupled damage physics equation for corrosion and fatigue is constructed, which includes the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue, including:

[0081] Based on the coupling effect of fatigue on corrosion, a physical equation for corrosion damage is constructed, which includes the effect of stress on corrosion rate.

[0082] Based on the coupling effect of corrosion on fatigue, a fatigue damage physics equation is constructed, which includes the influence of corrosion depth on crack propagation rate and the influence of hydrogen atoms on crack propagation rate.

[0083] Specifically, based on the coupling effect of fatigue on corrosion, a physical equation for corrosion damage is first constructed.

[0084] Alternating loads generate localized high strain rates at the crack tip, damaging the passivation film and increasing activity. According to the Gutman model, dynamic corrosion current... for:

[0085] (4)

[0086] in, For micro-strain at the crack tip, For reference strain.

[0087] Due to microstrain at the crack tip Difficult to monitor directly, we utilize linear elasticity to map microscopic strain into macroscopic stress. Functions:

[0088] (5)

[0089] in, The mechanochemical coupling factor characterizes the effect of stress on the corrosion rate. These are the material-related fitting coefficients.

[0090] Furthermore, the saturation effect caused by the introduction of corrosion product coverage. , The saturation index Given the current corrosion depth, This represents the maximum corrosion depth.

[0091] Substituting the above formula into formula 2, we get:

[0092] (6)

[0093] make The final corrosion evolution equation is obtained:

[0094] (7)

[0095] This equation shows that instantaneous stress The larger the value, the higher the corrosion rate. The faster.

[0096] Next, based on the coupling effect of corrosion on fatigue, a physical equation for fatigue damage is constructed.

[0097] Corrosion accelerates crack propagation through a dual mechanism of "geometric effect" and "environmental effect".

[0098] "Geometric effect": Stress concentration caused by corrosion pits, which can be considered as surface notches. According to fracture mechanics, its stress concentration factor... for:

[0099] (8)

[0100] in, For corrosion depth, Let be the radius of curvature of the pit bottom. Use this to correct the effective stress intensity factor. :

[0101] (9)

[0102] Assuming the morphology of corrosion pits has self-similarity, The term is equivalent to relative depth ( The function is introduced, and a geometric coupling factor is introduced. :

[0103] (10)

[0104] in, The geometric coupling factor characterizes the effect of corrosion depth on crack propagation rate.

[0105] "Environmental Effect": Hydrogen embrittlement occurs when hydrogen atoms produced during corrosion diffuse to the crack tip. According to the hydrogen-induced cracking theory, hydrogen concentration reduces the material's fracture energy, increasing the crack propagation rate. The environmental effect can be simplified as a linear strengthening factor:

[0106] (11)

[0107] in, As a hydrogen embrittlement sensitizer, it combines hydrogen concentration and material sensitivity to characterize the effect of hydrogen atoms on crack propagation rate.

[0108] Substituting the two corrections mentioned above (Equation 10 and Equation 11) into the pure fatigue evolution equation (Equation 3), we get:

[0109]

[0110] Right now

[0111] (12)

[0112] After simplification, the final crack propagation evolution equation is obtained:

[0113] (13)

[0114] This equation shows that the corrosion depth The deeper or the harsher the environment ( (The larger the crack growth rate) The faster.

[0115] Based on the above derivation, the remaining life prediction model for a floating platform mooring system consists of the following set of coupled equations:

[0116] (14)

[0117] in, As a mechanochemical coupling factor, it characterizes how stress σ accelerates corrosion by increasing strain rate; The geometric coupling factor characterizes the corrosion depth. How to address stress concentration To accelerate crack propagation; As a hydrogen embrittlement sensitizer, it characterizes how hydrogen generated by corrosion accelerates crack propagation by reducing material toughness.

[0118] To improve the numerical stability and convergence speed of neural network training, the physical equation formula 14 derived in the previous steps is made dimensionless.

[0119] First, define the dimensionless state variable:

[0120] (15)

[0121] in, This is the critical crack length for fracture. For the design life of the mooring system, This represents the yield stress limit of the material.

[0122] Substituting the variables in Equation 15 into Equation 14, we obtain the standard form applicable to Physics-Guided Machine Learning (PGML) models, namely the physical equations for corrosion damage and fatigue damage:

[0123] (16)

[0124] in, The corrosion rate is dimensionless. The coefficients for the corrosion model are the overall coefficients. Let the dimensionless state variable be the corrosion depth. The saturation index The mechanochemical coupling factor characterizes the effect of stress on the corrosion rate. For normalized load, These are the material-related fitting coefficients.

[0125] The crack propagation rate is a dimensionless value. These are the comprehensive coefficients of the fatigue model. For material constants, Let be the dimensionless state variable representing the crack length. The geometric coupling factor characterizes the effect of corrosion depth on crack propagation rate. Geometric dimension ratio, It is a hydrogen embrittlement sensitizer, characterizing the effect of hydrogen atoms on crack propagation rate.

[0126] The physical meanings and derivation relationships of the dimensionless parameters in Formula 16 are as follows:

[0127] (Corrosion Model Composite Coefficient): Integrates timescale and material electrochemical properties, representing the maximum relative corrosion rate achievable within the design life. .

[0128] (Fatigue model comprehensive coefficient): integrates time scale, material fatigue properties (Paris parameters), and geometric properties, among which, .

[0129] (Geometric Dimension Ratio): A structural characteristic parameter of the mooring chain, reflecting the proportional relationship between the maximum corrosion depth and the critical crack length, used to correct the geometric coupling term. .

[0130] (Normalized load): When corresponding to the physical equation of corrosion damage, it represents the normalized instantaneous stress; when corresponding to the physical equation of fatigue damage, it represents the normalized stress amplitude. .

[0131] This disclosure establishes an accurate physical equation for corrosion-fatigue coupled damage, describing the dynamic process of their interaction from a mechanistic perspective.

[0132] After obtaining Equation 16, it is necessary to verify whether the model conforms to common sense physics under boundary conditions.

[0133] Pure corrosion verification: Let the input load be Check equation 16 Does it approach 0, and Whether it degenerates into a corrosion rate that decays only over time.

[0134] Pure fatigue verification: increasing corrosion depth Examine the geometric coupling terms in equation 16. Whether it approaches 1, degenerates into the standard Paris law.

[0135] Step 102: Based on the coupled damage physical equations of corrosion and fatigue, determine the physical equation of the comprehensive damage degree.

[0136] In one embodiment, a physical equation for determining the overall damage degree is derived based on the coupled damage physics equations of corrosion and fatigue, including:

[0137] Based on the coupled damage physics equations of corrosion and fatigue, the corrosion depth and crack length at the current moment are determined.

[0138] Based on the current corrosion depth and crack length, the physical equation for the overall damage degree is determined; the physical equation for the overall damage degree is shown in the following formula:

[0139]

[0140] in, The overall damage level at the current moment. The current corrosion depth. The crack length at the current moment. The weight of the current corrosion depth. The weight is the crack length at the current moment.

[0141] Specifically, in order to unify physical damage of two different dimensions, corrosion and fatigue, into a scalar index that can be used for life prediction, this disclosure establishes a comprehensive damage degree. :

[0142] (17)

[0143] in, Given the current corrosion depth, To allow for corrosion limits, This represents the current crack length. This is the critical crack length for fracture. The weighting coefficient reflects the risk proportion of the two failure modes in a specific sea area.

[0144] when At that time, the mooring system was determined to be faulty. The time t corresponds to the end of the service life, which is also the total service life of the mooring system. Subtracting the current service life from the total service life gives the remaining service life.

[0145] Because Equation 16 is highly nonlinear and strongly coupled, it cannot be directly superimposed as in traditional methods. This disclosure uses a numerical integration method to solve for the damage evolution trajectory.

[0146] State recursion: Using the fourth-order Runge-Kutta method (RK4), based on the dimensionless state at the current time step. , and load To recursively calculate the state at the next time step:

[0147] (18)

[0148] Overall damage output: The calculated damage level (Relative corrosion depth) and Substituting (relative crack length) into Formula 17, we obtain the physical equation for the overall damage degree:

[0149] (19)

[0150] Based on this equation, the remaining life curve of the mooring system as a function of time can be obtained.

[0151] To ensure that Equation 16 accurately describes the characteristics of specific materials and sea areas, a "step-by-step calibration" strategy is adopted to determine the model parameters. .

[0152] First, determine the basic physical parameters through laboratory testing and calibration. .

[0153] Corrosion model comprehensive coefficient The results were determined by fitting the curve through a full immersion static corrosion test.

[0154] Fatigue model comprehensive coefficient The test was conducted using a crack propagation test (CT specimen) under standard air conditions, and calibrated based on the Paris formula.

[0155] Saturation index Through long-term immersion tests, and plotting "corrosion depth-time" (Current Distance as a Percentage of Corrosion) data, a corrosion depth-time curve was constructed. The curve is used to determine this.

[0156] Geometric coupling factor The difference in fatigue life between pre-corroded pitted specimens and specimens without pits is determined by fatigue testing of the pre-corroded pitted specimens.

[0157] Then, environmentally sensitive parameters are determined through inversion of on-site data. .

[0158] For deep-sea high-pressure, low-temperature, and microbial environments that are difficult to simulate in a laboratory, environmentally sensitive parameters (mechanical-chemical coupling factor) are considered. Hydrogen embrittlement sensitive factor Inversion identification is used to construct an optimization objective function. This makes the overall damage degree of the damage trajectory predicted by the model... Approximating the overall damage degree of sparse damage points obtained from actual on-site testing :

[0159] (20)

[0160] After multiple iterations, when and When the difference between them is the smallest, Corresponding mechanochemical coupling factor and hydrogen embrittlement sensitive factors That is the required value.

[0161] This disclosure also allows for parameter sensitivity analysis.

[0162] calculate The physical parameters in Predicted final remaining life Sensitivity coefficient To identify key control factors:

[0163] (twenty one)

[0164] Substituting each parameter into Formula 21 yields the sensitivity coefficient for each parameter. For example, if the calculated value is... The one with the highest sensitivity coefficient should be the focus of subsequent monitoring, particularly for hydrogen concentration detection, and efforts should be intensified. The weight.

[0165] Step 103: Embed the physical equations of coupled damage and the physical equations of comprehensive damage into the initial prediction model.

[0166] The core of this disclosure lies in using Equations 16 and 19 as "physical knowledge kernels" and embedding them into the training process of the initial prediction model to construct a physics-guided machine learning model. The initial prediction model in this disclosure can be a neural network model. It should be noted that Equations 15 and 18 can also be embedded into the initial prediction model in this disclosure.

[0167] This disclosure aims to construct a hybrid neural network model that deeply integrates physical mechanisms and data-driven approaches. Unlike traditional "black box" models that rely solely on data regression, the network architecture of this disclosure includes an explicit "physical constraint module" to ensure that the prediction results strictly follow the corrosion-fatigue coupled damage physical equation in Equation 16.

[0168] Step 104: Train the initial prediction model to obtain the target prediction model.

[0169] In one embodiment, training an initial prediction model to obtain a target prediction model includes:

[0170] Obtain a sample training set, which includes time-series monitoring data of the sample mooring system;

[0171] Input the sample training set into the initial prediction model;

[0172] Construct a hybrid loss function for the initial prediction model, which includes a data fitting loss function, a physical consistency loss function, and an initial condition loss function;

[0173] The initial prediction model is trained based on the hybrid loss function to obtain the target prediction model.

[0174] Specifically, the training set includes actual measured sample data, simulated sample data obtained through simulation software, and experimental sample data obtained through laboratory experiments.

[0175] To address the scarcity of measured fault data in deep-sea engineering, a strategy of using simulation data as the primary source and measured data as a supplementary source is adopted to construct a sample training set.

[0176] Firstly, simulation sample data can be obtained through hydrodynamic model simulation.

[0177] A complete hydrodynamic model of the system, including the platform, mooring chain, and anchor points, was established. The model is as follows: Figure 2 As shown, the long-term joint distribution of wind, waves, and currents in the target sea area (e.g., scattering diagram) is input, and extensive time-domain coupled analysis and simulation are performed. The mooring failure model is as follows: Figure 3 As shown, after mooring failure, the stress-time history of each mooring chain during its service life (or a representative period) is output as the load for subsequent fatigue analysis.

[0178] This disclosure utilizes a hydrodynamic model to simulate a large amount of sample data, enabling high-fidelity simulation of environmental loads and system dynamic responses. By establishing a time-domain analysis model coupled with wind, waves, and current using marine engineering software, it provides realistic and reliable load inputs for damage analysis.

[0179] The dimensionless coupled damage physics equation (Formula 16) is solved by numerical integration to generate the corresponding theoretical damage evolution label, which serves as the source of "physical prior knowledge" for the initial prediction model.

[0180] Then, data from laboratory corrosion-fatigue coupling tests and discrete measurement points obtained from random on-site inspections of the mooring chains (such as the depth of corrosion pits in the 5th year) were collected. Crack length in the 10th year The data is then dimensionless to obtain the measured labels. The data corresponding to the theoretical damage evolution labels and the measured labels can serve as the basis for subsequent data fitting of the loss function.

[0181] The network architecture of the initial prediction model consists of the following three parts:

[0182] 1. Basic Prediction Network

[0183] The basic prediction network uses a Long Short-Term Memory (LSTM) network as the backbone network to fit the mapping relationship between state variables and unknown parameters.

[0184] The input vector of the input layer of the basic prediction network .

[0185] in, This refers to dimensionless service time. For normalized load, For relevant environmental monitoring indicators (such as temperature, seawater pH value, etc.).

[0186] Output vector of the output layer of the basic prediction network .

[0187] in, For the predicted dimensionless corrosion depth, The predicted dimensionless crack length, The network outputs learnable environmental parameters (mechanical-chemical coupling factor and hydrogen embrittlement sensitivity factor, respectively), which are allowed to change dynamically with time and environmental conditions.

[0188] In this disclosure, time-varying environmental parameters that are difficult to measure accurately (such as those that vary with the seasons) are addressed. and Instead of setting it as a fixed constant, it is used as the output channel of the neural network. During training, the network automatically adjusts... and The values ​​of these two parameters are set to minimize the mixed loss function. This enables the use of data to inversely correct the physical model parameters, making the model adaptive.

[0189] It should be explained that the output of the basic prediction network here is not the final output of the initial prediction model, but an intermediate output within the initial prediction model, while the final output of the initial prediction model is the overall damage degree.

[0190] 2. Physical Mechanism Embedded Module

[0191] This module does not contain trainable parameters; its function is to construct physical residuals using Automatic Differentiation (AD) to provide the foundational data for the subsequent physical consistency loss function. This module constructs multiple point sets by randomly sampling within the input layer of the base prediction network. These points do not have corresponding damage labels and are only used to calculate the physical equation residuals (i.e., to check whether the model conforms to physical laws).

[0192] This module consists of two parts: automatic differential calculation and physical equation calculation.

[0193] Automatic Differentiation Calculation: Utilizing the automatic differentiation function of deep learning frameworks to perform automatic differentiation on the output of the basic prediction network. and Regarding input time Differentiate to obtain the numerical derivative of the neural network. and .in, and This refers to the corrosion rate and crack propagation rate predicted by the initial prediction model.

[0194] Physical equation calculation: Calculating the state values ​​predicted by the basic prediction network. and parameter values Substitute directly into the right-hand side of Formula 16 to calculate the theoretical derivative:

[0195] Theoretical corrosion rate:

[0196] (twenty two)

[0197] Theoretical crack propagation rate:

[0198] (twenty three)

[0199] in, and These are the second target values ​​corresponding to the corrosion rate and crack propagation rate, respectively.

[0200] 3. Construction of hybrid loss function

[0201] Hybrid loss functions include data fitting loss function, physical consistency loss function, and initial condition loss function.

[0202] In one embodiment, the data fitting loss function includes the error between the corrosion depth and crack length predicted by the initial prediction model and their respective first target values;

[0203] The physical consistency loss function includes the error between the corrosion rate and crack propagation rate predicted by the initial prediction model and their respective second target values.

[0204] To impose physical mechanisms onto neural networks, a hybrid loss function consisting of three parts was constructed. The goal of model training is to minimize the mixture loss function. The network weights and physical parameters are optimized simultaneously through backpropagation.

[0205] Hybrid loss function Determined by the following formula:

[0206] (twenty four)

[0207] in, To fit the loss function to the data, Let physical consistency loss function be used. The initial conditional loss function, and These are the weights of the physical consistency loss function and the initial condition loss function, respectively.

[0208] Data fitting loss function This is used to measure the error between the predicted values ​​(corrosion depth and crack length) obtained through the initial prediction model and the measured data or high-fidelity simulation data (first target value), where the data fitting loss function is used. Determined by the following formula:

[0209] (25)

[0210] in, The number of sample points for measured data or high-fidelity simulation data. , () refers to measured data or high-fidelity simulation data.

[0211] Physical consistency loss function This measure assesses whether the evolutionary trajectory predicted by the initial prediction model violates the physical laws defined in Equation 16, preventing the neural network from "overfitting" or producing predictions that violate physical laws in areas with scarce data. This item does not require labeled data and can be calculated at any time point.

[0212] (26)

[0213] in, The number of points randomly sampled within the input layer of the base prediction network. and The mathematical form defined by Equation 16 ensures that the neural network is forcibly constrained to the physical manifold of corrosion-fatigue coupling.

[0214] With this architecture, the model can be calibrated using data wherever data is available. ), using physical equations where no data is available ( This allows for the simulation of the remaining lifetime, thus enabling high-precision prediction of the remaining lifetime.

[0215] Initial condition loss function Used to force neural networks to meet specific physical starting conditions.

[0216] Initial state definition: For newly deployed mooring chains, At that time, the actual initial corrosion depth Actual initial crack length (in This refers to the inherent initial crack length or detection limit of the material. It is the critical crack length.

[0217] Initial condition loss function As shown in the following formula:

[0218] (27)

[0219] in, This is the predicted corrosion depth at the initial moment. This is the predicted crack length at the initial moment.

[0220] To avoid the neural network getting trapped in local optima, a two-stage strategy of physical pre-training and data fine-tuning is adopted:

[0221] In the Physics-Only Pretraining phase, only... and The goal is to enable the neural network to first learn to "solve differential equations," grasp the basic physical characteristics of the coupled evolution of corrosion and fatigue, and ensure that the model can output results that conform to physical common sense even in areas without data coverage.

[0222] In the data-driven fine-tuning phase, measured data is introduced. And retain at the same time As a regularization term, its purpose is to correct idealization biases in pure physical models using real data and to calibrate environmental parameter outputs. This allows the predicted curve to accurately approximate the actual damage path.

[0223] After determining the hybrid loss function, the Bayesian optimization algorithm is used to automatically optimize the number of network layers, the number of neurons, and the physical loss weights, with the goal of minimizing the prediction error on the validation set, in order to obtain the trained target prediction model.

[0224] Step 105: Obtain the time-series monitoring data of the mooring system under test.

[0225] After obtaining the target prediction model, the remaining life of the mooring system under test can be predicted based on the target prediction model.

[0226] To predict remaining useful life, it is first necessary to obtain time-series monitoring data of the mooring system under test, such as the current service time, normalized load, and relevant environmental monitoring indicators.

[0227] Step 106: Input the time-series monitoring data of the mooring system under test into the target prediction model to determine the comprehensive damage degree of the mooring system under test.

[0228] After the time-series monitoring data of the mooring system under test is input into the target prediction model, the target prediction model will output a curve showing the overall damage degree over time. From this curve, the overall damage degree of the mooring system under test at different service times can be seen.

[0229] Step 107: Determine the remaining life of the mooring system under test based on the overall damage level.

[0230] In one embodiment, determining the remaining lifetime of the mooring system under test based on the overall damage level includes:

[0231] Determine whether the overall damage level is greater than or equal to 1;

[0232] When the overall damage level is greater than or equal to 1, it indicates that the mooring system under test has failed. The time corresponding to the current overall damage level is determined as the total service life of the mooring system under test.

[0233] Determine the remaining lifespan of the mooring system under test based on its total service life.

[0234] In step 106, it is mentioned that the target prediction model will output a curve of the overall damage degree changing over time. When the overall damage degree in the curve is greater than or equal to 1, it indicates that the mooring system under test has failed. The time corresponding to when the overall damage degree is equal to 1 is taken as the total service life of the mooring system under test. Since the current service life of the mooring system under test is known, the remaining life can be obtained by subtracting the current service life from the total service life.

[0235] In this disclosure, damage can be predicted from year T to year T+N by training the model using data from the mooring system over the previous T years, thus verifying the model's ability to predict future trends. The model's stability is tested on simulation data of a "once-in-a-century" severe sea state to ensure that the model will not exhibit physically impossible "negative damage" or "infinite damage divergence" phenomena under extreme conditions.

[0236] The model disclosed herein possesses both the ability to extrapolate the mechanism of a physical model and the ability of machine learning to learn from new data. It is applicable to mooring systems in different sea areas, platform types, and service stages, and can self-improve as new data accumulates. It can provide direct and quantitative scientific basis for the integrity management, condition-based maintenance decision-making, and life extension assessment of floating platform mooring systems.

[0237] The solutions disclosed herein can be widely applied in the following fields: 1. Deep-sea oil and gas field exploration and development. This disclosure can provide accurate remaining life predictions for the mooring systems of in-service floating production, storage and offloading (FPSO) units, semi-submersible production platforms, and deep-draft stand-alone (SPAR) platforms, supporting the scientific decision-making shift from "periodic replacement" to "condition-based maintenance" and "safe life extension." It significantly reduces the risk of production stoppages and huge replacement costs caused by overly conservative design or unexpected failures, improving the economic benefits of the oilfield throughout its entire life cycle. 2. Cutting-edge deep-sea technology. This disclosure can be applied to mooring and anchoring systems for new marine structures such as deep-sea aquaculture platforms, deep-sea mining systems, offshore wind power floating foundations, and deep-sea tourism facilities. It provides advanced tools for the long-term reliability design and verification in extreme deep-sea environments. It reduces the trial-and-error costs and engineering risks of new equipment development, accelerating the technological maturity and industrialization of deep-sea resource development and space utilization equipment. 3. Data analysis and artificial intelligence. This disclosure can serve as a reference paradigm for solving similar problems of "unclear mechanisms, insufficient data, and inaccurate predictions" in process industries, major equipment, and infrastructure. It promotes the widespread application of PGML technology in industrial intelligent diagnosis and prediction, and facilitates the deep integration of traditional industries with artificial intelligence. 4. In the field of materials and structural testing in deep-sea environments, this disclosure can serve as a benchmark method and virtual testbed for evaluating the performance of deep-sea engineering materials (such as ultra-high strength steel and composite materials) and new anti-corrosion technologies under complex load environments, providing advanced data support and theoretical tools for the research and development and standardization of deep-sea material systems. 5. In the field of marine engineering and shipbuilding, this disclosure can be integrated and developed into a professional marine engineering life prediction and risk assessment software module, supplementing or replacing some functions of similar commercial software, and enhancing the independent controllability of my country's marine engineering industry chain.

[0238] This disclosure also provides a mooring system life prediction device. Figure 4This is a schematic diagram of the structure of the mooring system life prediction device provided in the embodiments of this disclosure, as shown below. Figure 4 As shown, the device includes:

[0239] Module 401 is used to construct the coupled damage physics equations of corrosion and fatigue, which include the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue.

[0240] The first determining module 402 is used to determine the physical equation of the comprehensive damage degree based on the coupled damage physical equation of corrosion and fatigue.

[0241] Embedding module 403 is used to embed the physical equations of coupled damage physical equations and comprehensive damage degree into the initial prediction model;

[0242] Training module 404 is used to train the initial prediction model to obtain the target prediction model;

[0243] The acquisition module 405 is used to acquire the time-series monitoring data of the mooring system under test.

[0244] The second determining module 406 is used to input the time-series monitoring data of the mooring system under test into the target prediction model in order to determine the comprehensive damage degree of the mooring system under test.

[0245] The third determination module 407 is used to determine the remaining life of the mooring system under test based on the comprehensive damage degree.

[0246] The specific details of each part of the above-mentioned device have been described in detail in the method section of the implementation plan. For any undisclosed details, please refer to the implementation plan of the method section, and therefore will not be repeated here.

[0247] According to embodiments of this disclosure, this disclosure also provides an electronic device and a readable storage medium.

[0248] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0249] like Figure 5As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the electronic device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0250] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, such as keyboard, mouse, etc.; output unit 507, such as various types of monitors, speakers, etc.; storage unit 508, such as disk, optical disk, etc.; and communication unit 509, such as network card, modem, wireless transceiver, etc. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0251] The computing unit 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as the mooring system lifetime prediction method. For example, in some embodiments, the mooring system lifetime prediction method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the mooring system lifetime prediction method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the mooring system lifetime prediction method by any other suitable means (e.g., by means of firmware).

[0252] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0253] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0254] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0255] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0256] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0257] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0258] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this disclosure can be achieved, and this is not limited herein.

[0259] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means two or more, unless otherwise explicitly specified.

[0260] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

Claims

1. A method for predicting the lifespan of a mooring system, characterized in that, The method includes: A coupled damage physics equation for corrosion and fatigue is constructed, which includes the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue. A physical equation for determining the comprehensive damage degree is derived from the coupled damage physical equations of corrosion and fatigue. The coupled damage physical equation and the comprehensive damage degree physical equation are embedded into the initial prediction model; The initial prediction model is trained to obtain the target prediction model; Acquire time-series monitoring data of the mooring system under test; The time-series monitoring data of the mooring system under test is input into the target prediction model to determine the comprehensive damage degree of the mooring system under test. Based on the comprehensive damage level, the remaining life of the mooring system under test is determined; The physical equation for determining the comprehensive damage degree based on the coupled damage physical equation of corrosion and fatigue includes: Based on the coupled damage physics equations of corrosion and fatigue, the corrosion depth and crack length at the current moment are determined. Based on the current corrosion depth and crack length, determine the physical equation for the overall damage degree; The constructed coupled damage physics equations of corrosion and fatigue, which include the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue, include: Based on the coupling effect of fatigue on corrosion, a physical equation for corrosion damage is constructed, which includes the effect of stress on corrosion rate. Based on the coupling effect of corrosion on fatigue, a fatigue damage physical equation is constructed, which includes the influence of corrosion depth on crack propagation rate and the influence of hydrogen atoms on crack propagation rate. The construction of the physical equation for corrosion damage includes: determining the physical equation for corrosion damage based on the following formula: in, The corrosion rate is dimensionless. The coefficients for the corrosion model are the overall coefficients. Let the dimensionless state variable be the corrosion depth. The saturation index The mechanochemical coupling factor characterizes the effect of stress on the corrosion rate. For normalized load, These are the material-related fitting coefficients; The construction of the fatigue damage physical equation includes: determining the fatigue damage physical equation based on the following formula: in, The crack propagation rate is a dimensionless value. These are the comprehensive coefficients of the fatigue model. For material constants, Let be the dimensionless state variable representing the crack length. The geometric coupling factor characterizes the effect of corrosion depth on crack propagation rate. Geometric dimension ratio, It is a hydrogen embrittlement sensitizer, characterizing the effect of hydrogen atoms on crack propagation rate.

2. The method according to claim 1, characterized in that, The physical equation for the overall damage degree is shown in the following formula: in, The overall damage level at the current moment. The current corrosion depth. The crack length at the current moment. The weight of the current corrosion depth. The weight is the crack length at the current moment.

3. The method according to claim 1, characterized in that, The step of training the initial prediction model to obtain the target prediction model includes: Obtain a sample training set, which includes time-series monitoring data of the sample mooring system; The sample training set is input into the initial prediction model; Construct a hybrid loss function for the initial prediction model, the hybrid loss function including a data fitting loss function, a physical consistency loss function, and an initial condition loss function; The initial prediction model is trained based on the hybrid loss function to obtain the target prediction model.

4. The method according to claim 3, characterized in that, The data fitting loss function includes the error between the corrosion depth and crack length predicted by the initial prediction model and their respective first target values; The physical consistency loss function includes the error between the corrosion rate and crack propagation rate predicted by the initial prediction model and their respective second target values.

5. The method according to claim 1, characterized in that, Determining the remaining lifetime of the mooring system under test based on the comprehensive damage level includes: Determine whether the overall damage degree is greater than or equal to 1; When the overall damage level is greater than or equal to 1, it indicates that the mooring system under test has failed, and the time corresponding to the current overall damage level is determined as the total service life of the mooring system under test. The remaining lifespan of the mooring system under test is determined based on the total service life.

6. A mooring system life prediction device, characterized in that, The device includes: A construction module is used to construct a coupled damage physics equation for corrosion and fatigue, wherein the coupled damage physics equation includes the coupling effect of fatigue on corrosion and the coupling effect of corrosion on fatigue. The first determining module is used to determine the physical equation of the overall damage degree based on the coupled damage physical equation of corrosion and fatigue. An embedding module is used to embed the coupled damage physical equation and the comprehensive damage degree physical equation into the initial prediction model; The training module is used to train the initial prediction model to obtain the target prediction model; The acquisition module is used to acquire time-series monitoring data of the mooring system under test; The second determining module is used to input the time-series monitoring data of the mooring system under test into the target prediction model in order to determine the comprehensive damage degree of the mooring system under test. The third determining module is used to determine the remaining life of the mooring system under test based on the comprehensive damage level. The first determining module is specifically used to determine the corrosion depth and crack length at the current moment based on the coupled damage physics equation of corrosion and fatigue; and to determine the physical equation of comprehensive damage degree based on the corrosion depth and crack length at the current moment. The construction module is used to construct a corrosion damage physical equation based on the coupling effect of fatigue on corrosion, the corrosion damage physical equation including the effect of stress on corrosion rate; and to construct a fatigue damage physical equation based on the coupling effect of corrosion on fatigue, the fatigue damage physical equation including the effect of corrosion depth on crack propagation rate and the effect of hydrogen atoms on crack propagation rate. The building module is used to determine the physical equation of corrosion damage based on the following formula: in, The corrosion rate is dimensionless. The coefficients for the corrosion model are the overall coefficients. Let the dimensionless state variable be the corrosion depth. The saturation index The mechanochemical coupling factor characterizes the effect of stress on the corrosion rate. For normalized load, These are the material-related fitting coefficients; The building module is used to determine the physical equation of fatigue damage based on the following formula: in, The crack propagation rate is a dimensionless value. These are the comprehensive coefficients of the fatigue model. For material constants, Let be the dimensionless state variable representing the crack length. The geometric coupling factor characterizes the effect of corrosion depth on crack propagation rate. Geometric dimension ratio, It is a hydrogen embrittlement sensitizer, characterizing the effect of hydrogen atoms on crack propagation rate.

7. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5.

8. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-5.

Images