Parameter calculation method for ship hull fatigue damage evolution model under multi-field coupling

Abstract

This invention provides a method for calculating parameters of a ship hull fatigue damage evolution model under multi-field coupling, relating to the field of marine structural fatigue damage prediction technology. This invention collects historical data of the environmental base field, discrete phase medium field, and ship motion field of the target sea area to construct a dynamic boundary sequence, and establishes a ship-ice debris-wave coupling model based on the CEL method to obtain stress time histories of key components such as the bow, sides, and bottom in real time. Then, it uses the rainflow counting method to process the stress time histories to generate multi-level load spectra, and combines fatigue tests at different temperatures to obtain corrected S-N curves. Finally, the load spectra are sequentially substituted into the corrected curves to calculate the remaining fatigue life of each key component under conditions considering historical cumulative damage and real-time temperature, achieving accurate prediction of ship hull structural fatigue damage in polar navigation environments.

Description

Technical Field

[0001] This invention relates to the field of marine structural fatigue damage prediction technology, specifically a method for calculating parameters of a ship hull fatigue damage evolution model under multi-field coupling. Background Technology

[0002] In polar and frigid navigation environments, ship structures are subjected to the combined effects of multiple coupled fields, including waves, ice fragments, and low temperatures, resulting in a highly complex and uncertain fatigue damage evolution process. Traditional ship fatigue assessment methods are typically based on single load conditions or simplified environmental assumptions, making it difficult to accurately reflect the impact of the synergistic effects of multiple physical fields on the fatigue life of ship structures under actual sea conditions. Especially under ice-covered navigation conditions, the coupling effect of ice fragment impact, wave loads, and ship motion significantly exacerbates stress concentration and cumulative damage in critical areas. Existing technologies lack systematic modeling and parameter calculation methods for this complex process, leading to insufficient accuracy in fatigue life prediction and failing to meet the practical needs of polar ship engineering design. Therefore, there is an urgent need for a parameter calculation method for a ship fatigue damage evolution model that can comprehensively consider multi-field coupling effects, dynamic environmental changes, and material fatigue performance degradation mechanisms to achieve accurate prediction of remaining fatigue life.

[0003] In the prior art, publication number CN118395815A discloses a method for calculating parameters of a fatigue damage evolution model for marine structures, which obtains the engineering stress of metal plates. Engineering strain curve, based on engineering stress The strain curve is used to calculate the yield strength, tensile strength, and elastic modulus of the metal sheet; the stress of the metal sheet under different strain amplitudes is obtained. Strain hysteresis curves, based on stress under different strain amplitudes. The parameters of the nonlinear isotropic hardening model for metal sheets are calculated using strain hysteresis curves; the back strain force of the metal sheets under various strain amplitudes is obtained. Plastic strain curve, based on the back strain force under loading at each strain amplitude. The parameters of the nonlinear kinematic hardening model for metal sheets are obtained from the plastic strain curve.

[0004] The main problems with the above approach are: it only relies on material performance testing of sheet metal under normal temperature and single mechanical loading conditions, without considering the impact of the coupling effects of multiple physical fields such as waves, sea ice, and low temperature in the marine environment on the material's fatigue performance. In actual marine engineering, structures are subjected to complex loads and environmental conditions for extended periods. Ignoring multi-field coupling effects will lead to a significant deviation between the predicted results of the fatigue damage evolution model and the actual usage scenario. Furthermore, it relies solely on material-level cyclic loading tests without considering the dynamic response of the structure under actual sea conditions, lacking refined modeling of the local stress response of the structure, making it difficult to accurately assess the fatigue damage evolution of key components. Although the hardening model parameters of the material are extracted, they are not effectively coupled with the structural fatigue life prediction model, lacking a complete mapping from material properties to structural response.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method for calculating the parameters of a ship hull fatigue damage evolution model under multi-field coupling, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] The calculation of parameters for a ship hull fatigue damage evolution model under multi-field coupling includes the following steps:

[0009] Step 1: Collect historical environmental baseline field data, discrete phase medium field data, and ship motion field data for the target sea area. Construct a dynamic boundary sequence based on the above data. The environmental baseline field data includes seawater density and seawater dynamic viscosity. The discrete phase medium field data includes ice coverage, ice thickness, ice elastic modulus, compressive strength, and fracture energy. The ship motion field data includes ship speed, heading angle, roll angle, and pitch angle.

[0010] Step 2: Construct a hull-ice debris-wave coupling model based on CEL, apply a dynamic boundary sequence to the coupling model, obtain the stress of key parts of the hull in real time, and extract the stress time history of each key part. The key parts of the hull include the bow, side and bottom.

[0011] Step 3: The stress time history of each key component is processed using the rainflow counting method to obtain the stress amplitude and stress mean under each stress cycle, thereby generating a multi-level load spectrum for each key component.

[0012] Step 4: Prepare specimens based on the steel used for the hull, conduct fatigue tests on the specimens at different temperature points, obtain the basic SN function at different temperature points, and correct the basic SN function based on historical cumulative damage;

[0013] Step 5: Set the initial cumulative damage to zero and the initial ambient temperature. Substitute the load spectra of different key parts into the modified SN function in sequence to calculate the remaining fatigue life of each key part when the load spectrum is applied each time.

[0014] Furthermore, the principle for constructing a dynamic boundary condition sequence is as follows:

[0015] Environmental baseline field data, discrete phase medium field data, and ship motion field data of the target sea area were collected daily over the past year. For each type of data, the continuously changing data was divided into several discrete states, and the probability of transitioning from one state to another was determined. A state transition probability matrix for each type of data was constructed, and a dynamic boundary sequence was constructed using the Markov chain Monte Carlo method.

[0016] Furthermore, the principle underlying the division of continuously changing data into several discrete states is as follows:

[0017] Based on the range of seawater density and seawater dynamic viscosity in historical data, the seawater is divided into three equal segments between the maximum and minimum values, and each segment is set as a state.

[0018] Ice coverage ranging from 0% to 100% is divided into five states: no ice (0-10% ice coverage); light ice (10-30% ice coverage); moderate ice (30-60% ice coverage); heavy ice (60-90% ice coverage); and full ice coverage (90-100% ice coverage).

[0019] Based on historical data on ice thickness, elastic modulus, compressive strength, and fracture energy, a corresponding range is obtained for each type of data. The range is then divided into five equal segments between the minimum and maximum values, with each segment representing a state.

[0020] Within the ship's speed range, it is divided into three equal segments, corresponding to three states: low speed, medium speed, and high speed.

[0021] The heading angle is divided into equal intervals within the range of 0 to 360°, with each angle range corresponding to a different state;

[0022] Based on the range of roll and pitch angles of the ship in historical data, the range between the maximum and minimum values ​​is divided into three equal segments, each segment corresponding to a state.

[0023] Furthermore, the principle of constructing the hull-ice debris-wave coupling model is as follows:

[0024] The three-dimensional geometric model of the hull is obtained and imported into the finite element software. The three-dimensional geometric model of the hull is discretized using a Lagrange mesh. The density, elastic modulus, Poisson's ratio and yield strength of the hull material are obtained and imported into the discretized three-dimensional geometric model to construct the Lagrange finite element model of the hull structure.

[0025] An Eulerian mesh containing the water area around the ship hull and the ice zone is constructed to build the Eulerian domain. The Eulerian mesh includes two material components: water material and ice material. The water material is defined using the Newtonian fluid constitutive model, and the ice material is defined using the Drucker-Prager yield criterion combined with the damage evolution criterion.

[0026] Several contact pairs are established between the surface of the Lagrange finite element model of the hull structure and the Eulerian domain. The contact pairs are the areas where the hull structure and the Eulerian domain come into contact with each other. The contact force of each contact pair is calculated based on the penalty function method. The contact force of each contact pair is applied to the Lagrange mesh node corresponding to the hull structure, causing the deformation and movement of the hull structure. The contact force is used to realize the coupling between the hull structure and ice fragments and waves, thus constructing a hull-ice fragment-wave coupled model.

[0027] Furthermore, the principle for extracting the stress time history of each key component is as follows:

[0028] A dynamic boundary sequence is applied to the coupled model, and the total time of the dynamic boundary sequence is discretized into several time points. For each key part, the stress at each time point is recorded, and the stress time history of that key part is obtained.

[0029] Furthermore, the principle for generating multi-level load spectra for each key component is as follows:

[0030] For each critical component, the stress time history is processed using the rainflow counting method to generate several stress cycles. Specifically, for each stress time history, the stress corresponding to three consecutive time points is read and recorded as follows: ,in This represents the stress corresponding to the first time point. This indicates the stress corresponding to the second time point. This represents the stress corresponding to the third time point. The determination is based on whether these three time points constitute a stress cycle. The specific principle is: if the stress is satisfied... The value is between and Between these three consecutive time points, a complete stress cycle is formed. The stress amplitude and mean stress of this stress cycle are calculated. The stress amplitude is half the absolute value of the difference between the stresses at the first two time points of the stress cycle, and the mean stress is half the sum of the stresses at the first two time points of the stress cycle. Then, the first two time points of the stress cycle are deleted from the time series, and the third time point is retained. The same method is used to continue to determine whether it can form a stress cycle with other time points. The entire time series is traversed in chronological order until all time points have been processed, generating several stress cycles.

[0031] To obtain the tensile strength of the ship hull steel, the stress amplitude of each stress cycle is extracted and corrected to the equivalent stress amplitude. The specific principle of the correction is as follows: for each stress cycle, calculate the ratio of the mean stress to the tensile strength under that stress cycle, divide 1 by this ratio, and then divide the stress amplitude of that stress cycle by the result of the above division to obtain the equivalent stress amplitude of that stress cycle.

[0032] The equivalent stress amplitudes of all stress cycles are sorted in descending order. The interval between the maximum and minimum equivalent stress amplitudes is divided into several equally spaced micro-intervals. The number of stress cycles occurring within each micro-interval is counted. Stress cycles falling within the same micro-interval are merged, and the equivalent stress amplitude of that micro-interval is set as the average of the equivalent stress amplitudes of all stress cycles within that micro-interval. This generates several simplified data points, specifically in the following form: ,in, Indicates the first The equivalent stress amplitude in a small interval, Indices representing small intervals Indicates the first The number of stress cycles in a small interval;

[0033] The load spectrum is divided using an 8-level standard. The minimum and maximum equivalent stress amplitudes are found from the simplified data points. The logarithmic difference between the maximum and minimum equivalent stress amplitudes is calculated and divided by the level of the load spectrum to obtain the logarithmic interval. Starting from the minimum equivalent stress amplitude, the logarithmic stress intervals for each level are defined sequentially. The left endpoint of each logarithmic stress interval is calculated as follows: the difference between the level and 1 is multiplied by the logarithmic interval, and the product is added to the minimum equivalent stress amplitude. Alternatively, the product of the level and the logarithmic interval is calculated and added to the minimum equivalent stress amplitude. The midpoint of each logarithmic interval is taken and converted to a linear value, which is used as the representative stress amplitude for that level. This process is repeated for all simplified data points until all simplified data points are assigned to the logarithmic intervals of the corresponding levels. The total number of stress cycles for each level is recorded, generating several load spectra. Each element in the load spectrum is in the form of... ,in, Indicates the first The representative stress amplitude of the load spectrum. Indicates the index of the load spectral order. Indicates the first The total number of stress cycles in the load spectrum.

[0034] Furthermore, the principle for obtaining the fundamental SN function at different temperatures is as follows:

[0035] Within the range of -60℃ to 20℃, a temperature point is selected at 10℃ intervals, and fatigue tests are conducted at each temperature point to generate the basic SN function corresponding to each temperature point. The basic SN function is a curve with the remaining fatigue life as the dependent variable, the stress amplitude as the independent variable, and temperature as the influencing parameter. The relationship between the first material constant and the second material constant and temperature is obtained by fitting the nonlinear least squares method, and the above relationship is substituted into the basic SN function.

[0036] Furthermore, the principle of modifying the basic SN function based on historical cumulative damage is as follows:

[0037] The historical cumulative damage represents the ratio of consumed fatigue life to total fatigue life.

[0038] Each specimen is processed on a fatigue testing machine to consume its fatigue life. Different pre-damage is applied to each prepared specimen, which represents the historical cumulative damage of the specimen. At the same temperature point, different stress amplitudes are applied to the specimen, and the remaining fatigue life of the specimen under different stress amplitudes is measured to obtain the SN function at that temperature point. Based on the fatigue test data of multiple temperature points, two damage coupling coefficients are calculated. The two damage coupling coefficients include an intercept attenuation coefficient and a slope amplification coefficient. The intercept attenuation coefficient is used to describe the degree of attenuation of the intercept of the basic SN function by historical cumulative damage, and the slope amplification coefficient is used to describe the degree of amplification of the slope of the basic SN function by historical cumulative damage. The first material constant and the second material constant are corrected by the intercept attenuation coefficient and the slope amplification coefficient, respectively. Then, the corrected first material constant and the second material constant are combined to obtain the SN function under known temperature and known historical cumulative damage.

[0039] Furthermore, the principle for calculating the remaining fatigue life of critical components at each applied load spectrum is as follows:

[0040] For each critical area, apply the first... Real-time temperature corresponding to the first-level load spectrum and the second-level load spectrum Substituting the current cumulative damage at the critical location before the application of the load spectrum into the corrected SN function, we obtain the material's stress amplitude at the current cumulative damage and real-time temperature. The remaining fatigue life at that time.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] This invention, by collecting three types of historical data—basic field, discrete phase medium field, and hull motion field—can accurately describe the temporal evolution and state transition probabilities between seawater properties, sea ice state, and hull attitude. This makes the simulation input no longer an isolated numerical value, but a continuous, statistically consistent dynamic process. By using the CEL method to couple the hull with an Eulerian domain containing waves and ice fragments, it can effectively avoid mesh distortion problems caused by ice fragmentation and water flow between ice fragments, waves, and the hull, truly reflecting the multi-physics field synergy in polar navigation environments. By using the rainflow counting method to sort and divide the equivalent stress amplitude and merge cycles within the same interval, it not only preserves the statistical characteristics of the load but also improves the efficiency of subsequent fatigue calculations.

[0043] This invention also reflects the nonlinear coupling effect of temperature and damage by conducting fatigue tests on specimens with different pre-damage levels at different temperature points. By introducing temperature correction, the prediction results can reflect the fatigue performance degradation of materials under polar low-temperature environments. By introducing damage correction, the prediction results can reflect the cumulative effect of material performance degradation during service. By gradually applying multi-level load spectra to the modified SN curve model, the fatigue consumption under that load level is calculated once after each level of load spectrum is applied, and the cumulative damage is updated in real time as the input for the next level of calculation, thus realizing an accurate mapping from material properties to structural-level remaining life. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the method flow of an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of the stress time history of key parts in an embodiment of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0047] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0048] Example:

[0049] Please see Figures 1 to 2 The present invention provides a technical solution:

[0050] A method for calculating parameters of a ship hull fatigue damage evolution model under multi-field coupling, comprising the following steps:

[0051] Step 1: Collect historical environmental baseline field data, discrete phase medium field data, and ship motion field data for the target sea area. Construct a dynamic boundary sequence based on the above data. The environmental baseline field data includes seawater density and seawater dynamic viscosity. The discrete phase medium field data includes ice coverage, ice thickness, ice elastic modulus, compressive strength, and fracture energy. The ship motion field data includes ship speed, heading angle, roll angle, and pitch angle.

[0052] In this embodiment, the principle for constructing the dynamic boundary condition sequence is as follows:

[0053] Environmental baseline field data, discrete phase medium field data, and ship motion field data of the target sea area were collected daily over the past year. For each type of data, the continuously changing data was divided into several discrete states, and the probability of transitioning from one state to another was determined. A state transition probability matrix for each type of data was constructed, and a dynamic boundary sequence was constructed using the Markov chain Monte Carlo method.

[0054] The collected environmental baseline field data includes seawater density and seawater dynamic viscosity. Seawater density directly affects wave momentum, energy, and wave pressure on the hull, and the hull's motion attitude in the waves is also related to seawater density. Seawater density affects the hull's equilibrium position, which in turn affects the stress distribution in key areas. Seawater dynamic viscosity determines the damping effect of seawater on the hull's motion and changes the hydrodynamic coupling between ice fragments and the hull. Daily seawater density and dynamic viscosity for the target sea area are extracted from the ocean reanalysis data GLORY.

[0055] The collected discrete-phase medium field data includes ice coverage, ice thickness, ice elastic modulus, compressive strength, and fracture energy. Ice coverage determines the probability of the ship encountering ice fragments, directly affecting the intensity and frequency of ship-ice fragment collisions. Ice thickness affects the mass and inertia of the ice fragments, and the impact energy during a ship-ice fragment collision is positively correlated with ice thickness. The elastic modulus of ice determines the deformation characteristics of the ice fragments during the collision process, affecting the energy absorption capacity of the ice before it breaks apart, and thus affecting the impact load spectrum experienced by the ship. The compressive strength of ice is a core parameter of the Drucker-Prager yield criterion for ice materials, determining whether the ice breaks apart under the pressure of the ship, directly affecting the peak value and duration of the contact force. The collision between ice and the ship's hull generates high-stress-amplitude load cycles; ice fracture can be used to control energy dissipation in the ice material damage evolution criterion, determining the expansion of ice after breakage and affecting the attenuation characteristics of contact force; the daily ice coverage of the target sea area is obtained from AMSR-2 satellite remote sensing data and expressed as a percentage; the daily average ice thickness of the target sea area is extracted from the CryoSat-2 ice thickness remote sensing product as the ice thickness for that day; the daily average ice temperature and average ice salinity of the target sea area are extracted from the CMEMS sea ice reanalysis product, and 0℃ is set as the reference temperature to obtain the elastic modulus of pure ice at the reference temperature. Based on the daily average ice temperature and average ice salinity, the daily ice elastic modulus is calculated using the following formula: ,in, This indicates the average ice temperature over a day. This indicates the average salinity of ice on that day. This indicates that the average ice temperature is The average salinity of ice is The corresponding elastic modulus of ice at that time, These are the coefficients representing the influence of average ice temperature and average ice salinity on the elastic modulus. The value ranges from 0.005 to 0.01, and a larger value indicates a stronger temperature sensitivity; for permafrost, salinity has a smaller impact. The value ranges from 0.005 to 0.006. For nascent ice, the salinity is the highest, and the effect of salinity is significant. The value range is 0.008~0.001; the compressive strength of ice is calculated based on the daily ice temperature. The specific principle is as follows: obtain the compressive strength of pure ice at 0℃, and correct it based on the ice temperature and ice salinity to obtain the actual daily compressive strength of ice. The calculation formula is: ,in, This indicates that the average ice temperature is The average salinity of ice is The corresponding ice compressive strength, These represent the influence coefficients of average ice temperature and average ice salinity on compressive strength, respectively. The value range is 0.02~0.05. The value range is 0.01 to 0.03, and The larger the value, the stronger the compressive strength of ice at low temperatures; The larger the value of , the greater the decrease in compressive strength when the salinity increases by the same amount, indicating a greater impact of salinity on compressive strength. The fracture energy of ice is calculated based on compressive strength and elastic modulus, using the following formula: ,in, This indicates that the compressive strength of ice is The elastic modulus of ice is The time corresponds to the fracture energy of ice. This represents an empirical coefficient, ranging from 0.1 to 0.3, and for multi-year ice, its fracture energy is typically higher. A larger value indicates that nascent ice is more brittle and its fracture energy is typically lower. The value is relatively small.

[0056] The ship's motion field data includes speed, heading angle, roll angle, and pitch angle. Speed ​​determines the relative velocity between the ship and ice debris, thus affecting the hydrodynamics and collision kinetic energy of the ship. The heading angle determines the ship's attitude relative to the wave propagation direction, affecting the contact angle between the ship and ice debris, and thus affecting the directional component of the collision force. Roll and pitch angles are dynamic response indicators of the ship under the combined action of waves and ice debris, reflecting the ship's instantaneous attitude. Changes in the ship's attitude will change the relative position and force direction of key parts, such as the bow, sides, and bottom. The average speed and average heading angle recorded daily by shipborne sensors and the navigation log are used as the daily speed and heading angle, respectively. The average roll angle amplitude and average pitch angle amplitude recorded daily by the shipborne inertial measurement unit (IMU) are used as the daily roll angle and pitch angle.

[0057] The principle underlying the division of continuously changing data into several discrete states is as follows:

[0058] Based on the range of seawater density and seawater dynamic viscosity in historical data, the seawater is divided into three equal segments between the maximum and minimum values, and each segment is set as a state.

[0059] Ice coverage ranging from 0% to 100% is divided into five states: no ice (0-10% ice coverage); light ice (10-30% ice coverage); moderate ice (30-60% ice coverage); heavy ice (60-90% ice coverage); and full ice coverage (90-100% ice coverage).

[0060] Based on historical data on ice thickness, elastic modulus, compressive strength, and fracture energy, a corresponding range is obtained for each type of data. The range is then divided into five equal segments between the minimum and maximum values, with each segment representing a state.

[0061] Within the ship's speed range, it is divided into three equal segments, corresponding to three states: low speed, medium speed, and high speed.

[0062] The heading angle is divided into equal intervals within the range of 0 to 360°, with each angle range corresponding to a different state;

[0063] Based on the range of roll and pitch angles of the ship in historical data, the range between the maximum and minimum values ​​is divided into three equal segments, each segment corresponding to a state.

[0064] Taking ice coverage as an example, the principle of determining the probability of transitioning from one state to another and constructing the state transition probability matrix is ​​as follows: divide the ice coverage range of 0~100% into 5 states. Each index represents a state, mapping each collected ice coverage rate in the historical data to its corresponding state; the state transition matrix is ​​defined as a 5×5 matrix, and the nth index in the matrix... Line 1 The elements of the column are , indicating from state Transition to state The probability; historical data consists of data collected every day within the past year, with adjacent two days' data treated as a time pair. Based on the state changes of adjacent two days' data, the corresponding state transition frequency is increased; all time pairs in the historical data are traversed, and the state transitions of all time pairs are statistically analyzed to calculate the frequency of each state transition; for each state... In time pairs from state Transition to state The number of times Then the transition probability is: ,in, Indicates from state The sum of the number of transitions to other states, and Similarly, a state transition probability matrix is ​​constructed for other data, with the number of rows and columns of the matrix being the same as the number of states into which the data is divided.

[0065] The core idea of ​​constructing dynamic boundary sequences is as follows: Treat the change of each data point as a Markov process evolving over time, describe its state transition law using a state transition probability matrix, and then generate a random time series that conforms to this law using the Monte Carlo method, thereby constructing dynamic boundary conditions with statistical consistency. Specifically, let the index of the data acquisition time point be... The total number of time points is For any data, set its value at the th... The state corresponding to the value at each time point is: Then extract the first from the state transition probability matrix. The line indicates the current state at the current time point. At time t, the probability of changing to any other state at the next time point, from the t... Random sampling is performed based on the probability of each row, and the result of the random sampling is used as the state value for the next time point. This process is repeated until the state value for the next time point is determined through random sampling for all time points. The total number of state values ​​for each data type at all time points constitutes the state sequence for that data. Similarly, for all data, a sequence of length [length missing] is generated. The state sequence, extracting the first from the state sequence of each data point. There is a set of elements, each corresponding to a numerical range for a given state. The midpoint of this range is taken as the value of the data at the 1st state. The specific values ​​at each time point are combined into a vector, denoted as the boundary condition vector. Each boundary condition vector contains the specific values ​​of different data at the same time point. The boundary condition vectors of all time points are arranged in chronological order to form a dynamic boundary sequence.

[0066] The core function of dynamic boundary sequences is to transform static, discrete environmental parameters into continuous, statistically consistent time-series inputs, thereby providing time-varying realistic boundary conditions for the hull-ice debris-wave coupled model. Through the Markov chain Monte Carlo method, the dynamic boundary sequences retain the state transition laws and duration distributions, making the simulation input no longer an isolated value, but a stochastic process that conforms to the statistical laws of the real marine environment.

[0067] Step 2: Construct a hull-ice debris-wave coupling model based on CEL, apply a dynamic boundary sequence to the coupling model, obtain the stress of key parts of the hull in real time, and extract the stress time history of each key part. The key parts of the hull include the bow, side and bottom.

[0068] In this embodiment, the principle of constructing the hull-ice debris-wave coupling model is as follows:

[0069] The three-dimensional geometric model of the hull is obtained and imported into the finite element software. The three-dimensional geometric model of the hull is discretized using a Lagrange mesh. The density, elastic modulus, Poisson's ratio and yield strength of the hull material are obtained and imported into the discretized three-dimensional geometric model to construct the Lagrange finite element model of the hull structure.

[0070] Lagrange meshes are a type of mesh that moves with material deformation, suitable for describing the deformation and stress response of solid structures. During discretization, appropriate element types and mesh densities are selected based on the structural characteristics and stress gradient changes of different parts of the hull. In critical areas such as the bow, sides, and bottom, refined shell or solid elements are used for meshing to improve the accuracy of stress calculations; in regions with gradual structural changes, the mesh is coarsened to reduce computational costs. Mechanical property parameters of the materials used in the hull, including density, elastic modulus, Poisson's ratio, and yield strength, are obtained through material testing. These mechanical property parameters are then assigned to the discretized mesh elements to construct a Lagrange finite element model of the hull structure.

[0071] The Lagrange finite element model of the hull structure reflects the shape, size, spatial arrangement and connection relationship of the main structural components of the hull, and can realistically reflect the structural form of the hull during actual navigation.

[0072] An Eulerian mesh containing the water area around the ship hull and the ice zone is constructed to build the Eulerian domain. The Eulerian mesh includes two material components: water material and ice material. The water material is defined using the Newtonian fluid constitutive model, and the ice material is defined using the Drucker-Prager yield criterion combined with the damage evolution criterion.

[0073] The Eulerian mesh is fixed in space, within which water and ice materials can flow, deform, or break. An Eulerian mesh encompassing the water area surrounding the ship and the ice zone is constructed. This mesh is fixed in space and covers the area where the ship may navigate. The Eulerian domain contains two main components: water and ice. Water is defined as a Newtonian fluid in the Eulerian domain, using a Newtonian fluid constitutive model. It is assumed that the shear stress of the fluid is proportional to the velocity gradient, with the proportionality coefficient being the dynamic viscosity. For ice, before reaching its yield strength, the Drucker-Prager yield criterion is used to reflect the characteristic of increased yield strength under compression. After reaching the yield strength, the ice material undergoes stiffness degradation and breakage. A damage evolution criterion is introduced, using damage variables to characterize the degree of stiffness degradation as the loading process progresses. The damage variable ranges from 0 to 1, where 0 represents no damage and 1 represents complete damage, with the damage degree gradually increasing from 0 to 1.

[0074] Several contact pairs are established between the surface of the Lagrange finite element model of the hull structure and the Eulerian domain. The contact pairs are the areas where the hull structure and the Eulerian domain come into contact with each other. The contact force of each contact pair is calculated based on the penalty function method. The contact force of each contact pair is applied to the Lagrange mesh node corresponding to the hull structure, causing the deformation and movement of the hull structure. The contact force is used to realize the coupling between the hull structure and ice fragments and waves, thus constructing a hull-ice fragment-wave coupled model.

[0075] In finite element method software, a master surface and a slave surface are assigned to each contact pair. The master surface is the interface of the material in the Eulerian domain, and the slave surface is the Lagrangian element surface of the hull structure. The principle of the penalty function method is: when penetration occurs between contact surfaces, a contact force proportional to the penetration amount is applied to restore the contact constraint. The steps for calculating the contact force of each contact pair based on the penalty function method are: at each time point, monitor the relative position between the hull surface and the Eulerian domain material surface. If the slave surface penetrates the master surface, it is determined to be in a contact state, and this point is a contact pair. The penetration depth is the shortest distance from the node where the slave surface penetrates the master surface to the master surface. The normal contact force of this contact pair is the normal contact force. The contact stiffness is the product of the contact stiffness and the penetration depth. The normal contact stiffness is 10 to 100 times the elastic modulus of the hull material. When the value is 10 to 20 times, it is suitable for situations where hydrodynamics dominate, the ice layer is thin, or the ice fragments are small. In this case, the contact force is small, avoiding numerical oscillations, and it is suitable for working conditions where wave loads dominate. When the value is 20 to 50 times, it is suitable for situations where the ice layer is of moderate thickness and the hull contacts the ice frequently. In this case, it balances accuracy and convergence and is suitable for typical polar navigation conditions. When the value is 50 to 100 times, it is suitable for working conditions where impact loads are significant, such as large ice fragments. In this case, it ensures that the contact force is large enough to limit penetration. The calculated contact force is applied to the corresponding position of the hull structure's Lagrangian grid as an external force boundary condition, causing deformation and motion of the hull structure.

[0076] The principle for extracting the stress time history of each key component is as follows:

[0077] A dynamic boundary sequence is applied to the coupled model, and the total time of the dynamic boundary sequence is discretized into several time points. For each key part, the stress at each time point is recorded, and the stress time history of that key part is obtained.

[0078] After applying a dynamic boundary sequence to the coupled model, the stress of each key component at each time point is obtained. The stress of the same key component at all time points is integrated to obtain the stress time history of that key component.

[0079] Stress time history reflects the continuous change of stress over time in key components such as the bow, sides, and bottom of the ship under multi-field coupling. The influencing factors of stress include periodic wave loads, impact loads, the superposition effect of loads, and changes in hull attitude. Periodic loads are low-frequency stress fluctuations caused by waves, while impact loads are high-amplitude, short-term stress peaks caused by collisions between ice fragments and the hull. The superposition effect of loads is the coupling effect when wave loads and ice loads act simultaneously. Changes in hull attitude indicate that changes in the hull's roll, pitch, and heading angles will lead to changes in the direction and amplitude of forces on key components. By extracting stress time history, the environmental base field, discrete phase medium field, and hull motion field are transformed into the actual stress response of key components of the hull structure, allowing for targeted assessment of the fatigue damage evolution process of different key components.

[0080] Table 1 shows the stress changes over time at key locations. The stress time history of the bow region exhibits impact load characteristics, with multiple stress peaks exceeding 130 MPa, reflecting the frequent direct impacts of ice fragments on the bow during polar navigation. When the hull collides with ice floes, the bow, as the first part to come into contact with the ice, bears the greatest impact energy and experiences the most intense stress response. The stress quickly drops back after reaching its peak, reflecting the transient characteristics of the ice collision: short impact time, concentrated energy, and rapid attenuation. After each impact, the stress drops to around 50 MPa, indicating that the bow is in a relatively stable navigation state between impacts, mainly bearing the conventional loads caused by waves. The stress time history of the hull side region exhibits continuous compressive load characteristics, with a lower peak stress than the bow, but a longer duration of high stress, reflecting the continuous compressive effect of the ice on the hull side. The load changes in the hull side region are relatively continuous and smooth because, during hull navigation, the ice floes slide along the hull side, generating continuous contact pressure, rather than an instantaneous frontal impact. The bottom region exhibits periodic undulating load characteristics, with 4-5 time points constituting one undulating cycle. The bottom is mainly affected by waves; when the hull navigates in waves, the hydrostatic pressure and wave dynamic pressure on the bottom change periodically with the alternation of wave crests and troughs.

[0081] Table 1. Stress Variation of Key Components

[0082]

[0083] Step 3: The stress time history of each key component is processed using the rainflow counting method to obtain the stress amplitude and stress mean under each stress cycle, thereby generating a multi-level load spectrum for each key component.

[0084] In this embodiment, the principle for generating multi-level load spectra for each key component is as follows:

[0085] For each critical component, the stress time history is processed using the rainflow counting method to generate several stress cycles. Specifically, for each stress time history, the stress corresponding to three consecutive time points is read and recorded as follows: ,in This represents the stress corresponding to the first time point. This indicates the stress corresponding to the second time point. This represents the stress corresponding to the third time point. The determination is based on whether these three time points constitute a stress cycle. The specific principle is: if the stress is satisfied... The value is between and Between these three consecutive time points, a complete stress cycle is formed. The stress amplitude and mean stress of this stress cycle are calculated. The stress amplitude is half the absolute value of the difference between the stresses at the first two time points of the stress cycle, and the mean stress is half the sum of the stresses at the first two time points of the stress cycle. Then, the first two time points of the stress cycle are deleted from the time series, and the third time point is retained. The same method is used to continue to determine whether it can form a stress cycle with other time points. The entire time series is traversed in chronological order until all time points have been processed, generating several stress cycles.

[0086] The stresses at three consecutive time points are sequentially read from the stress time history and denoted as follows: If all three conditions are met: If these three time points constitute a complete stress cycle, then the formula for calculating the stress amplitude of a complete stress cycle is: , The formula for calculating the stress amplitude and mean stress is: , Represent the average stress, and record it. After this stress cycle, along with the corresponding stress amplitude and mean stress, the stress at the first two time points is removed from the stress time history. The stress at the third time point is retained and combined with the stress at other time points. The above judgment and calculation process is repeated until all time points are processed. After traversing the entire stress time history, a series of stress cycles are obtained, as well as the stress amplitude and stress mean corresponding to each stress cycle.

[0087] The generated stress cycles reflect the actual fatigue loads borne by key parts of the hull under multi-field coupling. During navigation, the stress time histories experienced by the hull are irregular and cannot be directly used for fatigue life calculation. By using the rainflow counting method, these stress time histories are decomposed into multiple complete stress cycles. At the same time, the periodic loads caused by waves, the high amplitude caused by ice collisions, the short-term impact loads, and the changes in stress direction and amplitude caused by hull motion are considered. Each cycle corresponds to a stress loading and unloading process. Moreover, the main stresses borne by different key parts are also different. The bow is mainly subjected to ice fragment impacts and wave pounding, the sides are subjected to ice compression and hydrodynamic forces, and the bottom is mainly subjected to wave pressure and floating ice friction.

[0088] To obtain the tensile strength of the ship hull steel, the stress amplitude of each stress cycle is extracted and corrected to the equivalent stress amplitude. The specific principle of the correction is as follows: for each stress cycle, calculate the ratio of the mean stress to the tensile strength under that stress cycle, divide 1 by this ratio, and then divide the stress amplitude of that stress cycle by the result of the above division to obtain the equivalent stress amplitude of that stress cycle.

[0089] The formula for calculating the equivalent stress amplitude of a stress cycle is:

[0090]

[0091] in, Indicates the equivalent stress amplitude. Indicates the stress amplitude of a stress cycle. This represents the mean stress value of a stress cycle. Indicates the tensile strength of the ship's hull steel;

[0092] In actual ship structures, the stress cycles experienced by critical components are often asymmetrical. The equivalent stress amplitude reflects the magnitude of the stress when an asymmetrical cyclic load is converted into a symmetrical cyclic load with the same fatigue damage. The larger the equivalent stress amplitude, the more severe the fatigue damage. The calculation of the equivalent stress amplitude is based on Goodman's theory, which describes the relationship between the stress amplitude and the mean stress at a given fatigue life. The specific relationship is as follows: , sorted out denominator It reflects the degree to which the mean stress weakens the fatigue load-bearing capacity of a material, using the tensile strength of the hull steel. As a correction factor, since tensile strength is the ultimate bearing capacity of a material under static load, reflecting the inherent property of a material to resist tensile failure, when... hour, This indicates that the current stress cycle is a symmetrical cycle and no correction of the stress amplitude is required. hour, The equivalent stress cycle is larger, indicating that the fatigue damage of the actual cycle is equivalent to a symmetrical cycle with a larger amplitude; when hour, The equivalent stress cycle is smaller, indicating that the fatigue damage of the actual cycle is equivalent to a symmetrical cycle with a smaller amplitude; when near When the denominator approaches 0, Approaching infinity indicates that near the static strength limit, even a very small stress amplitude can lead to rapid failure.

[0093] The equivalent stress amplitudes of all stress cycles are sorted in descending order. The interval between the maximum and minimum equivalent stress amplitudes is divided into several equally spaced micro-intervals. The number of stress cycles occurring within each micro-interval is counted. Stress cycles falling within the same micro-interval are merged, and the equivalent stress amplitude of that micro-interval is set as the average of the equivalent stress amplitudes of all stress cycles within that micro-interval. This generates several simplified data points, specifically in the following form: ,in, Indicates the first The equivalent stress amplitude in a small interval, Indices representing small intervals Indicates the first The number of stress cycles in a small interval;

[0094] Set the number of micro-intervals to 50. Between the maximum and minimum values, there are 50 small intervals of equal length, and the length of each small interval is... ,in, This represents the maximum value of the equivalent stress amplitude. This represents the minimum value of the equivalent stress amplitude.

[0095] The number of stress cycles output by the rainflow counting method is too large. Directly using all of them for subsequent fatigue life calculations would lead to excessive computational load and low efficiency. By dividing the equivalent stress amplitude into several small intervals and merging them, a large number of discrete stress cycles are compressed into a few representative simplified data points, reducing the data scale of subsequent load spectrum construction and life calculation. Each simplified data point corresponds to a small interval, containing the average value of the equivalent stress cycle amplitude within the small interval. The number of stress cycles within a small interval , This reflects the average damage level across all stress cycles within a small interval. It reflects the frequency information of stress cycles within a small interval. The higher the frequency, the more frequently the load of that amplitude level appears in the critical part, and the greater its contribution to the cumulative fatigue damage. All the simplified data points together constitute the amplitude distribution of the equivalent fatigue load spectrum borne by the critical parts of the hull under dynamic boundary conditions.

[0096] The load spectrum is divided using an 8-level standard. The minimum and maximum equivalent stress amplitudes are found from the simplified data points. The logarithmic difference between the maximum and minimum equivalent stress amplitudes is calculated and divided by the level of the load spectrum to obtain the logarithmic interval. Starting from the minimum equivalent stress amplitude, the logarithmic stress intervals for each level are defined sequentially. The left endpoint of each logarithmic stress interval is calculated as follows: the difference between the level and 1 is multiplied by the logarithmic interval, and the product is added to the minimum equivalent stress amplitude. Alternatively, the product of the level and the logarithmic interval is calculated and added to the minimum equivalent stress amplitude. The midpoint of each logarithmic interval is taken and converted to a linear value, which is used as the representative stress amplitude for that level. This process is repeated for all simplified data points until all simplified data points are assigned to the logarithmic intervals of the corresponding levels. The total number of stress cycles for each level is recorded, generating several load spectra. Each element in the load spectrum is in the form of... ,in, Indicates the first The representative stress amplitude of the load spectrum. Indicates the index of the load spectral order. Indicates the first The total number of stress cycles in the load spectrum.

[0097] From the generated simplified data points Find the minimum equivalent stress amplitude. and maximum equivalent stress amplitude The formula for calculating the logarithmic interval is: ;from Begin by defining the first... The formula for calculating the left endpoint of the logarithmic stress interval for level 1 is: The formula for calculating the right endpoint is: The midpoint of the logarithmic interval is: Convert the midpoint to a linear value to obtain the first... Representative stress amplitude of the load spectrum ; Traverse all simplified data points ,according to Within which logarithmic interval does the simplified data point fall, and what is the number of iterations for that data point? Accumulating these iterations to the total number of iterations for the corresponding series in the logarithmic interval yields an 8-level loading spectrum, where each element is of the form: , .

[0098] Fatigue load amplitudes often span multiple orders of magnitude. Using logarithmic coordinates for classification more reasonably reflects the contributions of low-amplitude and high-amplitude cycles to fatigue damage. Logarithmically, equal intervals correspond to geometrically ordered stress amplitude changes, consistent with the nonlinear characteristics of material fatigue performance. An 8-level load spectrum is a commonly used simplified form in fatigue analysis, derived from the combination of rainflow counting and cumulative fatigue damage theory. By compressing a large number of continuous load cycles into a few representative amplitude levels, the statistical characteristics of the load are preserved while significantly reducing the amount of data required for subsequent fatigue life calculations. Taking the midpoint of each logarithmic interval as a representative value minimizes the equivalent damage error of all cycles within that interval in a statistical sense. The load spectrum reflects the statistical characteristics of fatigue loads borne by key components of the hull under multi-field coupling. The severity of damage at each load level, To represent the frequency of occurrence of each load level, the load spectrum divides the equivalent stress amplitude into 8 levels at logarithmic intervals from minimum to maximum, visually demonstrating the stress amplitude distribution range borne by key parts of the hull. The number of cycles corresponding to each load level reflects the frequency of occurrence of loads with different amplitudes. Typically, it exhibits the typical fatigue load characteristics of frequent occurrence of low-amplitude loads and sparse occurrence of high-amplitude loads. The load spectrum is the final manifestation of the coupling effect of the environmental basic field, the discrete phase medium field, and the hull motion field. Differences in different sea states, ice conditions, and navigation conditions are ultimately transformed into changes in amplitude and frequency in the load spectrum.

[0099] Step 4: Prepare specimens based on the steel used for the hull, conduct fatigue tests on the specimens at different temperature points, obtain the basic SN function at different temperature points, and correct the basic SN function based on historical cumulative damage;

[0100] In this embodiment, the principle for obtaining the basic SN function at different temperatures is as follows:

[0101] Within the range of -60℃ to 20℃, a temperature point is selected at 10℃ intervals, and fatigue tests are conducted at each temperature point to generate the basic SN function corresponding to each temperature point. The basic SN function is a curve with the remaining fatigue life as the dependent variable, the stress amplitude as the independent variable, and temperature as the influencing parameter. The relationship between the first material constant and the second material constant and temperature is obtained by fitting the nonlinear least squares method, and the above relationship is substituted into the basic SN function.

[0102] To characterize the effect of low temperature environment on the fatigue performance of materials, fatigue tests were conducted at nine temperature points: -60℃, -50℃, -40℃, -30℃, -20℃, -10℃, 0℃, 10℃, and 20℃. At each temperature point, five different stress amplitude levels were selected, and five specimens were subjected to fatigue tests at each stress amplitude to obtain the remaining fatigue life data of each specimen.

[0103] The basic SN function takes a three-parameter form, with the remaining fatigue life N as the dependent variable, the stress amplitude S as the independent variable, and temperature T as an influencing parameter. The expression for the basic SN function is as follows: ,in, Represents the first material constant. Represents the second material constant. Indicates the stress amplitude. This represents the remaining fatigue life at that stress amplitude; at each temperature point, based on several sets of stress amplitude and remaining fatigue life data measured at that temperature point, let the index of the temperature point be... ,but Based on the least squares method, the temperature points are fitted. corresponding and ;

[0104] According to different temperature points and This results in several datasets, with the data in each dataset in the following format: , ,in Indicates the first The temperature at each temperature point;

[0105] The fitted relationship between the first material constant and temperature is as follows: The relationship between the second material constant and temperature is: ,in Let each represent its respective fitting coefficient; substitute the data from the dataset and fit the results using the least squares method. Substituting the fitting coefficients into the equation yields... ,Will Substituting the expression for the basic SN function, we obtain the basic SN function applicable to any temperature T.

[0106] First material constant This represents the baseline of the remaining fatigue life of the material under unit stress amplitude and determines the intercept of the basic SN function. The larger the value, the greater the remaining fatigue life of the material under the same stress amplitude, and the better its fatigue resistance; the second material constant. This reflects the sensitivity of fatigue life to changes in stress amplitude and determines the slope of the basic SN function. The larger the value, the more sensitive the material's fatigue life is to changes in stress amplitude. One determines the baseline level of the remaining fatigue life, and the other determines the rate at which the fatigue life decays with stress. Combining the two to generate the basic SN function determines the remaining fatigue life of the material under any temperature and stress amplitude without damage, and reflects the fatigue law of short life under high stress amplitude and long life under low stress amplitude.

[0107] The principle of modifying the basic SN function based on historical cumulative damage is as follows:

[0108] The historical cumulative damage represents the ratio of consumed fatigue life to total fatigue life.

[0109] Each specimen is processed on a fatigue testing machine to consume its fatigue life. Different pre-damage is applied to each prepared specimen, which represents the historical cumulative damage of the specimen. At the same temperature point, different stress amplitudes are applied to the specimen, and the remaining fatigue life of the specimen under different stress amplitudes is measured to obtain the SN function at that temperature point. Based on the fatigue test data of multiple temperature points, two damage coupling coefficients are calculated. The two damage coupling coefficients include an intercept attenuation coefficient and a slope amplification coefficient. The intercept attenuation coefficient is used to describe the degree of attenuation of the intercept of the basic SN function by historical cumulative damage, and the slope amplification coefficient is used to describe the degree of amplification of the slope of the basic SN function by historical cumulative damage. The first material constant and the second material constant are corrected by the intercept attenuation coefficient and the slope amplification coefficient, respectively. Then, the corrected first material constant and the second material constant are combined to obtain the SN function under known temperature and known historical cumulative damage.

[0110] The basic SN curve reflects the fatigue life of a material without damage. However, in actual service, materials generally bear a certain fatigue load, resulting in cumulative damage. Therefore, the influence of cumulative damage is introduced into the basic SN curve to construct a modified SN function. For each specimen, the stress level and number of cycles are randomly set on a fatigue testing machine to consume the fatigue life of each specimen, and the historical cumulative damage of each specimen is calculated. This represents the historical cumulative damage of the specimen. At the same temperature point, different stress amplitudes are applied to specimens with historical cumulative damage, and their remaining fatigue life is measured. The corrected SN function takes the form: ;in This represents the remaining fatigue life after introducing cumulative damage. , These are the first and second material constants after introducing cumulative damage, respectively. , , This represents the intercept attenuation coefficient. This represents the slope amplification factor; the material constants at each temperature point are known under non-damaging conditions. , By keeping other conditions constant, the material constants at each temperature point under the condition of introducing cumulative damage are obtained. , Then the intercept attenuation coefficient under this cumulative damage Slope amplification factor This led to the construction of two datasets, one for all historical cumulative damage. and ;in Indicates the first A historical accumulation of damage, An index representing historical cumulative damage. Indicates temperature as Historical cumulative damage The corresponding intercept attenuation coefficient, Indicates temperature as Historical cumulative damage The slope amplification factor corresponding to the time;

[0111] The function for fitting the intercept attenuation coefficient is: The function for fitting the slope amplification factor is: ,in Let represent the fitting coefficients, respectively. Substitute the dataset into the inputs and use the least squares method to obtain the fitting coefficients. ,Will Substitute to get and The fitting formula, through and right , Make corrections to obtain and This leads to the modified SN function.

[0112] The modified SN function considers both historical cumulative damage and temperature effects. After damage accumulation, the remaining fatigue life of a material depends not only on the current load and environment but also on the proportion of fatigue life already consumed. The greater the damage, the weaker the material's fatigue resistance. By applying different pre-damage conditions to the specimens, the intercept attenuation coefficient and slope amplification coefficient are introduced to modify the intercept and slope of the SN function. This reflects the nonlinear weakening effect of the fatigue loads already borne by the material during service on its remaining life. The greater the historical cumulative damage, the lower the intercept of the SN curve, indicating a decrease in initial life; the steeper the slope, the more sensitive it is to changes in stress amplitude. The modified SN function achieves the goal of predicting the remaining life of a material at a given temperature point based on its known historical cumulative damage.

[0113] Step 5: Substitute the load spectra of different key parts into the modified SN function in sequence to calculate the remaining fatigue life of each key part when the load spectrum is applied each time.

[0114] In this embodiment, the principle for calculating the remaining fatigue life of key components at each applied load spectrum is as follows:

[0115] Set the initial cumulative damage to zero, and apply the first [damage] to each critical site. Real-time temperature corresponding to the first-level load spectrum and the second-level load spectrum Substituting the current cumulative damage at the critical location before the application of the load spectrum into the corrected SN function, we obtain the material's stress amplitude at the current cumulative damage and real-time temperature. The remaining fatigue life at that time.

[0116] The corrected SN function is: , The current cumulative damage of key components is calculated using real-time temperature data. Thus, the application of the first The remaining fatigue life of key components after the first load spectrum.

[0117] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0118] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0119] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0120] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for calculating the parameters of a hull fatigue damage evolution model under the action of multiple fields, characterized in that, The specific steps include: Step 1: Collect historical environmental baseline field data, discrete phase medium field data, and ship motion field data for the target sea area. Construct a dynamic boundary sequence based on the above data. The environmental baseline field data includes seawater density and seawater dynamic viscosity. The discrete phase medium field data includes ice coverage, ice thickness, ice elastic modulus, compressive strength, and fracture energy. The ship motion field data includes ship speed, heading angle, roll angle, and pitch angle. Step 2: Construct a hull-ice debris-wave coupling model based on CEL, apply a dynamic boundary sequence to the coupling model, obtain the stress of key parts of the hull in real time, and extract the stress time history of each key part. The key parts of the hull include the bow, side and bottom. Step 3: The stress history of each key component is processed using the rainflow counting method to obtain the stress amplitude and average stress under each stress cycle, thereby generating a multi-level load spectrum for each key component. Step 4: Prepare specimens based on the steel used for the hull, conduct fatigue tests on the specimens at different temperature points, obtain the basic SN function at different temperature points, and correct the basic SN function based on historical cumulative damage; Step 5: Substitute the load spectra of different key parts into the modified SN function in sequence to calculate the remaining fatigue life of each key part when the load spectrum is applied each time. The modified S-N function has the form: ; in, This represents the remaining fatigue life after introducing cumulative damage. Indicates the stress amplitude. Indicates temperature. This indicates the historical cumulative damage of the specimen. , These are the first and second material constants after introducing cumulative damage, respectively. , , This represents the first material constant when there is no cumulative damage. This represents the second material constant when there is no cumulative damage. This represents the intercept attenuation coefficient. Indicates the slope amplification factor; It is known that the material constant of each temperature point is obtained under the condition of no damage , , The index represents the temperature point, and the material constant of each temperature point is obtained under the condition of introducing cumulative damage under the control of other conditions , The intercept attenuation coefficient under this cumulative damage is: , and the slope amplification coefficient is: ; Two datasets were then constructed, one for all historical cumulative damage. and ;in Indicates the first A historical accumulation of damage, An index representing historical cumulative damage. Indicates temperature as Historical cumulative damage The corresponding intercept attenuation coefficient, Indicates temperature as Historical cumulative damage The slope amplification factor corresponding to the time; The function for fitting the intercept attenuation coefficient is: The function for fitting the slope amplification factor is: ,in Let represent the fitting coefficients, respectively. Substitute the dataset into the inputs and use the least squares method to obtain the fitting coefficients. ,Will Substitute to get and The fitting formula, through and right , Make corrections to obtain and This leads to the modified SN function.

2. The method according to claim 1, characterized in that: The principle behind constructing the dynamic boundary condition sequence in step 1 is as follows: Environmental baseline field data, discrete phase medium field data, and ship motion field data of the target sea area were collected daily over the past year. For each type of data, the continuously changing data was divided into several discrete states, and the probability of transitioning from one state to another was determined. A state transition probability matrix for each type of data was constructed, and a dynamic boundary sequence was constructed using the Markov chain Monte Carlo method.

3. The method of claim 2, wherein the method is characterized by: The principle underlying the division of continuously changing data into several discrete states is as follows: Based on the range of seawater density and seawater dynamic viscosity in historical data, the seawater is divided into three equal segments between the maximum and minimum values, and each segment is set as a state. Ice coverage ranging from 0% to 100% is divided into five states: no ice (0-10% ice coverage); light ice (10-30% ice coverage); moderate ice (30-60% ice coverage); heavy ice (60-90% ice coverage); and full ice coverage (90-100% ice coverage). Based on historical data on ice thickness, elastic modulus, compressive strength, and fracture energy, a corresponding range is obtained for each type of data. The range is then divided into five equal segments between the minimum and maximum values, with each segment representing a state. Within the ship's speed range, it is divided into three equal segments, corresponding to three states: low speed, medium speed, and high speed. The heading angle is divided into equal intervals within the range of 0 to 360°, with each angle range corresponding to a different state; Based on the range of roll and pitch angles of the ship in historical data, the range between the maximum and minimum values ​​is divided into three equal segments, each segment corresponding to a state.

4. The method of claim 1, wherein the method is characterized by: The principle behind constructing the hull-ice debris-wave coupling model in step 2 is as follows: The three-dimensional geometric model of the hull is obtained and imported into the finite element software. The three-dimensional geometric model of the hull is discretized using a Lagrange mesh. The density, elastic modulus, Poisson's ratio and yield strength of the hull material are obtained and imported into the discretized three-dimensional geometric model to construct the Lagrange finite element model of the hull structure. An Eulerian mesh containing the water area around the ship hull and the ice zone is constructed to build the Eulerian domain. The Eulerian mesh includes two material components: water material and ice material. The water material is defined using the Newtonian fluid constitutive model, and the ice material is defined using the Drucker-Prager yield criterion combined with the damage evolution criterion. Several contact pairs are established between the surface of the Lagrange finite element model of the hull structure and the Eulerian domain. The contact pairs are the areas where the hull structure and the Eulerian domain come into contact with each other. The contact force of each contact pair is calculated based on the penalty function method. The contact force of each contact pair is applied to the Lagrange mesh node corresponding to the hull structure, causing the deformation and movement of the hull structure. The contact force is used to realize the coupling between the hull structure and ice fragments and waves, thus constructing a hull-ice fragment-wave coupled model.

5. The method of claim 1, wherein the method is characterized by: The principle behind extracting the stress time history of each key component in step 2 is as follows: A dynamic boundary sequence is applied to the coupled model, and the total time of the dynamic boundary sequence is discretized into several time points. For each key part, the stress at each time point is recorded, and the stress time history of that key part is obtained.

6. The method for calculating parameters of a ship hull fatigue damage evolution model under multi-field coupling as described in claim 5, characterized in that: The principle behind generating the multi-level load spectrum for each key component in step 3 is as follows: For each critical component, the stress time history is processed using the rainflow counting method to generate several stress cycles. Specifically, for each stress time history, the stress corresponding to three consecutive time points is read and recorded as follows: ,in This represents the stress corresponding to the first time point. This indicates the stress corresponding to the second time point. This represents the stress corresponding to the third time point. The determination is based on whether these three time points constitute a stress cycle. The specific principle is: if the stress is satisfied... The value is between and Between these three consecutive time points, a complete stress cycle is formed. The stress amplitude and mean stress of this stress cycle are calculated. The stress amplitude is half the absolute value of the difference between the stresses at the first two time points of the stress cycle, and the mean stress is half the sum of the stresses at the first two time points of the stress cycle. Then, the first two time points of the stress cycle are deleted from the time series, and the third time point is retained. The same method is used to continue to determine whether it can form a stress cycle with other time points. The entire time series is traversed in chronological order until all time points have been processed, generating several stress cycles. To obtain the tensile strength of the ship hull steel, the stress amplitude of each stress cycle is extracted and corrected to the equivalent stress amplitude. The specific principle of the correction is as follows: for each stress cycle, calculate the ratio of the mean stress to the tensile strength under that stress cycle, divide 1 by this ratio, and then divide the stress amplitude of that stress cycle by the result of the above division to obtain the equivalent stress amplitude of that stress cycle. The equivalent stress amplitudes of all stress cycles are sorted in descending order. The interval between the maximum and minimum equivalent stress amplitudes is divided into several equally spaced micro-intervals. The number of stress cycles occurring within each micro-interval is counted. Stress cycles falling within the same micro-interval are merged, and the equivalent stress amplitude of that micro-interval is set as the average of the equivalent stress amplitudes of all stress cycles within that micro-interval. This generates several simplified data points, specifically in the following form: ,in, Indicates the first The equivalent stress amplitude in a small interval, Indices representing small intervals Indicates the first The number of stress cycles in a small interval; The load spectrum is divided using an 8-level standard. The minimum and maximum equivalent stress amplitudes are found from the simplified data points. The logarithmic difference between the maximum and minimum equivalent stress amplitudes is calculated and divided by the level of the load spectrum to obtain the logarithmic interval. Starting from the minimum equivalent stress amplitude, the logarithmic stress intervals for each level are defined sequentially. The left endpoint of each logarithmic stress interval is calculated by multiplying the difference between the level and 1 by the logarithmic interval, and then adding the product to the minimum equivalent stress amplitude. The right endpoint is calculated by multiplying the level by the logarithmic interval, and then adding the product to the minimum equivalent stress amplitude. The midpoint of each logarithmic interval is taken and converted to a linear value, which is used as the representative stress amplitude for that level. This process is repeated for all simplified data points until all simplified data points are assigned to the logarithmic intervals of the corresponding levels. The total number of stress cycles for each level is recorded, generating several load spectra. Each element in the load spectrum is in the form of... ,in, Indicates the first The representative stress amplitude of the load spectrum. Indicates the index of the load spectral order. Indicates the first The total number of stress cycles in the load spectrum.

7. The method for calculating parameters of a ship hull fatigue damage evolution model under multi-field coupling as described in claim 6, characterized in that, The principle behind obtaining the basic SN function at different temperatures in step 4 is as follows: Within the range of -60℃ to 20℃, a temperature point is selected at 10℃ intervals, and fatigue tests are conducted at each temperature point to generate the basic SN function corresponding to each temperature point. The basic SN function is a curve with the remaining fatigue life as the dependent variable, the stress amplitude as the independent variable, and temperature as the influencing parameter. The relationship between the first material constant and the second material constant and temperature is obtained by fitting the nonlinear least squares method, and the above relationship is substituted into the basic SN function.

8. The method for calculating parameters of a ship hull fatigue damage evolution model under multi-field coupling as described in claim 7, characterized in that, The principle behind modifying the basic SN function based on historical cumulative damage in step 4 is as follows: The historical cumulative damage represents the ratio of consumed fatigue life to total fatigue life. Each specimen is processed on a fatigue testing machine to consume its fatigue life. Different pre-damage is applied to each prepared specimen, which represents the historical cumulative damage of the specimen. At the same temperature point, different stress amplitudes are applied to the specimen, and the remaining fatigue life of the specimen under different stress amplitudes is measured to obtain the SN function at that temperature point. Based on the fatigue test data of multiple temperature points, two damage coupling coefficients are calculated. The two damage coupling coefficients include an intercept attenuation coefficient and a slope amplification coefficient. The intercept attenuation coefficient is used to describe the degree of attenuation of the intercept of the basic SN function by historical cumulative damage, and the slope amplification coefficient is used to describe the degree of amplification of the slope of the basic SN function by historical cumulative damage. The first material constant and the second material constant are corrected by the intercept attenuation coefficient and the slope amplification coefficient, respectively. Then, the corrected first material constant and the second material constant are combined to obtain the SN function under known temperature and known historical cumulative damage.

9. The method for calculating parameters of a ship hull fatigue damage evolution model under multi-field coupling as described in claim 8, characterized in that, The principle behind calculating the remaining fatigue life of critical components in step 5 for each applied load spectrum is as follows: For each critical area, apply the first... Real-time temperature corresponding to the first-level load spectrum and the second-level load spectrum Substituting the current cumulative damage at the critical location before the application of the load spectrum into the corrected SN function, we obtain the material's stress amplitude at the current cumulative damage and real-time temperature. The remaining fatigue life at that time.

Images