A method for predicting linear wave loads in ship structural design

Abstract

This invention provides a method for predicting linear wave loads in ship structural design, relating to the field of ship engineering structural design technology. Based on optimized Latin hypercube sampling and three-dimensional potential flow calculation, this invention constructs a dataset of hydrodynamic pressure fields and load components under regular wave conditions. It extracts the basis function system and establishes a linear mapping relationship between the combination coefficients and the loads. For irregular wave sea states, it interpolates to determine the state transition matrix and control matrix. Combined with the main wave disturbance force, it performs a time-series recursion of the combination coefficients, triggering local transient calculations at preset intervals, and performing sparse verification corrections on the recursive state. Finally, it outputs the load history in real time through the mapping relationship and statistically analyzes the extreme values ​​to generate a wave load design parameter table. This method replaces full-order solutions with low-dimensional recursion, triggering a small number of full-order verifications at preset intervals. While maintaining prediction accuracy comparable to full-order solutions, it significantly reduces computation time, providing efficient and reliable load input for ship structural design.

Description

Technical Field

[0001] This invention relates to the field of ship engineering structural design technology, specifically a method for predicting linear wave loads in ship structural design. Background Technology

[0002] In the process of ship structural design, accurately predicting the wave loads that the hull will experience when sailing in waves is the core prerequisite for carrying out longitudinal strength verification, fatigue life assessment and structural dimension optimization. The accuracy of wave load prediction directly determines the safety margin and economy of the hull beam component dimensions. If the predicted load value is too low, it will lead to insufficient structural design strength, which may cause major safety accidents such as hull beam buckling, fatigue cracks or even fractures during actual navigation. If the predicted load value is too high, it will cause an unnecessary increase in steel consumption, increase construction costs and reduce the ship's cargo carrying capacity and fuel economy.

[0003] In existing technologies, there are three main approaches to ship wave load prediction. The first approach is frequency domain calculation based on three-dimensional linear potential flow theory. This approach treats the hull as a rigid body, solves the Laplace equation in regular waves to obtain the hydrodynamic pressure distribution on the wetted surface of the hull, and then obtains the load transfer function of each check profile through integration. Combined with a wave spectrum model, short-term and long-term statistical forecasts are made. This method has high computational efficiency and is the most widely used in ship design units. The second approach is time-domain nonlinear method based on the coupling of computational fluid dynamics and finite element method. This approach directly solves the viscous flow field control equations and simultaneously considers the reaction of the hull's elastic deformation to the flow field. It can capture strong nonlinear load components such as slamming, wave inrush, and flutter, and has high prediction accuracy. The third approach is direct load measurement based on model tests. By installing a force balance or segmented force beam on a scaled-down self-propelled ship model, irregular wave conditions are simulated in a water tank, and the load history is directly recorded. This method is considered the most reliable means of obtaining wave loads. However, the aforementioned existing technologies all have their own insurmountable limitations in practical engineering applications. Although the three-dimensional linear potential flow frequency domain method is computationally fast, its theoretical basis is based on the assumptions of small-amplitude waves, ideal fluids, and linearization of the ship's hull without speed. When a ship sails at a high speed in severe sea conditions, the geometry of the wetted surface of the hull changes significantly with the wave surface, and the nonlinear effect of the free surface is significantly enhanced. At this time, there is a non-negligible systematic deviation between the load prediction results obtained by the frequency domain method based on the principle of linear superposition and the actual load. While the high-precision computational fluid dynamics coupling method can partially solve the above nonlinear problems, completing the full-time domain simulation of hundreds of wave cycles under an irregular wave state usually requires a very long computation time. This is completely impractical in ship design practice with tight design cycles and a large number of working condition combinations. More importantly, there is no existing technology that can simultaneously balance the computational efficiency of the three-dimensional potential flow method and the prediction accuracy of the high-fidelity simulation method in the design stage.

[0004] 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

[0005] The purpose of this invention is to provide a method for predicting linear wave loads in ship structural design, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for predicting linear wave loads in ship structural design, comprising the following steps: Step 1: Obtain the data parameters of the target ship. Based on the design speed, wave height and heading angle range, and using optimized Latin hypercube sampling, generate regular wave conditions. Use the three-dimensional potential flow algorithm to solve the hydrodynamic pressure field on the wetted surface of the hull and the load components of the verification profile under each regular wave condition to form the design load dataset. Step 2: Perform intrinsic orthogonal decomposition on the hydrodynamic pressure field, extract the first few modes that satisfy the energy proportion threshold as the basis function system, project to obtain the combination coefficients, establish a linear mapping relationship model between the combination coefficients and the load components, and generate a load mapping file; Step 3: Input the sea state parameters of the irregular wave sea state to be evaluated, call the load mapping file, interpolate the combination coefficients of adjacent regular wave conditions according to the sea state parameters, determine the state transition matrix and control matrix, calculate the main wave interference force of the hull based on the JONSWAP wave energy spectrum model and take into account the phase compensation of local beam variation, and establish the state recursive calculation formula. Step 4: The combination coefficients are progressively recursively calculated with a fixed recursive time step. When the number of recursive steps reaches the preset interval determined by the ratio of the spectral peak period to the recursive step size, the reconstruction verification is triggered. The three-dimensional potential flow solver is called to solve the verification pressure field at full order and the pressure deviation is compared with the reconstructed pressure field. Based on the pre-calibrated deviation correction mapping relationship, the correction amount of the combination coefficients is obtained and the recursion continues. The linear mapping relationship model is used to convert the load component time history and generate a wave load parameter table after statistical and extreme value envelope processing.

[0007] Furthermore, the process of generating the design load dataset specifically includes: acquiring the data parameters of the target ship, which include retrieving the hull plate shape point coordinate data of the target ship from the ship design database, as well as the weight and longitudinal position data of the center of gravity of each station distributed along the ship's length. The shape point coordinate data is used to define the geometric boundary of the wetted surface of the hull, and the weight and longitudinal position data of the center of gravity are used to determine the inertial force distribution characteristics of the hull beams. Within a multidimensional parameter space defined by the lower and upper limits of design speed, wave height, and heading angle, optimized Latin hypercube sampling is used to generate a preset number of regular wave conditions. For each regular wave condition, a hydrodynamic solver based on three-dimensional potential flow theory is used to solve the wetted surface mesh of the hull, obtaining the amplitude and phase distribution of pulsating pressure at each wetted surface mesh node. The amplitude and phase distribution of pulsating pressure are then used as the hydrodynamic pressure field corresponding to that regular wave condition. The pressure distribution is integrated along the ship's length, and the vertical bending moment, horizontal bending moment, and torque at a pre-specified check profile are extracted as load components. The regular wave condition is associated with and stored with the corresponding hydrodynamic pressure field and load components to form a design load dataset.

[0008] Furthermore, the process of performing intrinsic orthogonal decomposition on the hydrodynamic pressure field, extracting the basis function system, and establishing a linear mapping relationship model specifically includes: arranging the hydrodynamic pressure fields under all regular wave conditions in the design load dataset into a high-dimensional sample matrix by columns; performing intrinsic orthogonal decomposition on the high-dimensional sample matrix to obtain a series of mutually orthogonal basis vectors; calculating the energy contribution rate of each basis vector in reconstructing the entire hydrodynamic pressure field in the high-dimensional sample matrix; accumulating each basis vector in descending order of energy contribution rate; and determining the basis vectors participating in the accumulation as the basis function system when the accumulated value exceeds the preset energy proportion threshold. For each hydrodynamic pressure field in the design load dataset, it is projected onto the basis function system to obtain a set of combination coefficients that correspond one-to-one with each hydrodynamic pressure field. At the same time, the load components corresponding to each hydrodynamic pressure field are extracted, and a correspondence is established between the combination coefficients of each hydrodynamic pressure field and the corresponding load components. Based on this correspondence, a linear mapping relationship model is established using the least squares method. The combination coefficients are then converted into load components through this linear mapping relationship model.

[0009] Furthermore, the process of generating and calling the load mapping file specifically includes: packaging and storing the distribution data of each basis vector in the basis function system on the wet surface grid nodes, the model parameters of the linear mapping relationship model, and the boundary parameters of the regular wave condition in the design load dataset into a binary format data file, and using this data file as the load mapping file; Input the sea state parameters of the irregular wave sea state to be evaluated, including speed, significant wave height, spectral peak period, and heading angle. Based on the input speed, retrieve and load the corresponding load mapping file. Using the combination coefficients as intermediate variables, use the speed, significant wave height, spectral peak period, and heading angle of the irregular wave sea state to be evaluated as search conditions. Perform multidimensional linear interpolation between the combination coefficients of adjacent regular wave working conditions according to the sea state parameters to determine the state transition matrix and control matrix. By processing the meaningful wave height and spectral peak period included in the sea state parameters using the JONSWAP wave energy spectrum model, the time history data of wave rise is obtained. Based on the time history data of wave rise, heading angle, and wetted surface geometry of the hull, and combined with the wave phase difference caused by local changes in the wetted surface beam of the hull for phase compensation, the wave principal disturbance force varying with time is calculated. Based on the state transition matrix, control matrix, and wave principal disturbance force, a state recursive calculation formula with combination coefficients as intermediate variables is established. This state recursive calculation formula expresses the recursive relationship between the combination coefficients of the next recursive time step, the combination coefficients of the current recursive time step, and the wave principal disturbance force of the current recursive time step.

[0010] Furthermore, the process of determining the state transition matrix and control matrix through multidimensional interpolation specifically includes: In the regular wave condition parameter space covered by the design load dataset, the spatial distance between the sea state parameters of the irregular wave sea state to be evaluated and each regular wave condition parameter is calculated. Several groups of regular wave conditions with the smallest spatial distance are identified as adjacent regular wave conditions. The combination coefficients corresponding to the adjacent regular wave conditions are extracted, as well as the time series change rate data of the combination coefficients of the adjacent regular wave conditions as a function of the number of recursion steps in their respective recursion processes. Using the normalized distance between each component of the sea state parameters of the irregular wave sea state to be evaluated and the adjacent regular wave condition parameters as weights, the time series change rate data of the adjacent regular wave conditions are weighted and averaged to obtain the state transition matrix corresponding to the irregular wave sea state to be evaluated. Using the normalized distance between each component of the sea state parameter of the irregular wave sea state to be evaluated and the parameters of the adjacent regular wave working conditions as weights, the combination coefficients of the adjacent regular wave working conditions under the action of the main wave disturbance force at unit wave height are weighted and averaged to obtain the control matrix corresponding to the irregular wave sea state to be evaluated.

[0011] Furthermore, the process of progressively executing state recursion along the time sequence of irregular wave sea states to update the combination coefficients and trigger reconstruction verification specifically includes: using a fixed recursion time step, processing the combination coefficients and the main wave disturbance force of the current recursion time step through the state recursion calculation formula step by step to obtain the combination coefficients of the next recursion time step; during the recursion execution process, determining whether the number of recursion steps executed since the start of the recursion has reached the preset interval step number, and triggering reconstruction verification when the preset interval step number is reached; During the reconstruction and verification, the three-dimensional potential flow solver is invoked to perform a full-order solution with the instantaneous wavefront rise and the instantaneous attitude of the hull at the current recursive time step as boundary conditions to obtain the verification hydrodynamic pressure field. At the same time, the combination coefficients of the current recursive time step are projected onto the basis function system to reconstruct the reconstructed pressure field of the current recursive time step. The pressure deviation between the verification hydrodynamic pressure field and the reconstructed pressure field is calculated. The pressure deviation is converted into a combination coefficient correction amount through a pre-calibrated deviation correction mapping relationship. The combination coefficient correction amount is added to the combination coefficients of the current recursive time step to obtain the corrected combination coefficients. The corrected combination coefficients are used as the starting point for the next recursive time step to continue the recursion. The pre-calibration method for the deviation correction mapping relationship is as follows: statistically analyze the pressure deviation and corresponding combination coefficient error between the hydrodynamic pressure field and the reconstructed pressure field under each regular wave condition in the design load dataset, and establish the mapping relationship between the pressure deviation and the combination coefficient error.

[0012] Furthermore, the process of converting the combination coefficients into load component time history and performing statistical and extreme value envelope processing using the linear mapping relationship model specifically includes: during the recursive execution process, each recursive time step calls the linear mapping relationship model to process the combination coefficients of the current recursive time step, and converts them into the vertical bending moment, horizontal bending moment and torque of the current recursive time step; The vertical bending moment, horizontal bending moment and torque obtained from each recursive time step are recorded sequentially according to the time series of irregular wave and sea state to form the time history of vertical bending moment load component, horizontal bending moment load component and torque load component. Zero-crossing analysis was performed on the time histories of vertical bending moment load components, horizontal bending moment load components, and torque load components, respectively. Load cycles that cross the mean were extracted from the time histories of each load component, and the amplitude of each load cycle was recorded. Based on the statistical results of the amplitude of each load cycle, the maximum value of each load component under the irregular wave sea state was calculated.

[0013] Furthermore, the process of generating a wave load parameter table for ship structural design specifically includes: comparing the maximum vertical bending moment, maximum horizontal bending moment, and maximum torque calculated under the current irregular wave state with the historical extreme value envelope of the corresponding load components under the sea state already evaluated for the ship type; if the current maximum value exceeds the range of the historical extreme value envelope, then the extreme value envelope of the ship type is updated, and a wave load parameter table containing the design values ​​of vertical bending moment, horizontal bending moment, and torque at the check section is output.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This invention generates a design load dataset covering multiple operating conditions by combining optimized Latin hypercube sampling with three-dimensional potential flow calculation. It also extracts the basis function system of the pressure field using intrinsic orthogonal decomposition, compressing the high-dimensional wet surface pressure distribution into a linear combination of a small number of dominant modes. On this basis, a linear mapping relationship model between the combination coefficients and load components is further established, so that the load prediction under irregular wave conditions no longer depends on the repeated solution of the full-order flow field, but is transformed into a time-series recursion of the low-dimensional combination coefficients. This order reduction process greatly reduces the computational cost of a single time step. Compared with the existing full-order time-domain simulation method, the overall computation time is significantly reduced within the same sea state duration, providing a feasible efficiency basis for rapid evaluation of multiple operating conditions in the design stage. This invention determines the state transition matrix and control matrix by interpolating irregular wave sea state parameters into the rate of change of multidimensional regular wave working condition combination coefficients, and introduces the main wave disturbance force based on the JONSWAP spectrum as an external excitation to construct a state recursive calculation formula that conforms to the laws of physical evolution. While performing the recursion, a reconstruction verification step triggered by a preset interval step is set. The verification pressure field is obtained by calling local transient calculation, and the deviation between it and the reconstructed pressure field is calculated. After being processed by the gain weight matrix, a correction increment is formed and fed back to the combination coefficients. This sparse correction mechanism autonomously suppresses the accumulation and divergence of recursion error without the need for external measured data, ensuring that the long-term load forecast results are consistent with the high-fidelity simulation trend. This invention combines intrinsic orthogonal decomposition reduced-order modeling with state recursion sparse verification. This method, without relying on the large amount of computational resources of model tests or full-order time-domain simulation, takes into account the computational efficiency of linear potential flow methods and the accuracy advantages of high-fidelity simulation, providing a practical and reliable technical tool for rapid load assessment and structural optimization iteration in the preliminary and detailed design stages of ships. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall method flow of the present invention; Figure 2 This is a fitting curve of the prior combination coefficients and the posterior combination coefficients of the present invention. Detailed Implementation

[0016] 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.

[0017] 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.

[0018] Example: Please see Figures 1-2 The present invention provides a technical solution: A method for predicting linear wave loads in ship structural design, comprising the following steps: Step 1: Obtain the data parameters of the target ship. Based on the design speed, wave height and heading angle range, and using optimized Latin hypercube sampling, generate regular wave conditions. Use the three-dimensional potential flow algorithm to solve the hydrodynamic pressure field on the wetted surface of the hull and the load components of the verification profile under each regular wave condition to form the design load dataset. The process of generating the design load dataset specifically includes: obtaining the data parameters of the target ship, which include the hull body shape point coordinate data of the target ship retrieved from the ship design database, as well as the weight and longitudinal position data of the center of gravity of each station distributed along the ship's length. The shape point coordinate data is used to define the geometric boundary of the wetted surface of the hull, and the weight and longitudinal position data of the center of gravity are used to determine the inertial force distribution characteristics of the hull beams. In this invention, the data parameters of the target vessel refer to the original design information retrieved from the ship design database, used to define the hull geometry and mass distribution. Specifically, the data parameters include the hull outer plating shape point coordinate data and the weight data and longitudinal center of gravity position data corresponding to each station along the ship's length. The shape point coordinate data is a set of three-dimensional coordinates of discretely distributed sampling points on the hull outer plating surface along the ship's length. These sampling points constitute a control point sequence defining the geometric boundary of the wetted surface of the hull, from which interpolation can be used to generate the complete curved surface of the wetted hull. The weight data refers to the sum of the hull structure weight and the loaded weight at each station along the ship's length. The longitudinal center of gravity position data refers to the longitudinal distance from the center of gravity to the midship or stern reference plane corresponding to the total measured weight at each station. The mass distribution curve and inertial property distribution of the hull beams along the ship's length are calculated from the weight data and longitudinal center of gravity position data at each station. This inertial property distribution includes the mass moment of inertia at each station of the hull beams. The design speed range is defined by the lower and upper speed limits, which are determined based on the speed range specified in the design specifications of the target vessel. The design wave height range is defined by the lower and upper wave height limits, which are determined based on wave dispersion statistics for the target vessel's navigation area, covering the main wave height ranges the target vessel will encounter during its operational cycle. The design heading angle range is defined by the lower and upper heading angle limits. The heading angle is the angle between the wave propagation direction and the vessel's navigation direction, and this angle range covers the vessel's navigation states when facing, following, and oblique waves. The speed range, wave height range, and heading angle range together span a three-dimensional operating parameter space. Within a multidimensional parameter space defined by the lower and upper limits of design speed, wave height, and heading angle, optimized Latin hypercube sampling is used to generate a preset number of regular wave conditions. This ensures that each parameter combination has uniform coverage within the multidimensional parameter space and can unbiasedly represent each sub-region, guaranteeing that the generated parameter combinations can unbiasedly represent all sub-regions within the multidimensional parameter space. Each parameter combination is defined as a regular wave condition, which refers to wave excitation calculation conditions consisting of a single determined speed value, a single determined wave height value, and a single determined heading angle value. For each regular wave condition, a hydrodynamic solver based on three-dimensional potential flow theory is used to solve the wetted surface mesh of the hull, obtaining the pulsating pressure amplitude and phase distribution at each wetted surface mesh node. The pulsating pressure amplitude and phase distribution are used as the hydrodynamic pressure field corresponding to the regular wave condition. For each regular wave condition, using the ship's speed, wave height, and heading angle as boundary conditions, a hydrodynamic solver based on three-dimensional potential flow theory is employed to calculate the hydrodynamic response of the wetted hull surface. Three-dimensional potential flow theory assumes the fluid to be an ideal fluid that is irrotational, inviscid, and incompressible, using the velocity potential function as the fundamental unknown to describe the interaction between the wave and the hull. The hydrodynamic solver numerically solves the velocity potential function using both surface boundary conditions and free surface boundary conditions, calculating the complex values ​​of the pulsating pressure generated at each grid node on the wetted hull surface under a unit amplitude regular wave incidence. The numerical value is a complex number containing a modulus component and an argument component, where the modulus component represents the oscillation amplitude of the pulsating pressure at the node, and the argument component represents the phase shift of the pulsating pressure at the node relative to the rise of the incident wave surface. The modulus component is extracted from the complex numerical value of the pulsating pressure as the amplitude of the pulsating pressure, and the argument component is extracted as the phase. The spatial distribution of the amplitude and phase of the pulsating pressure at each wetted surface grid node is taken as a whole and defined as the hydrodynamic pressure field corresponding to this regular wave condition. The hydrodynamic pressure field is a two-dimensional scalar field that corresponds one-to-one with the wetted surface grid node, and the value of each node is determined by the two components of the pulsating pressure amplitude and phase. The pressure distribution is integrated along the ship's length, and the vertical bending moment, horizontal bending moment, and torque at a pre-specified check profile are extracted as load components. The regular wave conditions are associated with and stored with the corresponding hydrodynamic pressure field and load components to form a design load dataset. After obtaining the hydrodynamic pressure field corresponding to each regular wave condition, the pressure distribution is integrated along the longitudinal integration path of the wetted surface of the hull in the hydrodynamic pressure field. The longitudinal integration path refers to one or more preset transverse strips of the wetted surface along the ship's length. During the integration process, for each longitudinal integration micro-segment, the pulsating pressure amplitude of each grid node within the transverse range of the wetted surface covered by the micro-segment is extracted, and the transverse vertical distance from each grid node to the neutral axis of the hull beam at the check section is included as the lever arm. The pulsating pressure amplitude is multiplied by the stress arm to obtain the contribution of each node to the bending moment of the check section. The bending moment contributions of all nodes are accumulated along the longitudinal direction to obtain the vertical bending moment unit wave amplitude response value of the check section. The verification section refers to one or more transverse section locations along the ship's length, pre-specified by the designers based on the structural strength assessment requirements of the hull. This location is selected as the midship section or other local extreme value locations of the bending moment envelope. The torque calculation also integrates the pressure distribution along the longitudinal integration path, and the transverse force arm contribution from the pressure at each node to the torsional center of the verification section is included in the transverse direction. The unit amplitude response value of the torque of the verification section is obtained through integration. The horizontal bending moment is calculated along the longitudinal integration path and the vertical force arm contribution from the pressure at each node to the neutral axis is included in the vertical direction. The unit amplitude response values ​​of the vertical bending moment, horizontal bending moment, and torque are collectively referred to as the load components of this regular wave condition. The load components refer to the internal force components borne at a specified section of the hull under given wave excitation conditions. The working parameters, hydrodynamic pressure field, and load components of each regular wave working condition are associated and stored in a one-to-one correspondence to form a design load dataset. The design load dataset is a structured database, in which each row corresponds to a regular wave working condition, including the speed, wave height, and heading angle as working condition parameter fields, all nodal pulsating pressure amplitude and phase data of the hydrodynamic pressure field under that working condition as pressure field data blocks, and vertical bending moment, horizontal bending moment, and torque values ​​as load component fields. The purpose of the design load dataset is to provide a sample set for the subsequent eigenorthogonal decomposition and order reduction processing, and at the same time to provide a regular wave working condition reference benchmark for multidimensional interpolation to determine the state transition matrix and control matrix.

[0019] Step 2: Perform intrinsic orthogonal decomposition on the hydrodynamic pressure field, extract the first few modes that satisfy the energy proportion threshold as the basis function system, project to obtain the combination coefficients, establish a linear mapping relationship model between the combination coefficients and the load components, and generate a load mapping file; In step 2 of the present invention, the hydrodynamic pressure field generated in step 1 is subjected to intrinsic orthogonal decomposition, the basis function system is extracted and a linear mapping relationship model is established. The purpose is to extract a low-dimensional modal basis that can characterize the main features of the spatial variation of the pressure field from a large number of high-dimensional hydrodynamic pressure field samples, and to establish a fast conversion relationship between the combination coefficients and the load components through the modal basis. The process of performing intrinsic orthogonal decomposition, extracting basis function system, and establishing linear mapping relationship model on hydrodynamic pressure field specifically includes: arranging the hydrodynamic pressure fields under all regular wave conditions in the design load dataset into a high-dimensional sample matrix by column; performing intrinsic orthogonal decomposition on the high-dimensional sample matrix to obtain a series of mutually orthogonal basis vectors; calculating the energy contribution rate of each basis vector in reconstructing all hydrodynamic pressure fields in the high-dimensional sample matrix; accumulating each basis vector in descending order of energy contribution rate; and determining the basis vectors participating in the accumulation as the basis function system when the accumulated value exceeds the preset energy proportion threshold. The hydrodynamic pressure fields under all regular wave conditions in the design load dataset are arranged in columns to form a high-dimensional sample matrix. The specific operation is as follows: each hydrodynamic pressure field is expanded into a column vector according to the unified numbering order of the wet surface grid nodes. The length of the column vector is equal to the total number of wet surface grid nodes multiplied by the sum of the amplitude and phase components of the pulsating pressure stored in each node. The column vectors corresponding to all regular wave conditions are arranged in columns and combined into a matrix with a number of rows much larger than the number of columns. This matrix is ​​the high-dimensional sample matrix. Each column of the high-dimensional sample matrix represents a complete snapshot of the hydrodynamic pressure field under a regular wave condition, and each row represents the value sequence of a certain pressure component of a certain grid node under all regular wave conditions. Eigenorthogonal decomposition of a high-dimensional sample matrix refers to processing the high-dimensional sample matrix using singular value decomposition (SVD) or covariance matrix eigenvalue decomposition (COVDE) algorithms to obtain a series of mutually orthogonal basis vectors. Essentially, eigenorthogonal decomposition extracts a set of orthogonal bases from the high-dimensional sample matrix, such that any column vector in the high-dimensional sample matrix can be represented as a linear combination of these orthogonal bases. The obtained basis vectors are arranged in descending order of their corresponding singular values. Each basis vector is a vector with the same length as the original hydrodynamic pressure field column vectors, representing a fundamental deformation mode of the hydrodynamic pressure field in space. Different basis vectors are mutually orthogonal, ensuring that the fundamental deformation modes are linearly independent. After obtaining a series of mutually orthogonal basis vectors, it is necessary to select the top few basis vectors that contribute the most to the reconstruction of the original hydrodynamic pressure field to form a basis function system. The selection is based on calculating the energy contribution rate of each basis vector. The energy contribution rate of a basis vector is defined as the proportion of the square of the singular value corresponding to that basis vector to the sum of the squares of all singular values. This proportion reflects the proportion of the original data variance that can be explained by reconstructing the high-dimensional sample matrix using only that basis vector. The basis vectors are accumulated in descending order of energy contribution rate. When the accumulated value exceeds the preset energy proportion threshold, the accumulation stops. The basis vectors participating in the accumulation are determined as the basis function system. The preset energy proportion threshold is a pre-set percentage value, which means the overall reconstruction accuracy requirement of the basis function system for the entire hydrodynamic pressure field in the original design load dataset. The basis function system consists of a finite number of basis vectors, which is much smaller than the dimension of the original hydrodynamic pressure field, thus realizing the order reduction expression of the high-dimensional pressure field data. For each hydrodynamic pressure field in the design load dataset, it is projected onto the basis function system to obtain a set of combination coefficients that correspond one-to-one with each hydrodynamic pressure field. At the same time, the load components corresponding to each hydrodynamic pressure field are extracted, and a correspondence is established between the combination coefficients of each hydrodynamic pressure field and the corresponding load components. Based on this correspondence, a linear mapping relationship model is established using the least squares method. The combination coefficients are then converted into load components through this linear mapping relationship model. The process of generating and calling the load mapping file specifically includes: packaging and storing the distribution data of each basis vector in the basis function system on the wet surface grid nodes, the model parameters of the linear mapping relationship model, and the boundary parameters of the regular wave condition in the design load dataset into a binary format data file, and using this data file as the load mapping file; The projection operation refers to calculating the inner product of the column vector of the hydrodynamic pressure field with each basis vector in the basis function system. The inner product values ​​of each order are arranged in the order of the basis vectors to form a set of combination coefficients corresponding to the hydrodynamic pressure field. The number of combination coefficients is equal to the number of basis vectors in the basis function system. Each combination coefficient represents the weight of the corresponding basis vector in the hydrodynamic pressure field. Therefore, any hydrodynamic pressure field can be approximately reconstructed by a linear combination of the basis function system and the corresponding combination coefficients. The combination coefficients obtained by projecting each hydrodynamic pressure field are then correlated with the vertical bending moment, horizontal bending moment and torque values ​​corresponding to the hydrodynamic pressure field, forming a set of sample pairs with combination coefficients as input and load components as output. Based on the aforementioned set of sample pairs, a linear mapping model is established using the least squares method. This model is a mathematical model that maps combination coefficients to load components. Specifically, it is a linear transformation matrix, where the number of rows equals the number of load components, and the number of columns equals the number of basis vectors in the basis function system. The goal of the least squares method is to calculate a linear transformation matrix based on the combination coefficient inputs and load component outputs of all sample pairs. This matrix minimizes the sum of squares of errors between the predicted and actual load component values ​​after converting the combination coefficients of each sample into predicted load component values. Once this linear mapping model is established, for any given set of combination coefficients, the corresponding check profile load components can be directly obtained through this model without needing to re-invoke the three-dimensional potential flow solver for full-order pressure field integration. The distribution data of each basis vector in the basis function system on the wetted surface grid nodes, the model parameters of the linear mapping relationship model, and the boundary parameters of the regular wave load in the design load dataset are packaged and stored as a binary data file, which is the load mapping file. The distribution data of each basis vector in the basis function system on the wetted surface grid nodes refers to the specific values ​​of the amplitude and phase components of the pulsating pressure on each grid node of the wetted surface corresponding to each basis vector in the basis function system. The model parameters of the linear mapping relationship model refer to the values ​​of each element in the linear transformation matrix obtained by the least squares method. The boundary parameters of the regular wave load refer to the lower limit, upper limit, lower limit, upper limit, lower limit, and upper limit of the heading angle corresponding to all regular wave loads in the design load dataset. The load mapping file is the final output of step 2 and the only ship type feature database file that needs to be called when performing the irregular wave load recursive calculation in step 3.

[0020] Step 3: Input the sea state parameters of the irregular wave sea state to be evaluated, call the load mapping file, interpolate the combination coefficients of adjacent regular wave conditions according to the sea state parameters, determine the state transition matrix and control matrix, calculate the main wave interference force of the hull based on the JONSWAP wave energy spectrum model and take into account the phase compensation of local beam variation, and establish the state recursive calculation formula. In step 3 of this invention, the sea state parameters of the irregular wave sea state to be evaluated are input and the load mapping file generated in step 2 is called. A state recursive calculation formula with combination coefficients as intermediate variables is established through multidimensional interpolation and wave force calculation, providing a complete calculation model for the time-series recursion in step 4. Input the sea state parameters of the irregular wave sea state to be evaluated, including speed, significant wave height, spectral peak period, and heading angle. Based on the input speed, retrieve and load the corresponding load mapping file. Using the combination coefficients as intermediate variables, use the speed, significant wave height, spectral peak period, and heading angle of the irregular wave sea state to be evaluated as search conditions. Perform multidimensional linear interpolation between the combination coefficients of adjacent regular wave working conditions according to the sea state parameters to determine the state transition matrix and control matrix. Irregular wave state to be evaluated refers to a wave excitation state defined by the actual ocean wave environment that needs to be evaluated in the structural design of the target ship. The difference between irregular wave state and regular wave state is that: regular wave state consists of waves with a single definite frequency, a single definite wave height, and a single definite direction, while irregular wave state is described by a wave spectrum model containing multiple frequency components, multiple wave height components, and multiple direction components. Sea state parameters are a set of quantitative indicators describing the characteristics of irregular wave sea states. Specifically, they include speed, meaningful wave height, spectral peak period, and heading angle. Meaningful wave height refers to the average value of the first third of the largest wave heights after arranging all wave heights recorded over a period of time in order of magnitude, representing the average energy level of the irregular wave sea state. Spectral peak period refers to the period value corresponding to the maximum value of the wave spectral density function, representing the periodic characteristics of the wave component with the most concentrated energy in the irregular wave sea state. After inputting the sea state parameters of the irregular wave sea state to be evaluated, the data content corresponding to the speed value in the load mapping file generated in step 2 is retrieved and loaded based on the speed value in the sea state parameters. The correspondence between the load mapping file and the speed has been established through file indexing when it is packaged and stored in step 2. Using combination coefficients as intermediate variables means that the combination coefficients serve as transitional state quantities connecting the input sea state parameters and the output state recursive calculation formula in this step. Their values ​​are not directly used to calculate the final load, but rather as a set of intermediate variables characterizing the instantaneous state of the ship's hydrodynamic pressure field to participate in the iterative process of state recursive calculation. The speed, meaningful wave height, spectral peak period, and heading angle of the irregular wave sea state to be evaluated are used as search conditions to locate the irregular wave working condition parameter space covered by the design load dataset. The regular wave working condition parameter space refers to the set of discrete parameter points composed of the speed, wave height, and heading angle of all regular wave working conditions in step 1. Each discrete point corresponds to a known regular wave working condition. In the parameter space of regular wave conditions, the spatial distance between the sea state parameters of the irregular sea state to be evaluated and the parameters of each regular wave condition is calculated. The spatial distance is defined as the square root of the sum of the squares of the differences between each component of the sea state parameter vector and the corresponding component of the regular wave condition parameter vector. Several groups of regular wave conditions with the smallest spatial distance are identified as adjacent regular wave conditions. Adjacent regular wave conditions refer to the set of regular wave conditions that enclose the sea state parameter point of the irregular sea state to be evaluated in the parameter space and form a minimum convex hull enclosing the point. The time series rate of change data corresponding to each adjacent regular wave condition is extracted. The time series rate of change data refers to the sequence of the rate of change of the combination coefficients under the regular wave condition with the recursive time step. Using the normalized distance between each component of the sea state parameters of the irregular sea state to be evaluated and the adjacent regular wave condition parameters as weights, the time series rate of change data of the adjacent regular wave conditions is weighted and averaged to obtain the state transition matrix corresponding to the irregular sea state to be evaluated. The normalized distance refers to the ratio obtained by dividing the spatial distance between the sea state parameter vector and an adjacent regular wave condition parameter vector by the sum of the spatial distances corresponding to all adjacent regular wave conditions, and using the ratio as the weight coefficient corresponding to that adjacent regular wave condition; the state transition matrix is ​​a square matrix whose order is equal to the number of basis vectors in the basis function system. This matrix acts on the combination coefficients of the current recursive time step and outputs the state transition components of the combination coefficients as the recursive time step evolves under the condition of no external excitation; Using the normalized distance between each component of the sea state parameters of the irregular wave state to be evaluated and the parameters of the adjacent regular wave state as weights, the combined coefficient response data of the adjacent regular wave states under the action of the main wave disturbance force at unit wave height are weighted and averaged to obtain the control matrix corresponding to the irregular wave state to be evaluated. The main wave disturbance force at unit wave height refers to the theoretical value of the wave disturbance force acting on the hull when the incident wave height is taken as a unit value. The combined coefficient response data refers to the incremental response of the combined coefficients under the action of the main wave disturbance force at unit wave height. The number of rows of the control matrix is ​​equal to the number of basis vectors in the basis function system, and the number of columns is equal to the dimension of the main wave disturbance force vector. This matrix acts on the main wave disturbance force at the current recursive time step and outputs the driving component of the main wave disturbance force on the change of the combined coefficients. By processing the meaningful wave height and spectral peak period included in the sea state parameters using the JONSWAP wave energy spectrum model, the time history data of wave rise is obtained. Based on the time history data of wave rise, heading angle, and wetted surface geometry of the hull, and combined with the wave phase difference caused by local changes in the wetted surface beam of the hull, phase compensation is performed to calculate the wave principal disturbance force that varies with time. Based on the state transition matrix, control matrix, and wave principal disturbance force, a state recursive calculation formula with combination coefficients as intermediate variables is established. This state recursive calculation formula expresses the recursive relationship between the combination coefficients of the next recursive time step, the combination coefficients of the current recursive time step, and the wave principal disturbance force of the current recursive time step. The JONSWAP wave energy spectrum model is a semi-empirical mathematical model describing the distribution of wave energy with frequency. Its input parameters are meaningful wave height and spectral peak period, and the output is the distribution of the wave spectral density function in the frequency domain. By discretizing the wave spectral density function at equal intervals in the frequency domain and generating a random phase for each discrete frequency point, the sine waves of each frequency component are superimposed using the harmonic superposition method to obtain the wave rise time history data that varies with time. The wave rise time history data is a sequence of wave vertical displacement values ​​arranged in chronological order, and the time interval is consistent with the recursive time step. Based on wave rise time history data, heading angle, and wetted surface geometry of the hull, a time-varying sequence of wave principal disturbance forces is calculated. The wave principal disturbance force refers to the time-varying external wave excitation force acting on the hull under irregular wave conditions. The calculation process is as follows: extract the instantaneous wave rise value at each recursive time step based on the wave rise time history data, determine the wave number component of the incident wave along the ship's length direction by combining the heading angle information, integrate the buoyancy generated by wave rise per unit length along the waterline length of the hull, and take into account the wave phase difference caused by local changes in the hull's beam during the integration process. Local beam variation refers to the change in transverse width of the wetted surface of the hull at different longitudinal positions. The beam variation will cause the wave front rise of the incident wave at the same moment to reach its peak at different longitudinal positions at different times. This time difference corresponds to a phase offset. When integrating, the wave front rise value at each longitudinal position needs to be phase compensated according to the corresponding phase offset before buoyancy is calculated. The resultant buoyancy force obtained by integration and the torque generated by it on the center of gravity of the ship are taken as the main wave disturbance force. The main wave disturbance forces of all time steps are arranged in chronological order to form the main wave disturbance force sequence. Based on the state transition matrix, control matrix, and wave principal disturbance force sequence, a state recursive calculation formula is established with combination coefficients as intermediate variables. The state recursive calculation formula is a discrete-time dynamic equation describing the evolution of combination coefficients with recursive time steps. The equation is expressed as follows: the combination coefficient of the next recursive time step is equal to the state transition matrix multiplied by the combination coefficient of the current recursive time step, plus the control matrix multiplied by the wave principal disturbance forces of each recursive time step. The addition operation in the state recursive calculation formula is element-wise addition of vectors, and the multiplication operation is matrix-vector multiplication. Through this state recursive calculation formula, given the combination coefficient of the initial recursive time step, the wave principal disturbance force sequence can be recursively calculated step by step to obtain the combination coefficient sequence of all time steps within the entire irregular wave sea state assessment period.

[0021] Step 4: The combination coefficients are progressively recursively calculated with a fixed recursive time step. When the number of recursive steps reaches the preset interval determined by the ratio of the spectral peak period to the recursive step size, the reconstruction verification is triggered. The three-dimensional potential flow solver is called to solve the verification pressure field at full order and the pressure deviation is compared with the reconstructed pressure field. Based on the pre-calibrated deviation correction mapping relationship, the correction amount of the combination coefficients is obtained and the recursion continues. The linear mapping relationship model is used to convert the load component time history and generate a wave load parameter table after statistical and extreme value envelope processing. In step 4 of the present invention, the state recursive calculation formula established in step 3 is executed step by step along the time sequence of irregular wave sea state. During the recursion process, the reconstruction verification is triggered at a preset interval step number to correct the combination coefficients. The combination coefficients are converted into load component time history in real time through the linear mapping relationship model established in step 2. Finally, the wave load parameter table is generated through statistical and extreme value envelope processing. The process of determining the state transition matrix and control matrix through multidimensional interpolation specifically includes: In the regular wave condition parameter space covered by the design load dataset, the spatial distance between the sea state parameters of the irregular wave sea state to be evaluated and each regular wave condition parameter is calculated. Several groups of regular wave conditions with the smallest spatial distance are identified as adjacent regular wave conditions. The combination coefficients corresponding to the adjacent regular wave conditions are extracted, as well as the time series change rate data of the combination coefficients of the adjacent regular wave conditions as a function of the number of recursion steps in their respective recursion processes. Using the normalized distance between each component of the sea state parameters of the irregular wave sea state to be evaluated and the adjacent regular wave condition parameters as weights, the time series change rate data of the adjacent regular wave conditions are weighted and averaged to obtain the state transition matrix corresponding to the irregular wave sea state to be evaluated. In the parameter space of regular wave conditions, each discrete point corresponds to a known regular wave condition. The coordinates of this point are determined by three coordinate components: speed, wave height, and heading angle. The spatial distance is defined as the square root of the sum of the squares of the differences in each coordinate component between the sea state parameter vector of the irregular wave condition to be evaluated and the parameter vector of the regular wave condition. Among the sea state parameters of the irregular wave condition to be evaluated—speed, significant wave height, spectral peak period, and heading angle—three components are selected for spatial distance calculation: speed, significant wave height, and heading angle. The periodic values ​​are used for the calculation of the JONSWAP wave energy spectrum model; the groups of regular wave conditions with the smallest spatial distance are identified as adjacent regular wave conditions. The criteria for determining adjacent conditions are as follows: sort all regular wave conditions in ascending order of spatial distance, select the regular wave conditions that are ranked first and whose spatial distance is less than a preset distance threshold as adjacent regular wave conditions, or select the sea state parameter points corresponding to the regular wave conditions that are ranked first and have a sufficient number, and determine the regular wave conditions that can form the minimum convex hull in the multidimensional parameter space and enclose the irregular wave sea state to be evaluated as adjacent regular wave conditions. Extract the time-series change rate data corresponding to adjacent regular wave operating conditions. The time-series change rate data refers to the change rate sequence of the combination coefficients under the regular wave operating condition with the recursive time step. For a regular wave operating condition, the time-series change rate of its combination coefficients is pre-calculated in the following way: Under the regular wave operating condition, state recursion is performed with a fixed recursive time step, the combination coefficient value of each recursive time step is recorded, the difference of the combination coefficients between adjacent recursive time steps is calculated, and the difference is divided by the fixed recursive time step to obtain the time-series change rate data of the combination coefficients under the regular wave operating condition with the recursive time step. Using the normalized distance between each component of the sea state parameter of the irregular wave sea state to be evaluated and the parameters of the adjacent regular wave working conditions as the weight, the combination coefficients of the adjacent regular wave working conditions under the action of the main wave disturbance force at unit wave height are weighted and averaged to obtain the control matrix corresponding to the irregular wave sea state to be evaluated. Normalized distance refers to the ratio obtained by dividing the spatial distance between the sea state parameter vector and an adjacent regular wave condition parameter vector by the sum of the spatial distances corresponding to all adjacent regular wave conditions. This ratio is used as the weight coefficient corresponding to the adjacent regular wave condition. The specific operation of weighted averaging is as follows: multiply each element in the time series rate of change data of each adjacent regular wave condition by the weight coefficient corresponding to the adjacent regular wave condition to obtain the weighted time series rate of change data. Add the weighted time series rate of change data of all adjacent regular wave conditions according to the corresponding element positions to obtain the combined coefficient state transition matrix. Using the normalized distance between each component of the sea state parameters of the irregular wave state to be evaluated and the parameters of adjacent regular wave states as weights, the combination coefficients of adjacent regular wave states under the action of the main wave disturbance force at unit wave height are weighted and averaged to obtain the control matrix corresponding to the irregular wave state to be evaluated. The main wave disturbance force at unit wave height refers to the theoretical value of the main wave disturbance force acting on the hull when the incident wave height is taken as unit wave amplitude. The combination coefficient under the action of the main wave disturbance force at unit wave height refers to the incremental response value generated by the combination coefficient when it is only subjected to the main wave disturbance force at unit wave height. This incremental response value is obtained in advance by calculating the response of the combination coefficient under the excitation of the main wave disturbance force at unit wave height in each adjacent regular wave state. The process of progressively executing state recursion along the time sequence of irregular wave sea states to update the combination coefficients and trigger reconstruction verification specifically includes: using a fixed recursion time step, processing the combination coefficients and the main wave disturbance force of the current recursion time step through the state recursion calculation formula step by step to obtain the combination coefficients of the next recursion time step; during the recursion execution process, determining whether the number of recursion steps executed since the start of the recursion has reached the preset interval step number, and triggering reconstruction verification when the preset interval step number is reached; The fixed recursive time step refers to the time interval between adjacent time points in the wave rise time history data generated in step 3. This time step also serves as the discrete time step of the state recursive calculation formula. With the fixed recursive time step, the combination coefficients and the main wave disturbance force of the current recursive time step are processed step by step through the state recursive calculation formula to obtain the combination coefficients of the next recursive time step. The specific form of the state recursive calculation formula is: the combination coefficient of the next recursive time step is equal to the state transition component obtained by multiplying the state transition matrix by the combination coefficient of the current recursive time step, plus the sum of the control input component obtained by multiplying the control matrix by the main wave disturbance force of the current recursive time step. The recursion step count refers to the number of state recursions executed from the recursion start time step to the current recursion time step. The combination coefficient of the recursion start time step is determined as follows: at the start of the recursion, the ship is in a calm water state, the main wave interference force is zero, and the combination coefficient is taken as a zero vector. A zero vector is a vector in which all element values ​​are zero. The preset interval step number is determined based on the ratio between the spectral peak period of the irregular wave sea state to be evaluated and the fixed recursive time step. The integer value between one-quarter and one-half of the ratio is taken. When the remainder of the recursive step number of the current recursive time step divided by the preset interval step number is zero, it is determined that the preset interval step number has been reached, and the reconstruction verification is triggered. During the reconstruction and verification process, the three-dimensional potential flow solver is invoked to perform a full-order solution using the instantaneous wavefront rise and the instantaneous attitude of the hull at the current recursive time step as boundary conditions, thereby obtaining the verification hydrodynamic pressure field. Simultaneously, the combination coefficients of the current recursive time step are projected onto the basis function system to reconstruct the reconstructed pressure field for the current recursive time step. The projection operation involves performing an inner product operation between the hydrodynamic pressure field of the current recursive time step and each basis vector in the basis function system to obtain the combination coefficients corresponding one-to-one with each basis vector. The reconstruction operation involves linearly combining the combination coefficients with the basis function system to obtain the reconstructed pressure field. The pressure deviation between the verification hydrodynamic pressure field and the reconstructed pressure field is calculated, that is, the pressure field difference between the verification hydrodynamic pressure field and the reconstructed pressure field at the corresponding wet surface grid node. The pressure deviation is converted into a combination coefficient correction amount through a pre-calibrated deviation correction mapping relationship. The combination coefficient correction amount is added to the combination coefficient of the current recursive time step to obtain the corrected combination coefficient. The corrected combination coefficient is used as the starting point for the next recursive time step to continue the recursion. Specifically, the formula for correcting the combination coefficients of the reconstruction verification is as follows: in, This serves as an index for the recursive time step under irregular wave sea conditions. For the first The combination coefficients, calculated using the state recursion formula at each recursive time step, have a dimension equal to the number of basis vectors in the basis function system. For the first The combined coefficients after reconstruction, verification, and correction for each recursive time step, dimension, and Similarly, the corrected combination coefficient is used as the first... The starting point for recursive calculation in each recursive time step. For the first The verification hydrodynamic pressure field obtained by local transient calculation at each recursive time step has a dimension equal to the total number of mesh nodes on the wet surface multiplied by the sum of the amplitude and phase components of the pulsating pressure stored in each node. The basis function system is formed by arranging the first few order basis vectors extracted in step 2 in columns. The number of rows is equal to the total number of wet surface grid nodes multiplied by the sum of the amplitude and phase components of the pulsating pressure stored in each node. The number of columns is equal to the number of basis vectors in the basis function system. Indicates the first At each recursive time step, the reconstructed pressure field is obtained by linear combination of the combination coefficients of the current recursive time step and the basis function system. The dimension is the same as that of the verified hydrodynamic pressure field. Indicates the first At each recursive time step, the pressure deviation between the verified hydrodynamic pressure field and the reconstructed pressure field is checked, with the same dimension as the verified hydrodynamic pressure field. The pre-calibrated deviation correction mapping matrix has a dimension equal to the number of basis vectors in the basis function system multiplied by the dimension of the verification hydrodynamic pressure field. The function of this matrix is ​​to map the pressure deviation into the correction increment of the combination coefficients. The specific data for the recursive time step and posterior combination coefficients are shown in Table 1.

[0022] Table 1. Data Statistics Table In this data analysis, the wave load state recursion and reconstruction verification process for 30 recursive time steps was evaluated. The data table includes the prior combination coefficients, reconstructed pressure field values, verified hydrodynamic pressure field values, pressure deviations, and posterior combination coefficients for each recursive time step. The purpose of these data is to reflect the mathematical relationship of the combination coefficients being corrected by verification feedback during the recursion process. Through data analysis, we observed that the sign of the pressure deviation determines the correction direction of the posterior combination coefficient: when the pressure deviation is positive (i.e., the verification pressure field is higher than the reconstructed pressure field), the posterior combination coefficient is adjusted upward relative to the prior combination coefficient; when the pressure deviation is negative (i.e., the verification pressure field is lower than the reconstructed pressure field), the posterior combination coefficient is adjusted downward. For example, the pressure deviation at recursive time step 7 is 0.263, and the prior combination coefficient of 1.109 is adjusted upward to 1.193, with a correction of +0.084; the pressure deviation at recursive time step 10 is -0.201, and the prior combination coefficient of 0.801 is adjusted downward to 0.754, with a correction of -0.047. Further observation reveals a positive correlation between the correction magnitude and the absolute value of the pressure deviation: the correction magnitudes corresponding to recursive time steps with larger absolute values ​​of pressure deviation (such as 0.263 in step 7, 0.224 in step 3, and 0.213 in step 13) are also larger (0.084, 0.076, and 0.073, respectively); while the correction magnitudes corresponding to recursive time steps with smaller absolute values ​​of pressure deviation (such as 0.118 in step 5 and 0.141 in step 19) are also smaller (0.042 and 0.050, respectively). This indicates that the linear correction relationship based on the deviation correction mapping matrix can effectively convert the pressure deviation into a reasonable correction amount for the combined coefficients. In summary, the posterior combination coefficients correct the prior combination coefficients under the drive of pressure deviation, thereby suppressing the cumulative error of the combination coefficients during the recursive process. This verifies the physical logic of correcting the combination coefficients based on pressure deviation in the reconstruction verification stage.

[0023] The triggering condition for reconstruction verification is: ,in, The preset interval number of steps, This represents the remainder operation, when... Divide by A reconstruction check is triggered when the remainder is zero; The pre-calibration method for the deviation correction mapping relationship is as follows: In the design load dataset, the pressure deviation and corresponding combination coefficient error between the verification hydrodynamic pressure field and the reconstructed pressure field under each regular wave condition are statistically analyzed. The mapping relationship between the pressure deviation and the combination coefficient error is established by linear regression, and matrix M is obtained. This calibration process is completed offline before the recursion begins, and M remains unchanged during the recursion process. The state recursion calculation formula for subsequent recursion time steps is still executed according to the original state transition matrix and control matrix until the next reconstructed verification is triggered. The process of converting the combination coefficients into load component time history and performing statistical and extreme value envelope processing using the linear mapping relationship model specifically includes: during the recursive execution process, each recursive time step calls the linear mapping relationship model to process the combination coefficients of the current recursive time step, and converts them into the vertical bending moment, horizontal bending moment and torque of the current recursive time step; The specific operation of calling the linear mapping relationship model is as follows: take the combination coefficients of the current recursive time step as the input vector, and perform matrix multiplication with the linear transformation matrix stored in the linear mapping relationship model. The resulting three-dimensional output vector is the vertical bending moment value, horizontal bending moment value and torque value of the current recursive time step. For the recursive time step that triggers the reconstruction verification, after the combination coefficient correction is completed, the linear mapping relationship model is called to convert the combination coefficients used to the corrected combination coefficients. The vertical bending moment, horizontal bending moment and torque obtained from each recursive time step are recorded sequentially according to the time series of irregular wave and sea state to form the time history of vertical bending moment load component, horizontal bending moment load component and torque load component. A time series refers to the sequential order of recursive time steps from the initial recursive time step to the final recursive time step. The sequence of vertical bending moment values ​​arranged in chronological order is taken as the time history of the vertical bending moment load component, the sequence of horizontal bending moment values ​​arranged in chronological order is taken as the time history of the horizontal bending moment load component, and the sequence of torque values ​​arranged in chronological order is taken as the time history of the torque load component. The time history of the load component is a continuous recorded curve of the load component value changing with time, and each data point in it corresponds to a recursive time step. Zero-crossing analysis was performed on the time histories of vertical bending moment load components, horizontal bending moment load components, and torque load components, respectively. Load cycles that cross the mean were extracted from the time histories of each load component, and the amplitude of each load cycle was recorded. Based on the statistical results of the amplitude of each load cycle, the maximum value of each load component under the irregular wave sea state was calculated. Zero-crossing analysis is a statistical method for identifying events that cross the mean between adjacent data points in a time-series curve. The mean is the arithmetic mean of all data point values ​​in the time-series curve. The execution process of zero-crossing analysis is as follows: scan the load component values ​​of adjacent recursive time steps one by one along the time series direction. When the load component value of the k-th recursive time step and the load component value of the (k+1)-th recursive time step are located on both sides of the mean, and the load component value of the k-th recursive time step is less than the mean and the load component value of the (k+1)-th recursive time step is greater than the mean, record the interval from the k-th recursive time step to the next recursive time step that crosses the mean in the same direction as a complete load cycle. Half of the difference between the maximum and minimum values ​​of the load components in the load cycle is taken as the amplitude of the load cycle. Record the amplitudes of all identified load cycles in the extraction order to form a load cycle amplitude sequence. The maximum value is calculated as follows: sort the load cycle amplitude sequence from largest to smallest, and take the first value after sorting as the maximum value of the load component under the irregular wave state; or fit the load cycle amplitude sequence into an extreme value probability distribution model, and calculate the maximum value based on the duration of the irregular wave state and the probability of extreme value occurrence. The process of generating a wave load parameter table for ship structural design specifically includes: comparing the maximum vertical bending moment, maximum horizontal bending moment, and maximum torque calculated under the current irregular wave state with the historical extreme value envelope of the corresponding load components under the sea state already evaluated for the ship type; if the current maximum value exceeds the range of the historical extreme value envelope, then the extreme value envelope of the ship type is updated, and a wave load parameter table containing the design values ​​of vertical bending moment, horizontal bending moment, and torque at the check section is output. The historical extreme value envelopes of the corresponding load components under the evaluated sea states for this ship type are stored as a data file. The historical extreme value envelope refers to the set of maximum values ​​formed by taking the envelopes of the maximum vertical bending moment, maximum horizontal bending moment, and maximum torque calculated for this ship type under various evaluated irregular wave sea states. The envelope operation means that for each load component, the maximum values ​​under all evaluated sea states are compared one by one, and only the maximum value under each sea state is retained as the envelope value of the load component. The comparison operation involves comparing the maximum vertical bending moment with the vertical bending moment envelope value in the historical extreme value envelope, comparing the maximum horizontal bending moment with the horizontal bending moment envelope value in the historical extreme value envelope, and comparing the maximum torque with the torque envelope value in the historical extreme value envelope. If the current maximum value exceeds the range of the historical extreme value envelope, that is, the current maximum value is greater than the envelope value of the corresponding load component in the historical extreme value envelope, then the extreme value envelope line of this ship type is updated, and the envelope value of the corresponding load component in the historical extreme value envelope is replaced with the current maximum value. The vertical bending moment design value is taken from the vertical bending moment envelope value in the updated extreme value envelope, the horizontal bending moment design value is taken from the horizontal bending moment envelope value in the updated extreme value envelope, and the torque design value is taken from the torque envelope value in the updated extreme value envelope. The wave load parameter table serves as the load boundary condition input item for the hull beam component size verification calculation, and can be directly called by ship structure designers.

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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 of prediction of linear wave loads in a ship structure design, c h a r a c t e r i s e d in that, The specific steps include: Step 1: Obtain the data parameters of the target ship. Based on the design speed, wave height and heading angle range, and using optimized Latin hypercube sampling, generate regular wave conditions. Use the three-dimensional potential flow algorithm to solve the hydrodynamic pressure field on the wetted surface of the hull and the load components of the verification profile under each regular wave condition to form the design load dataset. Step 2: Perform intrinsic orthogonal decomposition on the hydrodynamic pressure field, extract the first few modes that satisfy the energy proportion threshold as the basis function system, project to obtain the combination coefficients, establish a linear mapping relationship model between the combination coefficients and the load components, and generate a load mapping file; Step 3: Input the sea state parameters of the irregular wave sea state to be evaluated, call the load mapping file, interpolate the combination coefficients of adjacent regular wave conditions according to the sea state parameters, determine the state transition matrix and control matrix, calculate the main wave interference force of the hull based on the JONSWAP wave energy spectrum model and take into account the phase compensation of local beam variation, and establish the state recursive calculation formula. Step 4: The combination coefficients are progressively recursively calculated with a fixed recursive time step. When the number of recursive steps reaches the preset interval determined by the ratio of the spectral peak period to the recursive step size, the reconstruction verification is triggered. The three-dimensional potential flow solver is called to solve the verification pressure field at full order and the pressure deviation is compared with the reconstructed pressure field. Based on the pre-calibrated deviation correction mapping relationship, the correction amount of the combination coefficients is obtained and the recursion continues. The linear mapping relationship model is used to convert the load component time history and generate a wave load parameter table after statistical and extreme value envelope processing.

2. The method for predicting linear wave loads in ship structural design according to claim 1, characterized in that: The process of generating the design load dataset specifically includes: obtaining the data parameters of the target ship, which include the hull body shape point coordinate data of the target ship retrieved from the ship design database, as well as the weight and longitudinal position data of the center of gravity of each station distributed along the ship's length. The shape point coordinate data is used to define the geometric boundary of the wetted surface of the hull, and the weight and longitudinal position data of the center of gravity are used to determine the inertial force distribution characteristics of the hull beams. Within a multidimensional parameter space defined by the lower and upper limits of design speed, wave height, and heading angle, optimized Latin hypercube sampling is used to generate a preset number of regular wave conditions. For each regular wave condition, a hydrodynamic solver based on three-dimensional potential flow theory is used to solve the wetted surface mesh of the hull, obtaining the amplitude and phase distribution of pulsating pressure at each wetted surface mesh node. The amplitude and phase distribution of pulsating pressure are then used as the hydrodynamic pressure field corresponding to that regular wave condition. The pressure distribution is integrated along the ship's length, and the vertical bending moment, horizontal bending moment, and torque at a pre-specified check profile are extracted as load components. The regular wave condition is associated with and stored with the corresponding hydrodynamic pressure field and load components to form a design load dataset.

3. The method for predicting linear wave loads in ship structural design according to claim 1, characterized in that: The process of performing intrinsic orthogonal decomposition on the hydrodynamic pressure field, extracting the basis function system, and establishing a linear mapping relationship model specifically includes: arranging the hydrodynamic pressure fields under all regular wave conditions in the design load dataset into a high-dimensional sample matrix by columns; performing intrinsic orthogonal decomposition on the high-dimensional sample matrix to obtain a series of mutually orthogonal basis vectors; calculating the energy contribution rate of each basis vector in reconstructing the entire hydrodynamic pressure field in the high-dimensional sample matrix; accumulating each basis vector in descending order of energy contribution rate; and determining the basis vectors participating in the accumulation as the basis function system when the accumulated value exceeds a preset energy percentage threshold. For each hydrodynamic pressure field in the design load dataset, it is projected onto the basis function system to obtain a set of combination coefficients that correspond one-to-one with each hydrodynamic pressure field. At the same time, the load components corresponding to each hydrodynamic pressure field are extracted, and a correspondence is established between the combination coefficients of each hydrodynamic pressure field and the corresponding load components. Based on this correspondence, a linear mapping relationship model is established using the least squares method. The combination coefficients are then converted into load components through this linear mapping relationship model.

4. The method for predicting linear wave loads in ship structural design according to claim 1, characterized in that: The process of generating and calling the load mapping file specifically includes: packaging and storing the distribution data of each basis vector in the basis function system on the wet surface grid nodes, the model parameters of the linear mapping relationship model, and the boundary parameters of the regular wave condition in the design load dataset into a binary format data file, and using this data file as the load mapping file; Input the sea state parameters of the irregular wave sea state to be evaluated, including speed, significant wave height, spectral peak period, and heading angle. Based on the input speed, retrieve and load the corresponding load mapping file. Using the combination coefficients as intermediate variables, use the speed, significant wave height, spectral peak period, and heading angle of the irregular wave sea state to be evaluated as search conditions. Perform multidimensional linear interpolation between the combination coefficients of adjacent regular wave working conditions according to the sea state parameters to determine the state transition matrix and control matrix. By processing the meaningful wave height and spectral peak period included in the sea state parameters using the JONSWAP wave energy spectrum model, the time history data of wave rise is obtained. Based on the time history data of wave rise, heading angle, and wetted surface geometry of the hull, and combined with the wave phase difference caused by local changes in the wetted surface beam of the hull for phase compensation, the wave principal disturbance force varying with time is calculated. Based on the state transition matrix, control matrix, and wave principal disturbance force, a state recursive calculation formula with combination coefficients as intermediate variables is established. This state recursive calculation formula expresses the recursive relationship between the combination coefficients of the next recursive time step, the combination coefficients of the current recursive time step, and the wave principal disturbance force of the current recursive time step.

5. The method for predicting linear wave loads in ship structural design according to claim 4, characterized in that: The process of determining the state transition matrix and control matrix through multidimensional interpolation specifically includes: In the regular wave condition parameter space covered by the design load dataset, the spatial distance between the sea state parameters of the irregular wave sea state to be evaluated and each regular wave condition parameter is calculated. Several groups of regular wave conditions with the smallest spatial distance are identified as adjacent regular wave conditions. The combination coefficients corresponding to the adjacent regular wave conditions are extracted, as well as the time series change rate data of the combination coefficients of the adjacent regular wave conditions as a function of the number of recursion steps in their respective recursion processes. Using the normalized distance between each component of the sea state parameters of the irregular wave sea state to be evaluated and the adjacent regular wave condition parameters as weights, the time series change rate data of the adjacent regular wave conditions are weighted and averaged to obtain the state transition matrix corresponding to the irregular wave sea state to be evaluated. Using the normalized distance between each component of the sea state parameter of the irregular wave sea state to be evaluated and the parameters of the adjacent regular wave working conditions as weights, the combination coefficients of the adjacent regular wave working conditions under the action of the main wave disturbance force at unit wave height are weighted and averaged to obtain the control matrix corresponding to the irregular wave sea state to be evaluated.

6. The method for predicting linear wave loads in ship structural design according to claim 1, characterized in that: The process of progressively executing state recursion along the time sequence of irregular wave sea states to update the combination coefficients and trigger reconstruction verification specifically includes: using a fixed recursion time step, processing the combination coefficients and the main wave disturbance force of the current recursion time step through the state recursion calculation formula step by step to obtain the combination coefficients of the next recursion time step; during the recursion execution process, determining whether the number of recursion steps executed since the start of the recursion has reached the preset interval step number, and triggering reconstruction verification when the preset interval step number is reached; During the reconstruction and verification, the three-dimensional potential flow solver is invoked to perform a full-order solution with the instantaneous wavefront rise and the instantaneous attitude of the hull at the current recursive time step as boundary conditions to obtain the verification hydrodynamic pressure field. At the same time, the combination coefficients of the current recursive time step are projected onto the basis function system to reconstruct the reconstructed pressure field of the current recursive time step. The pressure deviation between the verification hydrodynamic pressure field and the reconstructed pressure field is calculated. The pressure deviation is converted into a combination coefficient correction amount through a pre-calibrated deviation correction mapping relationship. The combination coefficient correction amount is added to the combination coefficients of the current recursive time step to obtain the corrected combination coefficients. The corrected combination coefficients are used as the starting point for the next recursive time step to continue the recursion. The pre-calibration method for the deviation correction mapping relationship is as follows: statistically analyze the pressure deviation and corresponding combination coefficient error between the hydrodynamic pressure field and the reconstructed pressure field under each regular wave condition in the design load dataset, and establish the mapping relationship between the pressure deviation and the combination coefficient error.

7. The method for predicting linear wave loads in ship structural design according to claim 1, characterized in that: The process of converting the combination coefficients into load component time history and performing statistical and extreme value envelope processing using the linear mapping relationship model specifically includes: during the recursive execution process, each recursive time step calls the linear mapping relationship model to process the combination coefficients of the current recursive time step, and converts them into the vertical bending moment, horizontal bending moment and torque of the current recursive time step; The vertical bending moment, horizontal bending moment and torque obtained from each recursive time step are recorded sequentially according to the time series of irregular wave and sea state to form the time history of vertical bending moment load component, horizontal bending moment load component and torque load component. Zero-crossing analysis was performed on the time histories of vertical bending moment load components, horizontal bending moment load components, and torque load components, respectively. Load cycles that cross the mean were extracted from the time histories of each load component, and the amplitude of each load cycle was recorded. Based on the statistical results of the amplitude of each load cycle, the maximum value of each load component under the irregular wave sea state was calculated.

8. The method for predicting linear wave loads in ship structural design according to claim 7, characterized in that: The process of generating a wave load parameter table for ship structural design specifically includes: comparing the maximum vertical bending moment, maximum horizontal bending moment, and maximum torque calculated under the current irregular wave state with the historical extreme value envelope of the corresponding load components under the sea state already evaluated for the ship type; if the current maximum value exceeds the range of the historical extreme value envelope, then the extreme value envelope of the ship type is updated, and a wave load parameter table containing the design values ​​of vertical bending moment, horizontal bending moment, and torque at the check section is output.

Images