Method and system for calculating the strength of a support strut structure of a floating wind turbine foundation

Abstract

The application provides a kind of floating wind turbine foundation strut structure strength calculation method and system, it is related to marine engineering structural analysis technical field, the application obtains the geometric size of floating wind turbine foundation strut, establishes parameterized finite element model, material attribute is regarded as random variable, calibrates key cross section stress concentration coefficient, data acquisition is carried out in key cross section deployment sensor, and the probability distribution model of strain extreme value is established by extreme value statistics;Limit state function is constructed with yield strength as random variable, the strain extreme value model is combined with stress concentration coefficient, the maximum stress is expressed as the product of elastic modulus, strain extreme value and stress concentration coefficient, the reliability index is calculated by using first order second moment method, the reliability index is compared with the preset target value, whether the strength is qualified is judged, and the reliability index and determination conclusion are output, to solve the problem that traditional method cannot evaluate the real strength of in-service structure by using monitoring data, and realize quantitative evaluation based on reliability.

Description

Technical Field

[0001] This invention relates to the field of marine engineering structural analysis technology, specifically to a method and system for calculating the strength of a floating wind turbine foundation strut structure. Background Technology

[0002] Floating offshore wind turbines are core equipment for developing deep-sea wind energy resources. Their basic structure typically adopts a four-column semi-submersible platform, with each column connected to the others by struts to form an integrated load-bearing system. As key components for transmitting wave loads and maintaining the overall rigidity of the platform, the struts bear alternating loads caused by environmental factors such as wind, waves, and currents over a long period of time. Their structural strength is directly related to the safe operation of the entire machine. In the design process of floating wind turbine foundations, accurate strength assessment of the struts is an important step in ensuring the safe service of the platform throughout its entire life cycle, and it is also the core basis for determining the geometric dimensions and material selection of the struts.

[0003] Current methods for assessing the structural strength of floating wind turbine foundation struts are mainly based on deterministic design principles. Existing technologies typically first select several characteristic working conditions, calculate extreme wave loads using the Morrison equation, then establish a beam system or solid model of the strut in finite element software, apply gravity, buoyancy, and wave forces for static analysis, compare the calculated paradigm equivalent stress with the material yield strength, and use the safety factor as the criterion. Some improved methods introduce weak spring boundary conditions to simulate the elastic support of the mooring system, or use linear wave theory to generate random wave time histories for transient analysis, in order to more accurately simulate real sea conditions.

[0004] Existing technologies still have significant shortcomings in practical applications. On the one hand, the wave loads used in traditional methods are calculated based on theoretical assumptions or extreme working conditions recommended by specifications, which differ from the actual loads borne by the existing wind turbine foundations in the actual service environment. Due to the inherent randomness and uncertainty of the marine environment, the load conditions calculated by theory cannot fully cover the actual stress state, resulting in a disconnect between the assessment input and the actual stress state, and failing to accurately reflect the stress level of the strut under real sea conditions. On the other hand, existing methods treat material properties as deterministic parameters, while actual engineering materials, especially yield strength and elastic modulus, have unavoidable statistical variability. Ignoring this variability makes it impossible for the strength assessment results to quantify the failure probability of the structure, and can only provide a deterministic safety factor judgment, making it difficult to scientifically characterize the safety margin of the strut structure under material parameter fluctuations.

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

[0006] The purpose of this invention is to provide a method and system for calculating the strength of a floating wind turbine foundation strut structure, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A method for calculating the strength of a floating wind turbine foundation strut structure, comprising the following steps: Step 1: Obtain the geometric dimensions of the key cross-sections of the foundation struts of the floating wind turbine under test, and establish a parametric finite element model. Treat the material properties of the foundation struts as random variables, including the elastic modulus and yield strength. Determine the probability distribution type and distribution parameters of each random variable. Perform static analysis using the parametric finite element model and calibrate the stress concentration factor at each key cross-section. Step 2: Collect strain time history data at key cross-sectional locations, perform extreme value statistics on the collected strain time history data, extract strain peak samples, and fit a probability distribution model of strain extreme values. Step 3: Establish the limit state function of the foundation strut based on the theory of mechanics of materials. Express the maximum stress in the limit state function as the product of the elastic modulus, the extreme value of strain, and the stress concentration factor. The elastic modulus follows a probability distribution type, and the extreme value of strain follows a probability distribution model. Use the first second moment method to calculate the reliability index of the foundation strut under the geometric parameters. Step 4: Compare the reliability index with the preset target reliability index. If the reliability index is higher than the target reliability index, the strength of the foundation strut structure is determined to meet the requirements. Otherwise, the strength of the foundation strut structure is determined to be insufficient. Output the reliability index value and strength judgment result as the evaluation conclusion.

[0008] Furthermore, the geometric dimensional parameters include the outer diameter and wall thickness of the basic strut, and the material property parameters include a first random variable and a second random variable. The first random variable is the elastic modulus, and the second random variable is the yield strength. The probability distribution types and corresponding distribution parameters, namely the mean and standard deviation, of the first and second random variables are determined according to the material statistics.

[0009] Furthermore, the key cross-sectional locations of the foundation strut include the connection node between the foundation strut and the column, the mid-span of the strut, and the stress concentration areas identified in the preliminary analysis during the design phase. A strain sensor is arranged at each of the above key cross-sectional locations. The sampling frequency of the strain sensor is not less than 20 Hz, and the acquisition period of the strain time history data covers the complete seasonal changes of sea state for no less than one year, so as to obtain strain time history data that can fully reflect the statistical law of the strut force under actual service environment. Extreme value statistics are performed on the collected strain time history data to extract strain peak samples and fit a probability distribution model of strain extreme values. Specifically, this includes: dividing the acquisition period into several time windows; preprocessing the strain time history data to remove invalid data segments; extracting the maximum strain value at each key cross-section position within each time window as strain peak samples; and collecting all strain peak samples to form a strain peak sample set. Using extreme value statistics theory and based on the Gumbel distribution assumption, the strain peak sample set is used as input data, and the location and scale parameters of the distribution are determined by the maximum likelihood estimation method. This yields a complete mathematical description of the probability of different values ​​of strain extreme values ​​throughout the acquisition period, which is the probability distribution model of strain extreme values. The strain extreme value refers to the third random variable obtained after extreme value statistics processing, which represents the maximum strain of the basic strut during the acquisition period. The mean and standard deviation of this third random variable are used as characteristic parameters in the probability distribution model of strain extreme values.

[0010] Furthermore, based on the theory of mechanics of materials, the limit state function of the basic strut is established, specifically including: based on the geometric dimension parameters, the section with the largest calculated bending normal stress among all key section positions is identified through finite element analysis, and this section is determined as the critical section; using the maximum stress at the critical section as the failure criterion, a limit state function is constructed with the difference between the second random variable and the maximum stress as the expression, where the maximum stress is the stress response value generated by the critical section under the action of the third random variable, and this stress response value is the load effect; The maximum stress in the limit state function is expressed as the product of the elastic modulus, the extreme value of strain, and the stress concentration factor. Specifically, this involves combining the probability distribution model of the third random variable with the stress concentration factor obtained by calibration through a parametric finite element model, based on Hooke's law and the cross-sectional geometric relationship in mechanics of materials. Simultaneously, considering the randomness of the first random variable, the maximum stress is expressed as the product of the first random variable, the third random variable, and the stress concentration factor. The first random variable follows a probability distribution type, the third random variable follows a probability distribution model, and the stress concentration factor is a deterministic coefficient, representing the ratio of the actual stress to the nominal stress at the critical section.

[0011] Furthermore, the reliability index of the limit state function under the current geometric dimensions is calculated using the first second-moment method. Specifically, this includes: expanding the limit state function into a Taylor series at the mean points of the first, second, and third random variables and retaining it down to the first-order term, where the mean point refers to the coordinate point corresponding to the mean of each random variable; based on the mean and standard deviation of each random variable, and the first-order partial derivatives of the limit state function with respect to each random variable at the mean points, calculating the mean and standard deviation of the limit state function respectively, and taking the ratio of the mean to the standard deviation as the reliability index characterizing the structural safety level under the current geometric dimensions.

[0012] Furthermore, the specific methods for determining the mean and standard deviation of the limit state function are as follows: The mean of the limit state function is obtained by multiplying the mean of the first random variable, the mean of the third random variable, and the stress concentration coefficient to obtain a product. Then, the difference between the mean of the second random variable and this product is calculated, and the difference is the mean of the limit state function. The standard deviation of the limit state function is synthesized by the standard deviations of the second, first, and third random variables according to the error propagation theory. The standard deviations of the first and third random variables are multiplied by their corresponding sensitivity coefficients during synthesis. The sensitivity coefficients refer to the degree of influence of the unit change of each random variable on the value of the limit state function. Specifically, the sensitivity coefficient corresponding to the first random variable is the product of the mean of the third random variable and the stress concentration coefficient, and the sensitivity coefficient corresponding to the third random variable is the product of the mean of the first random variable and the stress concentration coefficient.

[0013] Furthermore, the reliability index is compared with a preset target reliability index, specifically including: the preset target reliability index refers to a threshold pre-set according to the structural safety level of the floating wind turbine foundation strut; when the calculated reliability index is greater than or equal to the target reliability index, the structural strength of the foundation strut is determined to meet the requirements, and an evaluation conclusion of qualified strength and a reliability index value are output; when the calculated reliability index is less than the target reliability index, the structural strength of the foundation strut is determined to be insufficient, and an evaluation conclusion of insufficient strength and a reliability index value are output.

[0014] The present invention also provides a calculation system for the structural strength of a floating wind turbine foundation strut, the system being used to execute the above-described method for calculating the structural strength of a floating wind turbine foundation strut, comprising: The parameter setting module is used to obtain the geometric dimension parameters at the key section positions of the foundation support rod of the floating wind turbine under test, and to establish a parametric finite element model. The material property parameters of the foundation support rod are used as random variables, including the elastic modulus and yield strength. The probability distribution type and distribution parameters of each random variable are determined. Static analysis is performed by the parametric finite element model to calibrate the stress concentration factor at each key section position. The model building module is used to collect strain time history data at key cross-sectional locations, perform extreme value statistics on the collected strain time history data, extract strain peak samples, and fit a probability distribution model of strain extreme values. The index calculation module is used to establish the limit state function of the foundation strut based on the theory of mechanics of materials. The maximum stress in the limit state function is expressed as the product of the elastic modulus, the extreme value of strain and the stress concentration factor. The elastic modulus follows a probability distribution type and the extreme value of strain follows a probability distribution model. The reliability index of the foundation strut under the geometric parameters is calculated using the first second moment method. The conclusion judgment module is used to compare the reliability index with the preset target reliability index. If the reliability index is higher than the target reliability index, it is determined that the strength of the foundation strut structure meets the requirements; otherwise, it is determined that the strength of the foundation strut structure is insufficient. The reliability index value and strength judgment result are output as the evaluation conclusion.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention captures the true stress state of a floating wind turbine foundation under actual service conditions in the form of strain peak samples by deploying strain sensors at key cross-sections of the strut and collecting data for at least one year. Then, it establishes a probability distribution model of strain extreme values ​​using extreme value statistics theory. This technical feature transforms the load input of the existing structure from deterministic values ​​based on theoretical assumptions or specifications to a statistical description based on measured data. This fundamentally solves the problem of the disconnect between load conditions and actual service environment in traditional strength assessment methods and realizes a quantitative characterization of the inherent randomness and uncertainty of the marine environment. This invention also introduces the probability distribution characteristics of material properties, treating yield strength and elastic modulus as random variables following a specific distribution type. Combined with the probability distribution model of strain extrema, it realizes the bidirectional randomness characterization of load effect and material resistance in the limit state function. On this basis, the reliability index of the strut structure under the current geometric parameters is calculated by the first second moment method. This index is compared with the preset target value, and whether the reliability index reaches the preset threshold is used as the criterion for judging whether the strength meets the requirements. This improves the strength assessment conclusion from the traditional qualitative judgment of safety factor to a quantitative evaluation based on failure probability, significantly enhancing the scientificity and reliability of the assessment results. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall method flow of the present invention; Figure 2 The theoretical value of the reliability index of this invention minus the average value of strain extreme values ​​(10) -6 )picture; Figure 3 The average extreme strain value of this invention (10) -6 - Reliability index observation value fitting curve; Figure 4 This is a flowchart of the overall system modules of the present invention. Detailed Implementation

[0017] 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.

[0018] 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.

[0019] Example: Please see Figures 1-3 The present invention provides a technical solution: A method for calculating the strength of a floating wind turbine foundation strut structure, comprising the following steps: Step 1: Obtain the geometric dimensions of the key cross-sections of the foundation struts of the floating wind turbine under test, and establish a parametric finite element model. Treat the material properties of the foundation struts as random variables, including the elastic modulus and yield strength. Determine the probability distribution type and distribution parameters of each random variable. Perform static analysis using the parametric finite element model and calibrate the stress concentration factor at each key cross-section. In a specific embodiment of the present invention, considering that the floating wind turbine foundation struts are subjected to complex wave loads and environmental loads during actual service, their structural strength depends not only on the geometric dimensions, but also on the randomness of material properties and stress concentration effects. Therefore, accurately obtaining the actual geometric parameters of the struts, reasonably describing the uncertainty of material properties, and accurately calibrating the stress concentration degree of key parts are the basis for subsequent reliability assessment. The geometric dimension parameters include the outer diameter and wall thickness of the basic strut. The material property parameters are random variables, including a first random variable and a second random variable. The first random variable is the elastic modulus and the second random variable is the yield strength. The probability distribution types that the first random variable and the second random variable follow, as well as the corresponding distribution parameters, namely the mean and standard deviation, are determined according to the material statistics. Material property parameters include elastic modulus and yield strength, which are defined as the first random variable (elastic modulus) and the second random variable (yield strength), respectively. They are considered random variables rather than deterministic constants because the properties of actual engineering materials have inherent statistical variability; that is, the elastic modulus and yield strength of steel produced in the same batch will fluctuate within a certain range. Based on the quality certificate provided by the material manufacturer or relevant material performance statistical databases, a sufficient number of sample data are collected, and statistical analysis methods are used to determine the probability distribution type (generally normal or log-normal) that these two random variables follow. The corresponding distribution parameters, namely the mean and standard deviation, are calculated. The mean represents the average level of material performance, while the standard deviation reflects the dispersion of performance. These distribution parameters will be used to describe the uncertainty of material performance in subsequent reliability calculations. The distribution type is determined by using the probability paper graph method or the KS test (Kolmogorov-Smirnov test) to test the goodness of fit of the candidate distributions. The distribution type that passes the significance level test and has the highest goodness of fit is selected as the probability distribution of the random variable. The distribution parameters (mean and standard deviation) are calculated from the sample data using the maximum likelihood estimation method or the method of moments. The stress concentration factor is a dimensionless parameter reflecting the degree of stress amplification at geometric discontinuities. It is defined as the ratio of the actual stress at the critical section to the nominal stress calculated based on simple beam theory. It should be noted that the stress concentration factor depends only on the geometry of the component and is independent of material properties and load magnitude; therefore, it can be considered a deterministic factor. The specific calibration process is as follows: On the established parametric finite element model, apply a unit load that is easy to calculate, such as a unit axial force or a unit bending moment. The load direction should be consistent with the main force direction of the strut in actual service. Run static analysis to extract the actual stress at each critical section location. Calculate the nominal stress of the same section under the same unit load according to the material mechanics formulas. The calculation of the nominal stress needs to be based on the obtained outer diameter and wall thickness to solve for the geometric properties of the section. Divide the actual stress by the nominal stress to obtain the stress concentration factor at the critical section location. Once this stress concentration factor is calibrated, it can be used as a constant in subsequent calculations to convert measured strain into actual stress.

[0020] Step 2: Collect strain time history data at key cross-sectional locations, perform extreme value statistics on the collected strain time history data, extract strain peak samples, and fit a probability distribution model of strain extreme values. In a specific embodiment of the present invention, the wave loads borne by the floating wind turbine foundation struts in the actual service environment have significant randomness and seasonal variation characteristics. Their stress state cannot be completely determined by theoretical calculation. Therefore, it is necessary to obtain the real strain response data of the struts through long-term field measurements, and use extreme value statistical theory to extract statistical laws that can characterize the extreme load effect from these data, so as to provide load input for subsequent reliability assessment. The key cross-sectional locations of the basic strut include the connection node between the basic strut and the column, the mid-span of the strut, and the stress concentration areas identified in the preliminary analysis during the design phase. A strain sensor is arranged at each of the above key cross-sectional locations. The sampling frequency of the strain sensor is not less than 20 Hz, and the acquisition period of the strain time history data covers the complete seasonal changes of sea state for no less than one year, so as to obtain strain time history data that can fully reflect the statistical law of the strut force under actual service environment. This location is completely consistent with the critical section location in step 1, specifically including the node area connecting the strut and the column, the mid-span of the strut, and other stress concentration areas identified by the finite element analysis during the preliminary design phase. The reason for placing sensors at each critical section location is that the stress response characteristics differ at different locations. Only by comprehensively monitoring all dangerous areas can we ensure that the subsequent assessment covers all weak points of the structure. At each of the above critical section locations, at least one fiber optic strain sensor should be placed. The selection of the fiber optic strain sensor must meet the durability and stability requirements for long-term outdoor monitoring. Its range should cover the maximum strain range that the foundation strut will experience. The sampling frequency of the sensor should be set to no less than 20 Hz. This frequency can effectively capture the dynamic response of the strut under wave loads and avoid the loss of peak data due to insufficient sampling rate. The acquisition period of strain time history data must cover the complete sea state seasonal changes for no less than one year, that is, at least one complete annual cycle, to fully reflect the statistical law of the strut stress under different seasons and different wind and wave conditions. The reason for requiring a long-term acquisition period of more than one year is that the marine environment has significant seasonal characteristics, and short-term monitoring cannot represent the long-term service conditions of the structure. To extract statistically significant peak samples from continuous strain time-history data, the entire acquisition period needs to be divided into multiple continuous and non-overlapping time windows. The rule for determining the length of the time window is as follows: based on the autocorrelation analysis of the strain time-history data, the autocorrelation function of the strain time-history data is calculated, and the lag time corresponding to the first decay of the autocorrelation function value to below 0.1 is taken as the minimum time window length to ensure that the strain peak samples extracted within each window are statistically independent. When autocorrelation analysis is lacking, a simplified rule based on the sea state characteristics of the sea area is used, that is, the time window length is set to 3 hours according to the decorrelation time scale of the meaningful wave height of the waves in the sea area. The principle of time window division is that the strain data within each window should be statistically independent, and the number of windows should be sufficient to ensure the sample size of the peak samples. The raw strain time history data within each time window are preprocessed to remove invalid data segments caused by sensor failure, signal transmission interruption, or environmental electromagnetic interference. The preprocessing criteria include: data exceeding the sensor's measurement range, data remaining constant for a long period (which may indicate sensor failure), and data showing obvious abnormal jumps. After preprocessing, valid strain time history data is retained within each time window. Specifically, data exhibiting a constant fluctuation range of less than 0.5% of the full scale for a period exceeding 10% of the time window length is considered sensor failure. Data exhibiting abnormal jumps where the amplitude change between adjacent sampling points exceeds 15% of the full scale and subsequent data does not recover to the normal fluctuation range is considered electromagnetic interference or signal transmission failure. Extreme value statistics are performed on the collected strain time history data to extract strain peak samples and fit a probability distribution model of strain extreme values. Specifically, this includes: dividing the acquisition period into several time windows; preprocessing the strain time history data to remove invalid data segments; extracting the maximum strain value at each key cross-section position within each time window as strain peak samples; and collecting all strain peak samples to form a strain peak sample set. Using extreme value statistics theory and based on the Gumbel distribution assumption, the strain peak sample set is used as input data, and the location and scale parameters of the distribution are determined by the maximum likelihood estimation method. This yields a complete mathematical description of the probability of different values ​​of strain extreme values ​​throughout the acquisition period. This complete mathematical description is the probability distribution model of strain extreme values. The strain extreme value refers to the third random variable obtained after extreme value statistics processing, representing the maximum strain of the basic strut during the acquisition period. The mean and standard deviation of this third random variable are used as characteristic parameters in the probability distribution model of strain extreme values. For each time window, the maximum strain value within that window is extracted at each key section location. The reason for extracting the maximum value instead of the average value or other statistics is that structural strength assessment focuses on the response under extreme loads, i.e. the maximum stress that occurs throughout the entire service life. The maximum strain value is a direct representation of this extreme response. The maximum strain values ​​extracted from all time windows and all key section locations are aggregated to form a strain peak sample set containing a sufficient number of samples. This strain peak sample set represents the overall distribution of extreme strain events that the strut will encounter in the actual service environment. Extreme value statistics is a specialized discipline that studies the distribution of maximum or minimum values ​​in random processes. It is applicable to inferring the probability of extreme events from a large amount of observation data. In this invention, the strain peak sample set is regarded as a set of independent and identically distributed extreme value observations. It is fitted based on the Gumbel distribution assumption. The Gumbel distribution is one of the commonly used distribution types in extreme value statistics, and it is particularly suitable for describing the asymptotic distribution of maximum value samples. The specific fitting process is as follows: Using the strain peak sample set as input data, the location and scale parameters of the Gumbel distribution are solved using the maximum likelihood estimation method. Maximum likelihood estimation is a classic parameter estimation method; its core idea is to find the parameter value that maximizes the probability of the current sample occurring. The location parameter calculated by this method reflects the central trend of the strain extreme value, while the scale parameter reflects its dispersion. Based on these two parameters, the mathematical relationship between the probabilities of different strain extreme values ​​throughout the entire service life can be fully described. This complete mathematical description is the probability distribution model of the strain extreme value. It should be noted that the strain extreme value is defined here as the third random variable. This random variable refers to the statistical quantity obtained after extreme value statistical processing, used to characterize the maximum strain occurring in the foundation strut during the complete acquisition period. Unlike the first random variable (elastic modulus) and the second random variable (yield strength) defined in step 1, the third random variable is not obtained based on material manufacturer statistics but is obtained through extreme value statistical fitting based on field measured data. The probability distribution model of the strain extreme value uses the mean and standard deviation of this third random variable as its characteristic parameters. These two parameters will be used as input data for subsequent reliability index calculations.

[0021] Step 3: Establish the limit state function of the foundation strut based on the theory of mechanics of materials. Express the maximum stress in the limit state function as the product of the elastic modulus, the extreme value of strain, and the stress concentration factor. The elastic modulus follows a probability distribution type, and the extreme value of strain follows a probability distribution model. Use the first second moment method to calculate the reliability index of the foundation strut under the geometric parameters. The core objective of step three is to construct a limit state function that describes the safety state of the foundation strut structure under test, based on the theory of mechanics of materials and probabilistic reliability methods. The material random variable determined in step one and the strain extreme value determined in step two are then introduced into this function. The reliability index is calculated using the first second-order moment method, thereby quantitatively assessing the structural safety of the strut under the current geometric parameters. Step three specifically includes three key steps: first, establishing the limit state function and determining the critical section; second, expressing the maximum stress as a function of random variables; and third, calculating the reliability index using the first second-order moment method. The limit state function of the foundation strut is established based on the theory of mechanics of materials. Specifically, it includes: based on the geometric dimension parameters, identifying the section with the largest calculated bending normal stress among all critical sections through finite element analysis, and determining this section as the critical section; using the maximum stress at the critical section as the failure criterion, constructing a limit state function with the difference between the second random variable and the maximum stress as the expression, where the maximum stress is the stress response value generated by the critical section under the action of the third random variable, and this stress response value is the load effect; In structural reliability analysis, the limit state function is a mathematical expression that describes the boundary between the safe state and the failure state of a structure. For the floating wind turbine foundation strut of this invention, bending stress is the main failure mode, so it is necessary to construct the limit state function based on bending normal stress. The critical section is determined from the key section locations identified in the first step. The critical section refers to the control section of the entire strut that experiences the most unfavorable stress and is most likely to fail. The specific method for determining the critical section is as follows: Based on the parametric finite element model established in the first step, which has already input the geometric parameters of the foundation strut to be tested, simulated loads that reflect actual service conditions are applied to the model, such as a combination of axial tensile and compressive loads and bending loads caused by waves. After performing static analysis, the calculated bending normal stress values ​​at each key section location identified in the first step are extracted. These key section locations include the strut-column connection node, the mid-span of the strut, and other stress concentration areas identified in the preliminary analysis during the design phase. The calculated bending normal stress values ​​at all key section locations are compared, and the section with the largest calculated value is taken as the critical section. This section will be used for subsequent analysis. The control section for reliability calculation is the part with the highest stress level and smallest safety margin in the entire strut. The simulated load that reflects the actual service conditions refers to the extreme wave load combination (including axial force and bending moment components) calculated according to the design sea conditions of the sea area where the floating wind turbine foundation is located and in accordance with relevant specifications. The magnitude and direction of this load combination represent the most unfavorable stress state that the strut will encounter during its service life. In step 1, a unit load is applied to calibrate the stress concentration factor K of each section. In step 3, an extreme design load is applied to compare the stress response of each section to determine the critical section. The two serve different analytical purposes and are both linear elastic analyses. The relative stress magnitude of the sections does not change due to the change in load amplitude. Therefore, the section with the largest K value calibrated in step 1 is the section with the largest stress response in step 3, i.e., the critical section. After identifying the critical section, a limit state function is constructed. The general form of the limit state function is the structural resistance minus the load effect. In this invention, the structural resistance is the material's yield strength, as yielding is the marker of steel entering a plastic state and is generally used as a criterion for strength failure. The load effect is the maximum stress generated at the critical section under external load. Therefore, the limit state function g is expressed as: g equals the yield strength minus the maximum stress at the critical section. It is important to clarify that in this expression, the yield strength is a random variable defined in the first step, representing the inherent dispersion of the material's resistance to plastic deformation. The maximum stress is the stress response value generated at the critical section under extreme strain, representing the actual stress level caused by the external load. This stress response value will be further expressed as a function of random variables in subsequent steps. The physical meaning of the limit state function g is: when g is greater than zero, it indicates that the structural resistance is greater than the load effect, and the structure is in a safe state; when g is equal to zero, it indicates that the structural resistance is equal to the load effect, and the structure is in a limit state; when g is less than zero, it indicates that the structural resistance is less than the load effect, and the structure fails. The maximum stress in the limit state function is expressed as the product of the elastic modulus, the extreme value of strain, and the stress concentration factor. Specifically, this involves: combining the probability distribution model of the third random variable with the stress concentration factor obtained by calibration through the parametric finite element model, based on Hooke's law and the cross-sectional geometric relationship in mechanics of materials; and considering the randomness of the first random variable, so that the maximum stress is expressed as the product of the first random variable, the third random variable, and the stress concentration factor. The first random variable is a random variable that follows the probability distribution type determined in step 1, and the third random variable is a random variable that follows the probability distribution model determined in step 2. The stress concentration factor is a deterministic factor, which is the ratio of the actual stress to the nominal stress at the critical section obtained by static analysis based on geometric dimension parameters using the parametric finite element model. The purpose of expressing the maximum stress in the limit state function as a function of random variables is to organically combine the strain extreme random variable obtained in the second step with the material random variable and stress concentration factor obtained in the first step, thus fully revealing the random transmission path from external load to structural response. According to Hooke's law in mechanics of materials, within the elastic range, the stress and strain of a material are directly proportional, and the proportionality constant is the elastic modulus. This relationship can be expressed as stress equals elastic modulus multiplied by strain. However, for actual engineering structures, due to abrupt changes in geometry, such as at joints and cross-sectional changes, the actual stress in local areas may be higher than the nominal stress calculated based on simple beam theory. This stress increase effect needs to be quantified using stress concentration factor. In this invention, the stress concentration factor is obtained through calibration of the parameterized finite element model established in the first step. The specific calibration process is as follows: a finite element model is established based on the actual geometric dimensions of the foundation strut to be tested, a simulated load is applied for static analysis, and the actual stress at the critical section is extracted; simultaneously, under the same load conditions, the nominal stress of the section is calculated according to the formulas of mechanics of materials; the ratio of the actual stress to the nominal stress is the stress concentration factor of the critical section. Therefore, the stress concentration factor is defined as the ratio of the actual stress at the critical section to the nominal stress calculated based on the formulas of mechanics of materials. It is a dimensionless deterministic coefficient that reflects the degree of stress amplification caused by geometrical abrupt changes. Its value is obtained through finite element analysis calibration and is regarded as a fixed value in this evaluation. Combining the two relationships above, the maximum stress at the critical section can be expressed as the product of three quantities: elastic modulus multiplied by strain extremum and then multiplied by stress concentration factor. In this product expression, the properties of each quantity need to be clearly defined. The elastic modulus is a random variable defined in the first step, which follows the probability distribution type determined in the first step based on material statistics, such as a normal distribution or a log-normal distribution, and has a definite mean and standard deviation. The strain extremum is a random variable defined in the second step, which follows the strain extremum probability distribution model established in the second step, and also has a definite mean and standard deviation. Through this expression, the originally deterministic calculation of the maximum stress is transformed into a function of random variables. The strain extremum random variable reflects the randomness of marine environmental loads, the elastic modulus random variable reflects the inherent dispersion of material properties, and the stress concentration factor reflects the deterministic influence of structural geometric characteristics. The product of these three quantities completely describes the entire process from external load input to local stress response output of the structure. The random variable expression for the maximum stress is: ,in The maximum stress at the critical section is a random variable representing the maximum stress response value that the foundation strut under test will exhibit under actual service conditions. Its function is to serve as a quantitative characterization of the load effect in the limit state function. This parameter is not directly taken, but is obtained by synthesizing the three quantities on the right side of the formula. The elastic modulus, defined in the first step, is the first random variable representing the material's ability to resist elastic deformation. Its physical meaning is the ratio of stress to strain within the elastic range. The range, probability distribution type, mean, and standard deviation of this parameter are determined based on statistical data provided by the material manufacturer. Specifically, it is determined according to the quality assurance certificate of the same batch of materials or an industry database. It follows a normal or log-normal distribution, and its mean is obtained. and standard deviation ; The strain extremum, defined in the second step as the third random variable, represents the maximum strain value that the strut will experience during its service life. Its physical meaning is the structural response strength caused by external marine environmental loads. Since this strain extremum is statistically derived from real strain data directly collected by sensors deployed at the critical section, its value objectively includes the local strain amplification effect caused by geometrical abrupt changes. The range, probability distribution type, mean, and standard deviation of this parameter are all obtained through extreme value statistical fitting in the second step. Specifically, based on long-term field measurement data, a Gumbel distribution is used for fitting, and the location and scale parameters of the distribution are determined using the maximum likelihood estimation method, thereby calculating the mean. and standard deviation ; The stress concentration factor is a deterministic coefficient obtained in the first step of finite element analysis calibration. It represents the amplification factor of the local stress increase caused by geometrical abrupt changes. Its physical meaning is the ratio of the actual stress to the nominal stress at the critical section. This parameter is determined as follows: a finite element model is established based on the actual geometric dimensions of the strut under test; a unit load is applied for static analysis; the actual stress at the critical section is extracted; and the nominal stress is calculated using the formulas of mechanics of materials. The actual stress is then divided by the nominal stress to obtain the stress concentration factor. , It is a value greater than 1, generally between 1.2 and 3.0, with the specific value depending on the severity of the geometric change, such as at a typical strut-column connection. The K-value will be between 1.5 and 2.0. Before the assessment begins, the K-value will be compared with the values ​​of each calibrated key section to identify the section with the largest K-value as the critical section. Strain sensors will be deployed at this section to ensure the accuracy of the collected strain data. It can accurately reflect the most unfavorable stress state of the structure; The reliability index of the limit state function under the current geometric parameters is calculated using the first second-moment method. Specifically, this includes: expanding the limit state function into a Taylor series at the mean points of the first, second, and third random variables and retaining it down to the first term, where the mean point refers to the coordinate point corresponding to the mean of each random variable; calculating the mean and standard deviation of the limit state function based on the mean and standard deviation of each random variable, and the first-order partial derivatives of the limit state function with respect to each random variable at the mean points; and taking the ratio of the mean to the standard deviation as the reliability index characterizing the structural safety level under the current geometric parameters. After obtaining a complete description of the limit state function and its internal random variables, it is necessary to calculate a reliability index that can quantitatively characterize the structural safety level. This invention uses the first second-moment method for solution. The basic idea of ​​this method is to linearize the limit state function at the mean points of each random variable, and use the mean and variance information of the random variables to approximate the statistical characteristics of the limit state function. It is necessary to clarify the three random variables in the limit state function and their statistical parameters. These three random variables are: the first random variable, the second random variable, and the third random variable, and their mean (denoted as...) , , ) and standard deviation (denoted as , , The mean point is the coordinate point corresponding to the mean value of each random variable, i.e., a point in space. (This has been determined by the first and second steps respectively.) , , The reason for choosing the mean point as the linearization expansion point is that the mean point is the region where the random variable is most likely to occur, and linearization at this point can obtain better approximation accuracy. The limit state function g is expanded using a Taylor series at the mean point. Taylor series expansion is a method of approximating a function using polynomials. Here, only the first-order terms (linear terms) are retained, ignoring higher-order terms (second order and above). The advantage of this approach is its computational simplicity and sufficient engineering accuracy when the coefficient of variation of the random variable is small. Based on the results of the Taylor expansion, approximate expressions for the mean and standard deviation of the limit state function g can be derived. The mean of the limit state function g... The calculation method is as follows: The average safety margin of a structure is equal to the average resistance (mean yield strength) minus the average load effect (mean elastic modulus multiplied by the mean strain extreme value and the stress concentration factor). The product of the mean elastic modulus and the mean strain extreme value represents the average nominal stress, and multiplying it by the stress concentration factor gives the average true stress. The specific methods for determining the mean and standard deviation of the limit state function are as follows: The mean of the limit state function is obtained by multiplying the mean of the first random variable, the mean of the third random variable, and the stress concentration coefficient to obtain a product. Then, the difference between the mean of the second random variable and this product is calculated, and the difference is the mean of the limit state function. The standard deviation of the limit state function is synthesized by the standard deviations of the second, first, and third random variables according to the error propagation theory. The standard deviations of the first and third random variables are multiplied by their corresponding sensitivity coefficients during synthesis. The sensitivity coefficients refer to the degree of influence of the unit change of each random variable on the value of the limit state function. Specifically, the sensitivity coefficient corresponding to the first random variable is the product of the mean of the third random variable and the stress concentration coefficient, and the sensitivity coefficient corresponding to the third random variable is the product of the mean of the first random variable and the stress concentration coefficient. The mean expression for the limit state function is: This formula is a component of the one-step second-moment method for calculating reliability indices. It is used to solve for the mean of the limit state function g. Its function is to calculate the average safety margin of the structure using the mean of each random variable as a representative value. The mean of the limit state function represents the magnitude of the average safety margin of the structure, which is the difference between the average resistance and the average load effect. The larger the value, the more sufficient the average safety reserve of the structure. The mean yield strength is the second random variable determined in the first step. The mean value represents the average level of a material's ability to resist plastic deformation. Its value comes from material statistics. For example, for Q355 steel, the mean yield strength is between 355 MPa and 400 MPa, depending on the statistical data in the material quality certificate. The mean value of the elastic modulus is the first random variable determined in the first step. The mean value, whose source and range are explained in Formula 1; The mean of the strain extrema is the third random variable determined in the second step. The mean value, derived from the statistical fitting results of extreme values, represents the average level of the maximum strain that will occur during the service life; due to The strain extreme values ​​are obtained through extreme value statistics from the real strain data directly collected by sensors deployed at the critical section. These values ​​objectively include the local strain amplification effect caused by geometric abrupt changes. This is the true maximum stress at the critical section, and there is no need to multiply it by the stress concentration factor. The mathematical structure of this formula embodies the basic idea of ​​reliability analysis: the average safety margin of a structure equals the average resistance ( Subtract the average load effect ( The average load effect is composed of the product of the material's average stiffness, the average strain extreme value, and the stress concentration factor. It reflects the expected value of the complete transmission path from external load to internal stress. The mean value of the strain extreme value comes from the field measured data at the critical section, which can truly reflect the stress level of the structure in the actual service environment. The standard deviation expression for the limit state function is: This formula is another core component of the one-step second-moment method for calculating reliability indices. It is used to solve for the standard deviation of the limit state function g. Its function is based on error propagation theory, comprehensively considering the impact of the discreteness of each random variable on the uncertainty of structural safety margin. The derivation of this formula is based on Taylor series expansion and the assumption of independent random variables, and is a standard processing method in probabilistic reliability analysis. The standard deviation of the limit state function represents the degree of uncertainty of the structural safety margin, that is, the fluctuation range of the safety margin due to the dispersion of material properties and the randomness of load. The smaller the value, the more stable the prediction of the safety margin. The standard deviation of the yield strength is the second random variable determined in the first step. The standard deviation represents the degree of dispersion of yield strength. Its value comes from material statistics and is generally calculated based on the coefficient of variation. For example, if the mean yield strength is 355 MPa and the coefficient of variation is 0.05, then the standard deviation is 17.75 MPa. The standard deviation of the elastic modulus is the first random variable determined in the first step. The standard deviation represents the degree of dispersion of the elastic modulus. Its value is also derived from material statistics. For example, if the mean elastic modulus is 206,000 MPa and the coefficient of variation is 0.03, then the standard deviation is 6,180 MPa. The standard deviation of the strain extrema is the third random variable determined in the second step. The standard deviation represents the degree of dispersion of strain extrema. Its value is derived from the statistical fitting results of the extrema. For example, if the mean of the strain extrema is 300 microstrains and the standard deviation is 50 microstrains, then... That is, 50 microstrain; In the formula, It is the elastic modulus The sensitivity coefficient represents Each unit change affects the magnitude of the change in the limit state function by influencing the maximum stress; similarly... It is the extreme value of strain The introduction of these two sensitivity coefficients allows the discreteness of different random variables to be weighted and synthesized according to their actual impact on the final result, which embodies the core idea of ​​error propagation theory. Standard deviation of the limit state function g The calculation requires synthesis according to error propagation theory. The basic principle of error propagation theory is: when a function is composed of multiple random variables, the variance of the function (the square of the standard deviation) is equal to the sum of the squares of the variances of each random variable multiplied by the sensitivity coefficient of that variable's influence on the function, plus the covariance term between the random variables. In the limit state function of this invention, since the three random variables (elastic modulus, yield strength, and strain extrema) originate from different physical processes, it can be reasonably assumed that they are independent of each other, therefore the covariance term is zero. The specific synthesis method is: the variance of the limit state function g... The square equals the variance of the yield strength. Square, plus the variance of the elastic modulus Sensitivity coefficient of squared elastic modulus The square of the value, plus the variance of the extreme strain values. Sensitivity coefficient of squared strain extrema The square of, where the sensitivity coefficient and It is determined by the value of the first-order partial derivative of the limit state function with respect to each random variable at the mean point; After calculation, the expression for the limit state function g in this invention is: ,right Taking the partial derivative gives The value at the mean point is Therefore, the sensitivity coefficient of the elastic modulus Equal to the mean value of strain extrema The physical meaning of this coefficient is: when the elastic modulus increases by one unit, the value of the limit state function will decrease. The unit is used because a larger elastic modulus will produce greater stress at the same strain, thus reducing the safety margin; for Taking the partial derivative gives The value at the mean point is Therefore, the sensitivity coefficient of strain extrema Equal to the mean elastic modulus The physical meaning of this coefficient is: when the strain extremum increases by one unit, the value of the limiting state function will decrease. The value is given by one unit because larger strain also generates larger stress; the sensitivity coefficient of the yield strength is 1 because the partial derivative of the limit state function with respect to the yield strength is equal to 1, which means that for every unit increase in yield strength, the value of the limit state function directly increases by one unit; the mean value of the limit state function is obtained. and standard deviation Next, reliability metrics Defined as: equal Divide by The physical meaning of this ratio is: the ratio of the structure's average safety margin (mean) to its uncertainty (standard deviation), a reliability index. It is a dimensionless numerical value that comprehensively reflects the overall impact of material randomness and load randomness on structural safety. The larger the value, the higher the safety margin relative to uncertainty, and the safer the structure. The smaller the value, the lower the proportion of safety margin to uncertainty, and the greater the possibility of structural failure; The reliability index expression is: This formula is the final expression for calculating the reliability index using the one-step second-moment method. It compares the average safety margin of the structure with its uncertainty to obtain a dimensionless reliability index. ;in As a reliability index, this parameter is the core result finally calculated in this invention. It represents the structural safety level of the strut under test under the current geometric parameters. Its physical meaning is the average safety margin of the structure. ) relative to its uncertainty ( Multiples of ) The larger the value, the higher the safety margin relative to uncertainty, and the safer the structure. The smaller the value, the lower the safety margin relative to uncertainty, and the greater the possibility of structural failure. In engineering practice... The value is generally between 2.0 and 5.0, depending on the safety level and design requirements of the structure; Specific data on some sample parameters and reliability indicators are shown in Table 1.

[0022] Table 1 Data Statistics Table By observing the data in the table, the second column, the average extreme strain value, represents the average level of maximum strain that the basic struts will experience during long-term monitoring, obtained through extreme value statistics. This parameter is a quantitative representation of the randomness of marine environmental loads; a larger value means a higher level of extreme loads borne by the struts. It should be noted that although the table does not have a separate column for stress concentration factors, the sensors used to collect strain data are all deployed at the critical sections calibrated by finite element analysis. Therefore, the average extreme strain value objectively includes the local stress concentration effect caused by geometric abrupt changes, and can truly reflect the stress level of the structure at the most unfavorable section. There is no need to multiply by an additional stress concentration factor in the reliability calculation. The third column, the theoretical value of the reliability index, is a theoretical result rigorously calculated according to the core mathematical formula of this invention. It integrates the randomness of material properties and the randomness of loads, and quantitatively characterizes the safety level of the strut under test under the current parameters. A larger value indicates a more sufficient safety margin for the structure, while a smaller value indicates a higher safety risk.

[0023] By observing the data in the first fifteen rows, a clear negative correlation can be found between the reliability index and the mean value of the extreme strain values. When the mean value of the extreme strain values ​​is at a low level, the theoretical value of the reliability index is generally high; while when the mean value of the extreme strain values ​​increases, the theoretical value of the reliability index decreases accordingly. This relationship reflects the basic principle of structural safety, that is, as the level of external load increases, the safety margin of the structure will inevitably decrease. For example, the average extreme strain value of sample 1 is 612.3 microstrain, which is relatively low, and its corresponding theoretical reliability index reaches 4.832. In contrast, the average extreme strain value of sample 5 is 891.2 microstrain, which is significantly higher, and its theoretical reliability index drops to 2.236, forming a stark contrast. This inverse relationship persists throughout the data series, reflecting the dominant influence of load level on structural safety. Even when the average extreme strain values ​​are similar, there are still reasonable fluctuations in the theoretical reliability index values ​​among different samples. This is mainly due to the random differences in the distribution parameters of elastic modulus and yield strength when each sample is statistically fitted to the extreme values. This reflects the superimposed influence of the randomness of material properties on the reliability level. This fluctuation characteristic conforms to the basic laws of probabilistic reliability analysis. The fluctuation of observed values ​​relative to theoretical values ​​can be analyzed using the residual column. In the first fifteen rows of data, the residuals can be positive or negative, with a value range of approximately -0.057 to 0.105. For example, the observed value of sample number 2 is 3.752, slightly higher than its theoretical value of 3.647, with a residual of 0.105; while the observed value of sample number 5 is 2.179, slightly lower than its theoretical value of 2.236, with a residual of -0.057. This fluctuation range is moderate and can truly reflect the random biases that may occur in actual monitoring. From the overall trend of the sample sequence, the observed values ​​of the reliability index fluctuate around the theoretical values ​​without any systematic deviation. For example, the theoretical value of sample number 1 is 4.832, and the observed value is 4.805, which are very close. The theoretical value of sample number 6 is 3.992, and the observed value is 4.055, which also maintains good consistency. This distribution characteristic makes it possible to restore the true relationship through fitting methods. In addition, observing the growth order of the sample numbers reveals that the data is not monotonically arranged according to a certain parameter, but exhibits random distribution characteristics. For example, the average strain extreme value of sample number 2 is 745.8 microstrain, while the average strain extreme value of sample number 3 is higher but the number is larger. This randomness avoids autocorrelation between data and is more consistent with the real situation of random occurrence of working conditions in actual sampling. The data from the first fifteen rows show that the changes in reliability index are mainly determined by the mean of strain extreme values, and the two are negatively correlated. That is, an increase in the mean of strain extreme values ​​will lead to a decrease in reliability index. At the same time, the inherent randomness of material properties (elastic modulus, yield strength) acts as a superposition factor, which makes the reliability index of samples with similar strain levels reasonably dispersed. The random fluctuations of the observed values ​​simulate the uncertainty in engineering practice, making the dataset both consistent with theoretical logic and have practical significance.

[0024] Step 4: Compare the reliability index with the preset target reliability index. If the reliability index is higher than the target reliability index, the strength of the foundation strut structure is determined to meet the requirements. Otherwise, the strength of the foundation strut structure is determined to be insufficient. Output the reliability index value and strength judgment result as the evaluation conclusion. The reliability index calculated in step 3 is compared with the preset target reliability index. Based on the comparison results, a clear judgment is made on the structural strength of the floating wind turbine foundation support rod under test, and an evaluation conclusion is output to provide a quantitative basis for engineering operation and maintenance decisions. Step 4 specifically includes two key links: first, the determination of the preset target reliability index, and second, the comparison of reliability indexes and strength judgment. The reliability index is compared with a preset target reliability index, specifically including: the preset target reliability index refers to a threshold pre-set according to the structural safety level of the floating wind turbine foundation strut; when the calculated reliability index is greater than or equal to the target reliability index, the structural strength of the foundation strut is determined to meet the requirements, and an evaluation conclusion of qualified strength and a reliability index value are output; when the calculated reliability index is less than the target reliability index, the structural strength of the foundation strut is determined to be insufficient, and an evaluation conclusion of insufficient strength and a reliability index value are output. The target reliability index is a pre-set numerical threshold before reliability assessment. Its purpose is to provide a standard for measuring reliability and to determine whether the current reliability level of the structure meets acceptable safety requirements. This threshold is not arbitrarily selected, but needs to be scientifically determined based on the structural safety level of the floating wind turbine foundation struts and relevant industry standards. The classification of structural safety levels mainly includes factors such as the degree of risk of personal injury and death, the magnitude of economic losses, the scope of environmental impact, and the impact on the overall operation of the offshore wind farm caused by the failure of the structure. For floating wind turbine foundations, the strut structure is a critical load-bearing component. Once it fails, it will lead to serious consequences such as wind turbine overturning, power generation interruption, and marine environmental pollution. Therefore, it is generally classified as a medium safety level or a high safety level. Different safety levels correspond to different target reliability requirements. The higher the safety level, the higher the required target reliability index. When determining the specific value of the target reliability index, it is necessary to refer to the current marine engineering structural design codes or reliability design standards. For example, ISO 2394 "General principles of structural reliability" published by the International Organization for Standardization, DNV-OS-C101 "Code for Design of Steel Structures" published by Det Norske Veritas, and relevant marine engineering codes in my country all provide suggested ranges of target reliability index values ​​for different structural types and safety levels. The suggested values ​​in these codes are based on a large amount of engineering practice, failure statistical analysis, and probabilistic reliability theory research results, and have wide industry recognition and applicability. In this invention, the specific value of the target reliability index is a fixed threshold pre-set in the design stage or assessment preparation stage, based on the requirements of the above-mentioned codes and combined with factors such as the actual structural characteristics of the floating wind turbine foundation strut, the design service life, and the severity of failure consequences. Once this threshold is set, it remains unchanged throughout the entire assessment process and serves as a benchmark for subsequent comparisons. It is important to clarify that the physical meaning of the target reliability index is: in order to meet the requirements for safe operation of the structure throughout its entire life cycle, the reliability index of the structure must reach a minimum level. If the calculated reliability index is higher than this level, it means that under the current material properties and load conditions, the structure has sufficient reliability and the probability of failure is lower than the acceptable level; if the calculated reliability index is lower than this level, it means that the reliability of the structure is insufficient, the probability of failure is higher than the acceptable level, and there is a safety risk that cannot be ignored. The reliability index calculated in step 3 is marked as... The preset target reliability index is marked as When comparing the size of two values, there are two scenarios; the first scenario is... Greater than or equal to If the calculated reliability index is not lower than the target reliability index, this indicates that, under the current geometric parameters, the reliability level of the foundation strut under test has reached the minimum standard of acceptable safety requirements. The physical meaning is that, after comprehensively considering the randomness of the material's elastic modulus and yield strength, the randomness of strain extrema caused by marine environmental loads, and the influence of structural geometric characteristics, the average safety margin of the strut relative to the uncertainty is sufficiently high, and the failure probability is controlled within an acceptable range. In this case, the structural strength of the foundation strut under test is deemed to meet the requirements. When this occurs, the output evaluation conclusion includes two parts: the first part is the qualitative judgment result, i.e., the conclusion that the strength is qualified; the second part is the quantitative support data, i.e., the calculated reliability index. The specific value can be used as a quantitative representation of the safety status of the strut structure for subsequent operation and maintenance records and comparative analysis. The second scenario is... Less than If the calculated reliability index is lower than the target reliability index, it indicates that, under the current geometric parameters, the reliability level of the foundation strut under test has failed to meet the minimum acceptable safety requirements. Physically, this means that after considering all random factors, the average safety margin of the strut relative to uncertainty is low, the failure probability exceeds the acceptable range, and the structure faces a significant safety risk. In this case, the structural strength of the foundation strut under test is deemed insufficient. When this occurs, the output assessment conclusion also includes two parts: the first part is the qualitative judgment result, i.e., the conclusion of insufficient strength; the second part is the quantitative support data, i.e., the calculated reliability index. The specific value indicates not only the existence of structural risks but also quantitatively reflects the degree of risk. That is, the lower the reliability index, the greater the risk and the more urgent the countermeasures required. The value is determined based on the structural safety level and relevant specifications. For example, for offshore wind power structures with a medium safety level, The value is set to 3.0 to 3.7; for higher safety levels, it is set to 3.7 to 4.2. This criterion transforms the probabilistic reliability analysis results into a clear basis for engineering decisions.

[0025] Please see Figure 4 The present invention also provides a calculation system for the structural strength of a floating wind turbine foundation strut, the system being used to execute the above-described method for calculating the structural strength of a floating wind turbine foundation strut, comprising: The parameter setting module is used to obtain the geometric dimension parameters at the key section positions of the foundation support rod of the floating wind turbine under test, and to establish a parametric finite element model. The material property parameters of the foundation support rod are used as random variables, including the elastic modulus and yield strength. The probability distribution type and distribution parameters of each random variable are determined. Static analysis is performed by the parametric finite element model to calibrate the stress concentration factor at each key section position. The model building module is used to collect strain time history data at key cross-sectional locations, perform extreme value statistics on the collected strain time history data, extract strain peak samples, and fit a probability distribution model of strain extreme values. The index calculation module is used to establish the limit state function of the foundation strut based on the theory of mechanics of materials. The maximum stress in the limit state function is expressed as the product of the elastic modulus, the extreme value of strain and the stress concentration factor. The elastic modulus follows a probability distribution type and the extreme value of strain follows a probability distribution model. The reliability index of the foundation strut under the geometric parameters is calculated using the first second moment method. The conclusion judgment module is used to compare the reliability index with the preset target reliability index. If the reliability index is higher than the target reliability index, it is determined that the strength of the foundation strut structure meets the requirements; otherwise, it is determined that the strength of the foundation strut structure is insufficient. The reliability index value and strength judgment result are output as the evaluation conclusion.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A method for calculating the strength of a floating wind turbine foundation strut structure, characterized in that, The specific steps include: Step 1: Obtain the geometric dimensions of the key cross-sections of the foundation struts of the floating wind turbine under test, and establish a parametric finite element model. Treat the material properties of the foundation struts as random variables, including the elastic modulus and yield strength. Determine the probability distribution type and distribution parameters of each random variable. Perform static analysis using the parametric finite element model and calibrate the stress concentration factor at each key cross-section. Step 2: Collect strain time history data at key cross-sectional locations, perform extreme value statistics on the collected strain time history data, extract strain peak samples, and fit a probability distribution model of strain extreme values. Step 3: Establish the limit state function of the foundation strut based on the theory of mechanics of materials. Express the maximum stress in the limit state function as the product of the elastic modulus, the extreme value of strain, and the stress concentration factor. The elastic modulus follows a probability distribution type, and the extreme value of strain follows a probability distribution model. Use the first second moment method to calculate the reliability index of the foundation strut under the geometric parameters. Step 4: Compare the reliability index with the preset target reliability index. If the reliability index is higher than the target reliability index, the strength of the foundation strut structure is determined to meet the requirements. Otherwise, the strength of the foundation strut structure is determined to be insufficient. Output the reliability index value and strength judgment result as the evaluation conclusion.

2. The method for calculating the strength of a floating wind turbine foundation strut structure according to claim 1, characterized in that: The geometric dimensional parameters include the outer diameter and wall thickness of the basic strut, and the material property parameters include a first random variable and a second random variable. The first random variable is the elastic modulus, and the second random variable is the yield strength. The probability distribution types that the first random variable and the second random variable follow, as well as the corresponding distribution parameters, namely the mean and standard deviation, are determined according to the material statistics.

3. The method for calculating the strength of a floating wind turbine foundation strut structure according to claim 2, characterized in that: The key cross-sectional locations of the basic strut include the connection node between the basic strut and the column, the mid-span of the strut, and the stress concentration areas identified in the preliminary analysis during the design phase. A strain sensor is arranged at each of the above key cross-sectional locations. The sampling frequency of the strain sensor is not less than 20 Hz, and the acquisition period of the strain time history data covers the complete seasonal changes of sea state for no less than one year, so as to obtain strain time history data that can fully reflect the statistical law of the strut force under actual service environment. Extreme value statistics are performed on the collected strain time history data to extract strain peak samples and fit a probability distribution model of strain extreme values. Specifically, this includes: dividing the acquisition period into several time windows; preprocessing the strain time history data to remove invalid data segments; extracting the maximum strain value at each key cross-section position within each time window as strain peak samples; and collecting all strain peak samples to form a strain peak sample set. Using extreme value statistics theory and based on the Gumbel distribution assumption, the strain peak sample set is used as input data, and the location and scale parameters of the distribution are determined by the maximum likelihood estimation method. This yields a complete mathematical description of the probability of different values ​​of strain extreme values ​​throughout the acquisition period, which is the probability distribution model of strain extreme values. The strain extreme value refers to the third random variable obtained after extreme value statistics processing, which represents the maximum strain of the basic strut during the acquisition period. The mean and standard deviation of this third random variable are used as characteristic parameters in the probability distribution model of strain extreme values.

4. The method for calculating the strength of a floating wind turbine foundation strut structure according to claim 3, characterized in that: The limit state function of the foundation strut is established based on the theory of mechanics of materials. Specifically, it includes: based on the geometric dimension parameters, identifying the section with the largest calculated bending normal stress among all critical sections through finite element analysis, and determining this section as the critical section; using the maximum stress at the critical section as the failure criterion, constructing a limit state function with the difference between the second random variable and the maximum stress as the expression, where the maximum stress is the stress response value generated by the critical section under the action of the third random variable, and this stress response value is the load effect; The maximum stress in the limit state function is expressed as the product of the elastic modulus, the extreme value of strain, and the stress concentration factor. Specifically, this involves combining the probability distribution model of the third random variable with the stress concentration factor obtained by calibration through a parametric finite element model, based on Hooke's law and the cross-sectional geometric relationship in mechanics of materials. Simultaneously, considering the randomness of the first random variable, the maximum stress is expressed as the product of the first random variable, the third random variable, and the stress concentration factor. The first random variable follows a probability distribution type, the third random variable follows a probability distribution model, and the stress concentration factor is a deterministic coefficient, representing the ratio of the actual stress to the nominal stress at the critical section.

5. The method for calculating the strength of a floating wind turbine foundation strut structure according to claim 4, characterized in that: The reliability index of the limit state function under the current geometric parameters is calculated using the first second-moment method. Specifically, this includes: expanding the limit state function into a Taylor series at the mean points of the first, second, and third random variables and retaining the result down to the first term, where the mean point refers to the coordinate point corresponding to the mean of each random variable; calculating the mean and standard deviation of the limit state function based on the mean and standard deviation of each random variable, and the first-order partial derivatives of the limit state function with respect to each random variable at the mean points; and taking the ratio of the mean to the standard deviation as the reliability index characterizing the structural safety level under the current geometric parameters.

6. The method for calculating the strength of a floating wind turbine foundation strut structure according to claim 5, characterized in that: The specific methods for determining the mean and standard deviation of the limit state function are as follows: The mean of the limit state function is obtained by multiplying the mean of the first random variable, the mean of the third random variable, and the stress concentration coefficient to obtain a product. Then, the difference between the mean of the second random variable and this product is calculated, and the difference is the mean of the limit state function. The standard deviation of the limit state function is synthesized by the standard deviations of the second, first, and third random variables according to the error propagation theory. The standard deviations of the first and third random variables are multiplied by their corresponding sensitivity coefficients during synthesis. The sensitivity coefficients refer to the degree of influence of the unit change of each random variable on the value of the limit state function. Specifically, the sensitivity coefficient corresponding to the first random variable is the product of the mean of the third random variable and the stress concentration coefficient, and the sensitivity coefficient corresponding to the third random variable is the product of the mean of the first random variable and the stress concentration coefficient.

7. The method for calculating the strength of a floating wind turbine foundation strut structure according to claim 6, characterized in that: The reliability index is compared with a preset target reliability index, specifically including: the preset target reliability index refers to a threshold pre-set according to the structural safety level of the floating wind turbine foundation strut; when the calculated reliability index is greater than or equal to the target reliability index, the structural strength of the foundation strut is determined to meet the requirements, and an evaluation conclusion of qualified strength and a reliability index value are output; when the calculated reliability index is less than the target reliability index, the structural strength of the foundation strut is determined to be insufficient, and an evaluation conclusion of insufficient strength and a reliability index value are output.

8. A calculation system for the structural strength of a floating wind turbine foundation strut, characterized in that: The system is used to execute a method for calculating the strength of a floating wind turbine foundation strut structure as described in any one of claims 1-7, including: The parameter setting module is used to obtain the geometric dimension parameters at the key section positions of the foundation support rod of the floating wind turbine under test, and to establish a parametric finite element model. The material property parameters of the foundation support rod are used as random variables, including the elastic modulus and yield strength. The probability distribution type and distribution parameters of each random variable are determined. Static analysis is performed by the parametric finite element model to calibrate the stress concentration factor at each key section position. The model building module is used to collect strain time history data at key cross-sectional locations, perform extreme value statistics on the collected strain time history data, extract strain peak samples, and fit a probability distribution model of strain extreme values. The index calculation module is used to establish the limit state function of the foundation strut based on the theory of mechanics of materials. The maximum stress in the limit state function is expressed as the product of the elastic modulus, the extreme value of strain and the stress concentration factor. The elastic modulus follows a probability distribution type and the extreme value of strain follows a probability distribution model. The reliability index of the foundation strut under the geometric parameters is calculated using the first second moment method. The conclusion judgment module is used to compare the reliability index with the preset target reliability index. If the reliability index is higher than the target reliability index, it is determined that the strength of the foundation strut structure meets the requirements; otherwise, it is determined that the strength of the foundation strut structure is insufficient. The reliability index value and strength judgment result are output as the evaluation conclusion.

Images