Model error evaluation method, device and equipment for offshore wind power single pile p-y curve method

Abstract

The application discloses a kind of offshore wind power single pile p-y curve method model error evaluation method, device and equipment, method includes: obtaining the real test pile data of offshore wind power single pile, creates three-dimensional finite element reference model;Respectively through real test pile data and three-dimensional finite element reference model calculation measured horizontal ultimate bearing capacity and first model horizontal ultimate bearing capacity, to calibrate reference model error;Adjust the model parameter of three-dimensional finite element reference model, obtain multiple simulation cases;Respectively through p-y curve method and three-dimensional finite element reference model to simulation case analysis obtains p-y horizontal ultimate bearing capacity and second model horizontal ultimate bearing capacity, to determine curve deviation;Through curve deviation and reference model error evaluation p-y curve model error.The technical scheme provided by the application can improve the accuracy of offshore wind power single pile p-y curve method model error evaluation under the condition of limited pile data.

Description

Technical Field

[0001] This invention relates to the field of marine geotechnical engineering, and specifically to a model error assessment method, apparatus, and equipment for the Py curve method of offshore wind power monopile. Background Technology

[0002] The most commonly used design method for offshore wind turbine monopiles is the Py curve method recommended by the American Petroleum Institute (API). This method assumes the monopile is an Euler-Bernoulli beam and discretizes the soil around the pile as a series of nonlinear stiffness springs. The Py curve describes the relationship between the horizontal displacement of the monopile at different depths in the soil and the soil resistance. The API-recommended Py curve was derived from field tests on slender flexible piles (depth ratio L / D ≥ 20, where L / D is the ratio of the pile's depth L below the mudline to its diameter D) used in offshore oil and gas platforms. However, the semi-rigid monopiles used in offshore wind turbines (depth ratio L / D ≤ ​​10) differ significantly from those used in offshore oil and gas platforms. Under horizontal loads, rigid or semi-rigid piles will rotate around a point below the mudline, and their horizontal displacement at the mudline is mainly caused by the rotation of the pile body; while flexible piles will bend, resulting in multiple inflection points below the mudline, and their horizontal displacement at the mudline is mainly caused by the bending of the pile body. Furthermore, the py curve is difficult to account for the contribution of the vertical shear stress on the pile side to the horizontal bearing capacity. Therefore, in the application of offshore wind power monopiles, it may lead to conservative design parameters and excessive safety redundancy.

[0003] From a reliability design perspective, under the ultimate limit state, the horizontal load at the pile top has a certain probability of exceeding the ultimate horizontal bearing capacity of a single pile (the horizontal load at the pile top corresponding to a horizontal displacement of 0.1 times the pile diameter at the mud surface is considered the ultimate horizontal bearing capacity of a single pile), and this probability is the failure probability. However, there is a certain deviation between the horizontal ultimate bearing capacity of wind turbine single piles calculated by the Py curve recommended by the current API and the measured value. Therefore, in order to accurately assess the failure probability of wind turbine single piles, it is necessary to introduce the model error of the Py curve method into the function of reliability calculation. The model error is defined as the ratio between the measured value of the horizontal ultimate bearing capacity of a single pile and the value calculated by the Py curve method. This measured value is usually determined through pile testing. By introducing the model error, the true failure probability of wind turbine single piles can be accurately assessed, thereby reducing the excessive safety redundancy caused by the Py curve method.

[0004] Regarding research on model errors for monopile piles, in 2005, Phoon et al., in their article "Characterisation of model uncertainties for laterally loaded rigid drilledshafts. Géotechnique" published in the journal *Geótechnique*, introduced a method for determining the model error of a calculation model for the horizontal bearing capacity of small-diameter rigid drilled piles. In 2014, Bian Xiaoya et al., in their article "Reliability Analysis of Serviceability Limit State of Foundation Piles Considering Parameter and Model Uncertainties" published in *Rock Mechanics*, derived calculation formulas for the ultimate limit state and serviceability limit state reliability indices of vertically loaded piles, considering the uncertainties of the calculation model and load. However, the above literature mainly focuses on slender flexible piles (L / D≥20) on land, for which test pile data is generally abundant. For large-diameter monopile foundations in offshore wind power, the pile diameter can reach 8-10m, making test pile costs prohibitively high, and test pile data is usually very scarce. Therefore, the methods mentioned above for studying model errors are not applicable to large-diameter monopile foundations in offshore wind power.

[0005] In 2020, Zhang Haiyang et al. published a paper titled "Correction of Py Curve for Large-Diameter Monopile Foundations at Sea" in the *Journal of Hydraulic Engineering*, proposing a Py curve correction method. This method calculates pile deformation and soil resistance at different depths below the soil using a finite element model and the Py curve, respectively. Using the finite element model's calculation results as the standard, the Py curve calculation formula is adjusted based on the error between the two results. However, this method lacks a clear explanation of how to assess the error in the horizontal ultimate bearing capacity of the pile top. Furthermore, the most accurate data regarding various aspects of large-diameter monopile foundations at sea comes from measured values ​​in pile testing. This paper directly uses the finite element model's calculation results as the standard to measure the error of the Py curve, thus still exhibiting inaccuracies.

[0006] Therefore, there is an urgent need for a method to evaluate the error of the Py curve method model for offshore wind power monopile under limited test pile data conditions. Summary of the Invention

[0007] In view of this, the present invention provides a method, apparatus and equipment for evaluating the model error of the Py curve method for offshore wind power monopile, thereby improving the accuracy of evaluating the model error of the Py curve method for offshore wind power monopile under limited test pile data conditions.

[0008] According to a first aspect, embodiments of the present invention provide a model error assessment method for offshore wind turbine monopile using the Py curve method. The method includes: acquiring actual test pile data for offshore wind turbine monopile, the test pile data including a load-displacement curve, the load-displacement curve being used to characterize the relationship between the horizontal load at the pile top and the horizontal displacement of the pile body at the mud surface; creating a three-dimensional finite element reference model for analyzing the horizontal bearing capacity of the offshore wind turbine monopile; calculating the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity under the same working conditions using the test pile data and the three-dimensional finite element reference model, respectively; and calculating the difference between the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity. The error of the three-dimensional finite element reference model relative to the reference model of the actual test pile data is calibrated. The model parameters of the three-dimensional finite element reference model are adjusted to obtain multiple simulation cases. The same simulation cases are analyzed and calculated sequentially using the py curve method and the three-dimensional finite element reference model to obtain the corresponding py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. Based on the difference between the py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, the curve deviation of the py curve method relative to the three-dimensional finite element reference model is determined. The py curve model error of the py curve method relative to the actual test pile data is evaluated by the curve deviation and the reference model error.

[0009] Optionally, the model parameters include pile parameters and soil constitutive parameters. The pile parameters include the pile depth L, pile diameter D, pile wall thickness t, and loading point height H. The soil constitutive parameters include: effective unit weight γ', initial reference void ratio e0, Poisson's ratio υ, critical state line slope M, compression coefficient λ, expansion coefficient κ, permeability coefficient K, initial overconsolidation parameter R, and overconsolidation control parameter m. R Initial structural parameters R * and structural control parameter m * R .

[0010] Optionally, the steps for calculating the py horizontal ultimate bearing capacity corresponding to each simulation case using the py curve method include: dividing the portion of a single pile from below the mud surface to the pile bottom into a preset number of calculation units, and creating a finite difference equation based on the calculation units and the current simulation case; initializing the horizontal load at the pile top and the horizontal displacement of each node, where the node is the endpoint of the calculation unit; substituting the initialized horizontal displacement of each node into the py curve to obtain the soil resistance of each node; substituting the soil resistance of each node into the finite difference equation to calculate the new horizontal displacement of each node; if the error between the new horizontal displacement of each node and the initialized horizontal displacement of each node is less than a preset threshold, and the new horizontal displacement of the pile body at the mud surface reaches 0.1 times the pile diameter, the initialized horizontal load at the pile top is taken as the py horizontal ultimate bearing capacity of the current simulation case; if the error between the new horizontal displacement of each node and the initialized horizontal displacement of each node is greater than or equal to the preset threshold, or the new horizontal displacement of the pile body at the mud surface does not reach 0.1 times the pile diameter, then returning to the steps of initializing the horizontal load at the pile top and the horizontal displacement of each node, and re-initializing.

[0011] Optionally, the curve deviation is the ratio of the second model's horizontal ultimate bearing capacity to the py horizontal ultimate bearing capacity; the benchmark model error is the ratio of the measured horizontal ultimate bearing capacity to the first model's horizontal ultimate bearing capacity.

[0012] Optionally, the step of evaluating the Py curve model error of the Py curve method relative to the actual test pile data through the curve deviation and the benchmark model error includes: estimating the probability distributions of the benchmark model error and the curve deviation, respectively; determining the first mean and the first coefficient of variation corresponding to the benchmark model error based on the estimated probability distribution of the benchmark model error, and determining the second mean and the second coefficient of variation corresponding to the curve deviation based on the estimated probability distribution of the curve deviation; calculating the third mean corresponding to the Py curve model error by multiplying the first mean and the second mean; calculating the third coefficient of variation corresponding to the Py curve model error by taking the square root of the sum of the squares of the first coefficient of variation and the second coefficient of variation; testing the probability distribution form corresponding to the Py curve model error using the Anderson-Darling test, and determining the Py curve model error by using the third mean, the third coefficient of variation, and the probability distribution form corresponding to the Py curve model error.

[0013] Optionally, the probability distributions of the baseline model error and the curve deviation are estimated to be log-normal distributions.

[0014] Optionally, the finite difference equation includes:

[0015]

[0016]

[0017]

[0018] In the formula, h is the length of the calculation unit; y i Let Mi represent the horizontal displacement of the i-th node; M0 be the bending moment acting on the pile at the mud surface; and V0 be the horizontal shear force acting on the pile at the mud surface, where M0 and V0 are derived from the initial horizontal load at the pile top; p i E represents the soil resistance acting at the i-th node; i I i Let be the bending stiffness of the i-th node, where the 0th node is the node at the mud surface, nodes above the mud surface are negative, and nodes below the mud surface are positive.

[0019] According to a second aspect, embodiments of the present invention provide a model error assessment device for the Py-curve method of offshore wind turbine monopiles. The device includes: a data acquisition module for acquiring actual test pile data of offshore wind turbine monopiles, the test pile data including a load-displacement curve, the load-displacement curve being used to characterize the relationship between the horizontal load at the pile top and the horizontal displacement of the pile body at the mud surface; a model creation module for creating a three-dimensional finite element reference model for analyzing the horizontal bearing capacity of offshore wind turbine monopiles; and a finite element model error calibration module for calculating the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity under the same working conditions using the test pile data and the three-dimensional finite element reference model, respectively, and based on the difference between the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity. The system includes: a calibration module for determining the error of the three-dimensional finite element reference model relative to the actual test pile data; a simulation data generation module for adjusting the model parameters of the three-dimensional finite element reference model to obtain multiple simulation cases; a curve deviation calibration module for sequentially analyzing and calculating the same simulation cases using the py curve method and the three-dimensional finite element reference model to obtain the corresponding py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, and determining the curve deviation of the py curve method relative to the three-dimensional finite element reference model based on the difference between the py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity; and a py curve method error evaluation module for evaluating the py curve model error of the py curve method relative to the actual test pile data through the curve deviation and the reference model error.

[0020] According to a third aspect, embodiments of the present invention provide a model error assessment device for the Py curve method of offshore wind power monopile, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the method described in the first aspect, or any optional embodiment of the first aspect.

[0021] According to a fourth aspect, embodiments of the present invention provide a computer-readable storage medium storing computer instructions for causing a computer to perform the method described in the first aspect, or any optional embodiment of the first aspect.

[0022] The technical solution provided in this application has the following advantages:

[0023] The technical solution provided in this application, given the scarcity of test pile data for offshore wind turbine monopile foundations, proposes a method for calculating the model error of the Py-curve method for evaluating offshore wind turbine monopile foundations, using a three-dimensional finite element reference model as a bridge. First, actual test pile data for offshore wind turbine monopile foundations is obtained, including load-displacement curves. Then, a three-dimensional finite element reference model is created for analyzing the horizontal bearing capacity of offshore wind turbine monopile foundations. Subsequently, the measured horizontal ultimate bearing capacity and the first model's horizontal ultimate bearing capacity under the same working conditions are calculated using the test pile data and the three-dimensional finite element reference model, respectively. The method is then used to calculate the difference between the measured horizontal ultimate bearing capacity and the first model's horizontal ultimate bearing capacity. This study calibrates the error of a 3D finite element (FEM) benchmark model relative to actual test pile data. Then, it adjusts the model parameters of the 3D FEM benchmark model to obtain multiple simulation cases. The same simulation cases are analyzed and calculated sequentially using both the Py curve method and the 3D FEM benchmark model to obtain the corresponding Py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. Based on the difference between the Py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, the curve deviation of the Py curve method relative to the 3D FEM benchmark model is determined. Finally, the Py curve model error of the Py curve method relative to actual test pile data is comprehensively evaluated by considering both the curve deviation and the benchmark model error. This approach compensates for the deficiencies in test pile data, thereby improving the accuracy of the Py curve method's error assessment relative to actual test pile data. This provides assistance for the reliability calculation of large-diameter monopiles in offshore wind power and the development of design methods based on reliability theory. Attached Figure Description

[0024] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings:

[0025] Figure 1 This diagram illustrates the steps of a model error assessment method using the Py curve method for offshore wind power monopile according to one embodiment of the present invention.

[0026] Figure 2 A schematic diagram of a PISA test pile project is shown in one embodiment of the present invention;

[0027] Figure 3A schematic diagram of the structure of a three-dimensional finite element reference model in one embodiment of the present invention is shown;

[0028] Figure 4 A schematic diagram illustrating the probability distribution of baseline model error and curve deviation in one embodiment of the present invention is shown;

[0029] Figure 5 This diagram illustrates the structure of a model error evaluation device for the Py curve method of offshore wind power monopile according to one embodiment of the present invention.

[0030] Figure 6 A schematic diagram of the structure of a model error assessment device for the Py curve method of offshore wind power monopile is shown in one embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please see Figure 1 In one embodiment, a model error evaluation method for the Py curve method of offshore wind power monopile specifically includes the following steps:

[0033] Step S101: Obtain test pile data for offshore wind turbine monopiles. The test pile data includes load-displacement curves, which are used to characterize the relationship between the horizontal load at the top of the pile and the horizontal displacement of the pile body at the mud surface.

[0034] Specifically, this embodiment first collects test pile data for offshore wind turbine monopiles through literature review and other methods. This data mainly consists of load-displacement curves, and also includes some other pile-related data, such as pile deformation. The test pile data is relatively small; the load-displacement curve refers to the relationship between the horizontal load at the pile top loading point and the horizontal displacement at the mud surface of the pile. Taking the PISA (Pile-Soil Analysis) project as an example, such as... Figure 2As shown, a wind turbine monopile can be divided into two parts: the section below the mudline to the pile bottom and the section above the mudline to the pile top flange. During horizontal loading, a jack is typically used to apply a horizontal reaction force to the pile top, while simultaneously measuring the horizontal displacement of the pile body at the mudline. Through this loading and measurement method, the relationship between the horizontal load at the pile top and the horizontal displacement of the pile body at the mudline can be obtained, resulting in a load-displacement curve. Currently, the load-displacement curve obtained from pile testing in domestic and international literature is typically the relationship curve between the horizontal load at the pile top and the horizontal displacement of the pile body at the mudline. Currently, most domestic and international literature considers the horizontal load at the pile top corresponding to a horizontal displacement of the pile body at the mudline reaching 0.1 times the pile diameter as the ultimate horizontal bearing capacity of a monopile. This embodiment of the invention also adopts this standard; therefore, the load-displacement curve can be used to determine the ultimate horizontal bearing capacity of a monopile.

[0035] Step S102: Create a three-dimensional finite element reference model for analyzing the horizontal bearing capacity of offshore wind turbine monopiles.

[0036] Step S103: Calculate the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity under the same working conditions using the actual test pile data and the three-dimensional finite element reference model, respectively. Based on the difference between the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity, calibrate the reference model error of the three-dimensional finite element reference model relative to the actual test pile data.

[0037] Specifically, to compensate for the lack of test pile data, this embodiment of the invention creates a three-dimensional finite element benchmark model for analyzing the horizontal bearing capacity of a single pile in offshore wind power. Simulation experiments are conducted using this three-dimensional finite element benchmark model to obtain simulation data, thereby expanding the data volume. Furthermore, the three-dimensional finite element benchmark model is used to represent the model error of the Py-curve method for bridge analysis relative to the actual test pile data.

[0038] Specific operations include: considering that the horizontal ultimate bearing capacity calculated by the three-dimensional finite element reference model also has errors compared to the actual value, this embodiment of the invention needs to pre-calibrate the reference model error of the three-dimensional finite element reference model relative to the actual test pile data using actual test pile data. This embodiment uses ABAQUS finite element software to establish a three-dimensional finite element reference model for a large-diameter single pile, such as... Figure 3As shown, the reference model can adopt a constitutive model based on the modified Cambridge model with upper and lower load surfaces. This constitutive model, based on the modified Cambridge model, introduces a lower load surface to consider the overconsolidation characteristics of the soil and an upper load surface to consider the structural characteristics of the soil. Therefore, it can accurately simulate the response of natural soil under complex working conditions. This constitutive model can serve as a unified constitutive model for clay and sand. Then, the measured horizontal ultimate bearing capacity and the first model's horizontal ultimate bearing capacity under the same working condition (here, the same working condition refers to the working condition represented by the measured pile data) are calculated using actual test pile data and the three-dimensional finite element reference model, respectively. The difference between the measured horizontal ultimate bearing capacity and the first model's horizontal ultimate bearing capacity is used to obtain the reference model error of the three-dimensional finite element reference model relative to the measured pile data. The difference calculation method can include subtraction, ratio, etc. Considering that the model error of the traditional Py curve method is defined as "the ratio between the measured value of the horizontal ultimate bearing capacity of a single pile and the value calculated by the Py curve method," to facilitate the calculation of the failure probability of a wind turbine single pile, this embodiment of the invention also uses a ratio to reflect the error between the finite element reference model and the measured pile data, as shown in the following formula:

[0039]

[0040] In the formula, P m The measured horizontal ultimate bearing capacity of a single pile is obtained from actual pile test data. It represents the horizontal load corresponding to a horizontal displacement of 0.1D at the mud surface, where D is the pile diameter; P FEM The horizontal ultimate bearing capacity of the first model, ε, is calculated from the three-dimensional finite element reference model. FEM This represents the baseline model error.

[0041] Step S104: Adjust the model parameters of the three-dimensional finite element reference model to obtain multiple simulation cases.

[0042] Specifically, this step is the preparatory stage to compensate for the lack of test pile data. By adjusting the model parameters of the three-dimensional finite element reference model, a large number of simulation cases are obtained. The model parameters include pile parameters and soil constitutive parameters. By changing the soil constitutive parameters of the reference model, various types of sites are constructed, and by changing the pile parameters of individual piles, various L / D wind turbine monopiles are formed. In this way, a large number of numerical test cases are constructed to compensate for the lack of test pile data. Specifically, in one embodiment, the pile parameters specifically include the pile embedment depth L, pile diameter D, pile wall thickness t, and loading point height H; the soil constitutive parameters specifically include: soil effective unit weight γ', initial reference void ratio e0, Poisson's ratio υ, critical state line slope M, compression coefficient λ, expansion coefficient κ, permeability coefficient K, initial overconsolidation parameter R, and overconsolidation control parameter m. R Initial structural parameters R * and structural control parameter m *R .

[0043] Step S105: Analyze and calculate the same simulation case sequentially using the py curve method and the three-dimensional finite element reference model to obtain the corresponding py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. Based on the difference between the py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, determine the curve deviation of the py curve method relative to the three-dimensional finite element reference model.

[0044] Specifically, based on the case parameters of the expanded cases in step S104, a large number of horizontal load-displacement simulation experiments and calculations are performed using a three-dimensional finite element reference model to obtain the second model's horizontal ultimate bearing capacity corresponding to the horizontal displacement of the pile body at the mud surface reaching 0.1D for each simulation case. Simultaneously, the case parameters corresponding to each case are substituted into the py curve method to calculate the corresponding py horizontal ultimate bearing capacity. Finally, by combining the differences between the py horizontal ultimate bearing capacity of each case and the second model's horizontal ultimate bearing capacity, the curve deviation of the py curve method relative to the three-dimensional finite element reference model is determined. In this embodiment, the curve deviation is calculated as the ratio of the second model's horizontal ultimate bearing capacity to the py horizontal ultimate bearing capacity. The average value of the ratios from multiple cases can be taken to achieve the purpose of fusion, as shown in the following formula:

[0045]

[0046] In the formula, η represents the curve deviation of the py curve method relative to the three-dimensional finite element reference model, and P FEM P represents the horizontal ultimate bearing capacity of the second model. py This represents the horizontal limit capacity of py.

[0047] Specifically, in this embodiment of the invention, the API-recommended Python curve method that can be used according to the actual application scenario is as follows:

[0048] The recommended Py curve for the Clay API can be represented as follows:

[0049]

[0050] In the formula, p is the horizontal soil resistance; y is the horizontal displacement; y c =2.5ε c D, ε c The strain that appears at 0.5 times the maximum stress in an undrained compression test without disturbing the soil sample, where D is the pile diameter; p u The ultimate soil resistance is expressed as

[0051]

[0052] S uγ' is the undrained shear strength of the soil; X is the effective unit weight of the soil; J is the depth below the mud surface; J is a dimensionless empirical constant, which, based on experimental and engineering experience, ranges from 0.25 to 0.5; X R The depth from below the mud surface to the bottom of the zone of reduced soil resistance is denoted as...

[0053]

[0054] For sandy soil, the API-recommended py curve can be represented as follows:

[0055]

[0056] Where X is the depth below the mud surface; A is a coefficient considering static or cyclic loading conditions, where A = (3 - 0.8H / D) ≥ 0.9 under static conditions and A = 0.9 under cyclic loading; k is the initial modulus of the foundation reaction force, which can be determined according to API specifications; p u The ultimate soil resistance can be expressed as

[0057]

[0058] In the formula, C1, C2, and C3 are the internal friction angles. The function value can be obtained according to the API specification.

[0059] Step S106: Evaluate the error of the py curve model relative to the actual test pile data using the curve deviation and benchmark model error.

[0060] Specifically, finally, the curve deviation and benchmark model error from the above steps are comprehensively analyzed, and the two types of error parameters are fused. For example, new error parameters are generated based on the interval between the two types of error parameters, thereby obtaining the Py curve model error of the Py curve method relative to the actual test pile data. The model error evaluation method of the Py curve method for offshore wind power monopile provided by the embodiments of the present invention makes up for the deficiencies of test pile data, and can significantly improve the accuracy of evaluating the error in calculating the horizontal ultimate bearing capacity of monopile using the Py curve method, thus providing assistance for the reliability calculation of large-diameter monopile for offshore wind power and the development of design methods based on reliability theory.

[0061] Specifically, in one embodiment, step S105 above includes the following steps:

[0062] Step 1: Divide the portion of a single pile from below the mud surface to the pile bottom into a preset number of calculation units, and create finite difference equations based on the calculation units and the current simulation case.

[0063] Specifically, this embodiment of the invention analyzes and calculates the constructed numerical test simulation cases using the finite difference method and the py curve to obtain the py horizontal ultimate bearing capacity of each simulation case. First, the portion of a single pile from below the mud surface to the pile bottom is divided into a predetermined number of calculation units to prepare for creating finite difference equations. Then, finite difference equations are created based on the calculation units and the current simulation case. In this embodiment, the finite difference equations are as follows:

[0064]

[0065]

[0066]

[0067] In the formula, h is the length of the calculation unit; y i represents the horizontal displacement of the i-th node, where a node represents the endpoint of the computational unit; M0 is the bending moment acting on the pile at the mud surface; V0 is the horizontal shear force acting on the pile at the mud surface, where M0 and V0 are obtained by converting the initial horizontal load at the pile top; p i E represents the soil resistance acting at the i-th node; i I i Let be the bending stiffness of the i-th node, where the 0th node is the node at the mud surface. Nodes above the mud surface have negative values, and nodes below the mud surface have positive values. It is important to note that the soil resistance p0 at the mud surface is related to the soil resistance of the adjacent nodes {y}. -2 ,y -1 The relationships between y0, y1, and y2 are all present, therefore two virtual nodes y are usually added above the mud surface. -2 and y -1 ,。 At this time p i and y i All of these are unknown quantities; the only known quantity is the horizontal load at the top of the pile.

[0068] Step 2: Initialize the horizontal load on the pile top and the horizontal displacement of each node. The node is the endpoint of the calculation unit.

[0069] Step 3: Substitute the initial horizontal displacement of each node into the py curve to obtain the soil resistance of each node.

[0070] Step 4: Substitute the soil resistance of each node into the finite difference equation to calculate the new horizontal displacement of each node.

[0071] Step 5: If the error between the new horizontal displacement of each node and the initial horizontal displacement of each node is less than the preset threshold, and the new horizontal displacement of the pile body at the mud surface reaches 0.1 times the pile diameter, the initialized horizontal load at the top of the pile will be used as the py horizontal ultimate bearing capacity of the current simulation case.

[0072] Step 6: If the error between the new horizontal displacement of each node and the initial horizontal displacement of each node is greater than or equal to the preset threshold, or if the new horizontal displacement of the pile body at the mud surface does not reach 0.1 times the pile diameter, then return to the step of initializing the horizontal load at the top of the pile and the horizontal displacement of each node, and re-initialize.

[0073] Specifically, the calculation process begins. For the current simulation case, the horizontal load P at the top of the pile is initialized and converted into known quantities M0 and V0. The unknown quantity is y. i and p i First, initialize the horizontal displacement y of each node. i Given an unknown vector y i Assign a small value, such as 10. -3 ; Given y i In the case of y i Substituting the model parameters of the current simulation case into the py curve relationship provided in the above embodiment, the soil resistance p of each node is calculated. i At this point, the vector {M} on the right-hand side of the finite difference equation is known. -0 ,V0,p0,……p n Based on the above finite difference equation, the new horizontal displacement {y} at each node can be calculated. -2 new ,y -1 new ,y0 new ,……,y n new ,y n+1 new ,y n+2 new}

[0074] Next, determine whether the new horizontal displacement and the initial horizontal displacement of each node satisfy the conditions:

[0075]

[0076] Where Tol is a preset threshold, for example, 10. -5 .

[0077] If the above conditions are met, continue to determine the displacement y0 of the pile body at the mud surface. newIf the value reaches 0.1D, where D is the pile diameter, and if it does, the current horizontal load P acting on the pile top is taken as the ultimate horizontal bearing capacity of the single pile. Otherwise, return to step two, re-initialize, and iterate until the above conditions are met. The calculation process provided by this embodiment of the invention can accurately calculate the ultimate horizontal bearing capacity of the current simulation case (py). The calculation process for other simulation cases is the same. The above differential calculation process can be implemented using a finite difference calculation program written in Python based on the py curve method recommended by the API.

[0078] Specifically, in one embodiment, step S106 above includes the following steps:

[0079] Step 7: Estimate the probability distributions of the baseline model error and the curve deviation, respectively.

[0080] Specifically, in reliability calculations, variables such as horizontal load, ultimate horizontal bearing capacity of a single pile, and model error are all treated as random variables. Directly assessing these errors is challenging, as these random variables typically follow a certain probability distribution with a mean and coefficient of variation. Therefore, this embodiment of the invention uses probability distributions to assess the model error of the py curve, thereby improving the accuracy of model error assessment. First, the probability distributions of the baseline model error and curve deviation are estimated, such as normal distribution, chi-square distribution, and Poisson distribution. It is worth noting that in this embodiment of the invention, such as... Figure 4 As shown, the probability distributions of the estimated baseline model error and curve deviation are both log-normal distributions.

[0081] Step 8: Determine the first mean and first coefficient of variation corresponding to the baseline model error based on the estimated baseline model error probability distribution, and determine the second mean and second coefficient of variation corresponding to the curve deviation based on the estimated curve deviation probability distribution.

[0082] Step 9: Calculate the third mean corresponding to the error of the py curve model by multiplying the first mean and the second mean.

[0083] Step 10: Calculate the third coefficient of variation corresponding to the error of the Py curve model by taking the square root of the sum of the squares of the first and second coefficients of variation.

[0084] Specifically, given that the probability distributions of the baseline model error and the curve deviation are both log-normal distributions, the first mean and the first coefficient of variation corresponding to the baseline model error, and the second mean and the second coefficient of variation corresponding to the curve deviation, can be calculated using mathematical statistics methods. The process of calculating the mean and coefficient of variation based on the known probability distributions is existing technology and will not be elaborated upon here.

[0085] Once the first mean and first coefficient of variation corresponding to the baseline model error, and the second mean and second coefficient of variation corresponding to the curve deviation are determined, this embodiment of the invention calculates the third mean and third coefficient of variation using the following formula, where the third mean is the mean corresponding to the py curve model error, and the third coefficient of variation is the coefficient of variation corresponding to the py curve model error.

[0086] E(ε py )=E(ε FEM )·E(η)

[0087]

[0088] In the formula, E(ε) FEM ) represents the first mean, E(η) represents the second mean, and COV(ε) represents the third mean. FEM ) represents the first coefficient of variation, COV(η) represents the second coefficient of variation, and ε py E(ε) represents the error of the py curve model relative to the actual test pile data, where py curve method is used. py ) represents the third mean, COV(ε) py The third coefficient of variation is represented by . The third mean is determined by multiplying the first mean and the second mean, and the third coefficient of variation is determined by taking the square root of the sum of the squares of the first and second coefficients of variation.

[0089] Step 11: Use the Anderson-Darling test to examine the probability distribution form corresponding to the error of the py curve model, and determine the error of the py curve model using the known third mean, third coefficient of variation, and probability distribution form.

[0090] Specifically, the Anderson-Darling test can be used to measure how well our data fits a specified distribution (the Anderson-Darling test is an existing technique and will not be elaborated upon here). Thus, the probability distribution form corresponding to the error of the Py curve model can be obtained through the Anderson-Darling test, combined with the third mean E(ε) calculated in step ten above. py ) and the third coefficient of variation COV(ε) py By performing reverse calculation, the model error ε of the py curve can be obtained. py This enables accurate evaluation of the error in the Py curve method model, providing a new approach to solving the problem of difficulty in evaluating the error in the Py curve method model.

[0091] Through the above steps, the technical solution provided in this application, in the case of scarce test pile data for offshore wind turbine monopile foundations, proposes a method for calculating the model error of the Py curve method for evaluating offshore wind turbine monopile foundations, using a three-dimensional finite element reference model as a bridge: First, obtain the actual test pile data of the offshore wind turbine monopile, including the load-displacement curve; then, create a three-dimensional finite element reference model for analyzing the horizontal bearing capacity of the offshore wind turbine monopile; subsequently, calculate the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity under the same working conditions using the actual test pile data and the three-dimensional finite element reference model, respectively, and calculate the error based on the difference between the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity. The differences were identified by calibrating the 3D finite element reference model relative to the benchmark model of the actual test pile data. Then, the model parameters of the 3D finite element reference model were adjusted to obtain multiple simulation cases. The same simulation cases were analyzed and calculated sequentially using the Py curve method and the 3D finite element reference model, respectively, to obtain the corresponding Py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. Based on the difference between the Py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, the curve deviation of the Py curve method relative to the 3D finite element reference model was determined. Finally, the Py curve model error of the Py curve method relative to the actual test pile data was comprehensively evaluated by considering the curve deviation and the benchmark model error. This approach compensates for the deficiencies in test pile data, thereby improving the accuracy of the Py curve method's error assessment relative to actual test pile data, and providing assistance for the reliability calculation of large-diameter monopiles in offshore wind power and the development of design methods based on reliability theory.

[0092] like Figure 5 As shown in the figure, this embodiment also provides a model error evaluation device for the Py curve method of offshore wind power monopile, the device comprising:

[0093] The data acquisition module 101 is used to acquire actual test pile data of offshore wind turbine monopiles. The test pile data includes load-displacement curves, which characterize the relationship between the horizontal load at the pile top and the horizontal displacement of the pile body at the mud surface. For details, please refer to the relevant description of step S101 in the above method embodiment, which will not be repeated here.

[0094] The model creation module 102 is used to create a three-dimensional finite element reference model for analyzing the horizontal bearing capacity of offshore wind turbine monopiles. For details, please refer to the relevant description of step S102 in the above method embodiment, which will not be repeated here.

[0095] The finite element model error calibration module 103 is used to calculate the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity under the same working conditions using actual test pile data and a three-dimensional finite element reference model, respectively. Based on the difference between the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity, it calibrates the reference model error of the three-dimensional finite element reference model relative to the actual test pile data. For details, please refer to the relevant description of step S103 in the above method embodiment, which will not be repeated here.

[0096] The simulation data generation module 104 is used to adjust the model parameters of the three-dimensional finite element reference model to obtain multiple simulation cases. For details, please refer to the relevant description of step S104 in the above method embodiment, which will not be repeated here.

[0097] The curve deviation calibration module 105 is used to analyze and calculate the same simulation case sequentially using both the py curve method and the three-dimensional finite element reference model, respectively, to obtain the corresponding py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. Based on the difference between the py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, the curve deviation of the py curve method relative to the three-dimensional finite element reference model is determined. For details, please refer to the relevant description of step S105 in the above method embodiment, which will not be repeated here.

[0098] The Py curve method error assessment module 106 is used to assess the Py curve model error of the Py curve method relative to the actual test pile data through curve deviation and benchmark model error. For details, please refer to the relevant description of step S106 in the above method embodiment, which will not be repeated here.

[0099] The model error evaluation device for the Py curve method of offshore wind power monopile provided in this embodiment of the invention is used to execute the model error evaluation method for the Py curve method of offshore wind power monopile provided in the above embodiment. Its implementation method and principle are the same. For details, please refer to the relevant description of the above method embodiment, which will not be repeated here.

[0100] Through the collaborative efforts of the aforementioned components, the technical solution provided in this application, in the context of scarce test pile data for offshore wind turbine monopile foundations, proposes a method for calculating the error of the Py-curve method for evaluating offshore wind turbine monopile foundations, using a three-dimensional finite element reference model as a bridge. First, actual test pile data of the offshore wind turbine monopile is obtained, including load-displacement curves. Then, a three-dimensional finite element reference model is created for analyzing the horizontal bearing capacity of the offshore wind turbine monopile. Subsequently, the measured horizontal ultimate bearing capacity and the first model's horizontal ultimate bearing capacity under the same working conditions are calculated using the actual test pile data and the three-dimensional finite element reference model, respectively. Finally, the method is used to calculate the measured horizontal ultimate bearing capacity and the first model's horizontal ultimate bearing capacity. The difference in bearing capacity was investigated to calibrate the baseline model error of the 3D finite element (FEM) benchmark model relative to the actual test pile data. Then, the model parameters of the 3D FEM benchmark model were adjusted to obtain multiple simulation cases. The same simulation cases were analyzed and calculated sequentially using the Py curve method and the 3D FEM benchmark model to obtain the corresponding Py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. Based on the difference between the Py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, the curve deviation of the Py curve method relative to the 3D FEM benchmark model was determined. Finally, the Py curve model error of the Py curve method relative to the actual test pile data was comprehensively evaluated by considering the curve deviation and the baseline model error. This approach compensates for the deficiencies in test pile data, thereby improving the accuracy of the Py curve method in evaluating the horizontal ultimate bearing capacity error relative to actual test pile data. This provides assistance for the reliability calculation of large-diameter monopiles in offshore wind power and the development of design methods based on reliability theory.

[0101] Figure 6 This invention illustrates a model error evaluation device for the Py curve method of offshore wind power monopile, comprising a processor 901 and a memory 902, which can be connected via a bus or other means. Figure 6 Taking the example of a connection between China and Israel via a bus.

[0102] Processor 901 can be a Central Processing Unit (CPU). Processor 901 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0103] The memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the above method embodiments. The processor 901 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory 902, thereby implementing the methods in the above method embodiments.

[0104] The memory 902 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor 901, etc. Furthermore, the memory 902 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 902 may optionally include memory remotely located relative to the processor 901, and these remote memories may be connected to the processor 901 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0105] One or more modules are stored in memory 902, and when executed by processor 901, they perform the methods described in the above method embodiments.

[0106] The specific details of the model error assessment equipment for the above-mentioned offshore wind power monopile Py curve method can be understood by referring to the relevant descriptions and effects in the above method embodiments, and will not be repeated here.

[0107] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The implemented program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0108] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A model error evaluation method for the Py curve method of offshore wind power monopile, characterized in that, The method includes: Obtain test pile data for offshore wind turbine monopiles, including load-displacement curves, which are used to characterize the relationship between the horizontal load at the top of the pile and the horizontal displacement of the pile body at the mud surface. Create a three-dimensional finite element benchmark model for analyzing the horizontal bearing capacity of monopiles in offshore wind power; The measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity under the same working conditions are calculated using the measured pile data and the three-dimensional finite element reference model, respectively. Based on the difference between the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity, the reference model error of the three-dimensional finite element reference model relative to the measured pile data is calibrated. By adjusting the model parameters of the three-dimensional finite element reference model, multiple simulation cases were obtained; The same simulation case was analyzed and calculated sequentially using the py curve method and the three-dimensional finite element reference model to obtain the corresponding py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. Based on the difference between the py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, the curve deviation of the py curve method relative to the three-dimensional finite element reference model was determined. The error of the py curve method relative to the actual test pile data is evaluated by the curve deviation and the benchmark model error.

2. The method according to claim 1, characterized in that, The model parameters include pile parameters and soil constitutive parameters. The pile parameters include pile depth, pile diameter, pile wall thickness, and loading point height. The soil constitutive parameters include: effective unit weight, initial reference void ratio, Poisson's ratio, slope of critical state line, compression coefficient, expansion coefficient, permeability coefficient, initial overconsolidation parameter, overconsolidation control parameter, initial structural parameter, and structural control parameter.

3. The method according to claim 1, characterized in that, The steps for calculating the ultimate horizontal bearing capacity of each simulation case using the Py curve method include: The portion of a single pile from below the mud surface to the pile bottom is divided into a predetermined number of calculation units, and finite difference equations are created based on the calculation units and the current simulation case. Initialize the horizontal load on the pile top and the horizontal displacement of each node, where the node is the endpoint of the calculation unit; Substitute the initial horizontal displacement of each node into the py curve to obtain the soil resistance of each node. Substitute the soil resistance at each node into the finite difference equation to calculate the new horizontal displacement at each node. If the error between the new horizontal displacement of each node and the initial horizontal displacement of each node is less than the preset threshold, and the new horizontal displacement of the pile body at the mud surface reaches 0.1 times the pile diameter, the initialized horizontal load at the top of the pile will be used as the py horizontal ultimate bearing capacity of the current simulation case. If the error between the new horizontal displacement of each node and the initial horizontal displacement of each node is greater than or equal to the preset threshold, or if the new horizontal displacement of the pile body at the mud surface does not reach 0.1 times the pile diameter, then return to the step of initializing the horizontal load at the top of the pile and the horizontal displacement of each node, and re-initialize.

4. The method according to claim 1, characterized in that, The curve deviation is the ratio of the second model's horizontal ultimate bearing capacity to the py horizontal ultimate bearing capacity; the benchmark model error is the ratio of the measured horizontal ultimate bearing capacity to the first model's horizontal ultimate bearing capacity.

5. The method according to claim 4, characterized in that, The method of evaluating the py curve model error of the py curve method relative to the actual test pile data through the curve deviation and the benchmark model error includes: Estimate the probability distributions of the baseline model error and the curve deviation, respectively; The first mean and the first coefficient of variation of the baseline model error are determined based on the estimated baseline model error probability distribution, and the second mean and the second coefficient of variation of the curve deviation are determined based on the estimated curve deviation probability distribution. The third mean corresponding to the error of the py curve model is calculated by multiplying the first mean and the second mean; The third coefficient of variation corresponding to the error of the py curve model is calculated by taking the square root of the sum of the squares of the first and second coefficients of variation. The probability distribution form corresponding to the error of the py curve model is tested by the Anderson-Darling test, and the py curve model error is determined by the third mean, the third coefficient of variation, and the probability distribution form corresponding to the error of the py curve model.

6. The method according to claim 5, characterized in that, The probability distributions of the baseline model error and the curve deviation are estimated to be log-normal.

7. The method according to claim 3, characterized in that, The finite difference equations include: In the formula, h is the length of the calculation unit; y i Let Mi represent the horizontal displacement of the i-th node; M0 be the bending moment acting on the pile at the mud surface; and V0 be the horizontal shear force acting on the pile at the mud surface, where M0 and V0 are derived from the initial horizontal load at the pile top; p i E represents the soil resistance acting at the i-th node; i I i Let be the bending stiffness of the i-th node, where the 0th node is the node at the mud surface, nodes above the mud surface are negative, and nodes below the mud surface are positive.

8. A model error evaluation device for the Py-curve method of offshore wind power monopile, characterized in that, The device includes: The data acquisition module is used to acquire test pile data of offshore wind turbine monopiles. The test pile data includes load-displacement curves, which are used to characterize the relationship between the horizontal load at the top of the pile and the horizontal displacement of the pile body at the mud surface. The model creation module is used to create a three-dimensional finite element benchmark model for analyzing the horizontal bearing capacity of offshore wind power monopiles. The finite element model error calibration module is used to calculate the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity under the same working conditions using the measured pile data and the three-dimensional finite element reference model, respectively, and to calibrate the reference model error of the three-dimensional finite element reference model relative to the measured pile data based on the difference between the measured horizontal ultimate bearing capacity and the first model horizontal ultimate bearing capacity. The simulation data generation module is used to adjust the model parameters of the three-dimensional finite element reference model to obtain multiple simulation cases; The curve deviation calibration module is used to analyze and calculate the same simulation case sequentially using the py curve method and the three-dimensional finite element reference model, respectively, to obtain the corresponding py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity, and to determine the curve deviation of the py curve method relative to the three-dimensional finite element reference model based on the difference between the py horizontal ultimate bearing capacity and the second model horizontal ultimate bearing capacity. The Py curve method error assessment module is used to assess the Py curve model error of the Py curve method relative to the actual test pile data by means of the curve deviation and the benchmark model error.

9. A model error assessment device for the Py-curve method of offshore wind power monopile, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method as described in any one of claims 1-7.

Images