Main scale optimization design method for floating foundation of floating fan

Abstract

The invention discloses a main scale optimization design method for a floating type foundation of a floating type draught fan, and relates to the field of ocean structure design optimization, and the method comprises the steps: determining a change interval of main scale parameters of the floating type foundation based on mother type main scale parameters, and randomly generating a plurality of groups of candidate main scale parameters, constructing a floating foundation model by using a parametric modeling method; determining candidate weight distribution parameters corresponding to the candidate main scale parameters, transferring a hydrodynamic analysis module, and performing hydrodynamic calculation on the floating foundation model by taking the candidate weight distribution parameters as input; and determining a comprehensive objective function and constraint conditions according to a calculation result, solving a main scale parameter set when the comprehensive objective function is optimal by adopting a multi-objective particle swarm optimization algorithm under the constraint conditions, and obtaining a floating foundation main scale optimization design scheme. An intelligent optimization algorithm is combined with an automatic water transfer hydrodynamic analysis module, so that the efficiency and precision of the main scale optimization design of the floating foundation are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of marine structure design optimization, and in particular to a method for optimizing the main dimensions of a floating foundation of a floating wind turbine. Background Art

[0002] The development and utilization of clean energy is a necessary means to achieve the dual carbon goals, and wind energy is an important one among them. With the gradual improvement of the development of wind power generation equipment on land and offshore, deep-sea areas with stable wind farms and abundant wind energy resources have gradually received attention in recent years. The design and development of wind power generation equipment plays a vital role in improving the performance of wind power generation equipment in deep-sea operations. Since deep-sea areas are far away from land and have deep waters, conventional jacket-type wind power equipment is difficult to meet operational requirements, and floating wind turbines are more suitable. However, under the combined action of wind, waves and currents, the hydrodynamic response characteristics of the floating foundation of floating wind turbines moored on the sea surface are complex. Improving the performance of floating wind turbines under wave action by optimizing the structural scale has become an urgent problem that needs to be solved.

[0003] The optimization design of traditional floating wind turbine foundations relies on manual parametric modeling. Performance calculations are performed under specific primary scale parameters, and then the results are fitted to analyze the effects of varying primary scale parameters. However, this approach relies heavily on manual experience and has limitations in the selection of initial solutions, which compromises the overall effectiveness of the optimization process. This leads to significant time and effort required during the design process, and also poses the problem of incomplete primary scale parameter design. Summary of the Invention

[0004] In response to the above-mentioned problems and technical requirements, this application proposes a method for optimizing the main dimensions of a floating foundation for a floating wind turbine. The technical solution of this application is as follows:

[0005] A method for optimizing the main dimensions of a floating foundation of a floating wind turbine comprises the following steps:

[0006] Determine target sea state parameters, determine the main scale parameters and weight distribution parameters of the mother template of the floating foundation to be optimized, and determine the variation range of the main scale parameters of the floating foundation based on the main scale parameters of the mother template; the weight distribution parameters indicate the weight distribution of each component of the floating foundation;

[0007] Randomly generate multiple sets of candidate main scale parameters within the variation range of the floating foundation main scale parameters, and construct multiple floating foundation models using a parametric modeling method;

[0008] Determine the candidate weight distribution parameters corresponding to each set of candidate principal scale parameters based on the parent model principal scale parameters and the parent model weight distribution parameters; call the hydrodynamic analysis module, use each set of candidate weight distribution parameters as input to perform hydrodynamic calculations on each floating foundation model, and obtain multiple sets of hydrodynamic performance indicators; the hydrodynamic performance indicators include wind-induced heel angle, natural period, and short-term forecast;

[0009] The comprehensive objective function for optimizing the main dimensions of the floating foundation is determined based on wind-induced heel angles and short-term forecasts. The comprehensive objective function characterizes the stability of the floating foundation model. The displacement of the floating foundation model is determined based on candidate weight distribution parameters. The constraints for optimizing the main dimensions of the floating foundation are determined based on the relationship between the natural period and the spectral peak period of the target sea state parameters, based on the condition that the displacement of the floating foundation model meets the design requirements.

[0010] Under the constraints, a multi-objective particle swarm optimization algorithm is used to solve the main scale parameter set that makes the comprehensive objective function reach the optimal value, and the main scale optimization design scheme of the floating foundation is obtained.

[0011] A further technical solution is that the comprehensive objective function includes a first objective function f that minimizes the wind-induced tilt angle. w (x), and the second objective function f that minimizes the short-term forecast r (x); determine the comprehensive objective function F(x) = (f w (x),f r (x)) T ;

[0012] Where x is the candidate main scale parameter vector and x=(x1,x2,...,x j ,...,x n ) T , x j is the jth principal scale variable in the candidate principal scale parameter vector x and x jmin ≤x j ≤x jmax , x jmin is the lower limit of the change interval of the jth main scale variable, x jmax is the upper limit of the variation interval of the jth main scale variable. Each component of the floating basic model includes several main scale variables. n is the total number of main scale variables. The integer parameter n≥1, (·) T Represents matrix transpose.

[0013] A further technical solution is to perform hydrodynamic calculation on the floating foundation model to obtain the hydrostatic stiffness matrix k of the floating foundation model, and determine the wind-induced inclination angle according to the hydrostatic stiffness matrix k. According to the wind-induced inclination angle g w Determine the first objective function f w (x) = gw ;

[0014] Where c is the dimensionless ultimate wind load, and [k] represents the hydrostatic stiffness value of the degree of freedom corresponding to the wind direction in the hydrostatic stiffness matrix k of the floating foundation model.

[0015] A further technical solution is to perform hydrodynamic calculation on the floating foundation model to obtain the RAO motion response data RAO of the floating foundation model at any degree of freedom i. i (ω), according to the RAO motion response data RAO of degree of freedom i i (ω) Determine the short-term prediction of the floating foundation model in degree of freedom i by combining the target sea state parameters

[0016] Short-term forecast based on degrees of freedom i Determine the second objective function Where S(ω) is the wave spectrum of the target sea state parameters, m is the total number of degrees of freedom, and the integer parameter m≥1.

[0017] A further technical solution is to perform hydrodynamic calculations on the floating foundation model to obtain the motion cycle data of the floating foundation model, and determine the natural period of the degree of freedom i based on the motion cycle data.

[0018] Determine the constraints for the optimal design of the main dimensions of the floating foundation:

[0019]

[0020] Among them, Δ x is the displacement of the floating foundation model corresponding to the candidate main scale parameter vector x, Δ max is the upper threshold of the water discharge, T p is the spectrum peak period of the target sea state parameter.

[0021] A further technical solution is that determining the candidate weight distribution parameters corresponding to the candidate main scale parameters includes:

[0022] Determine the weight of the master form of each component of the floating foundation model according to the master form weight distribution parameters;

[0023] Determine the volume change of each component of the floating foundation model according to the change of the candidate main scale parameters relative to the main scale parameters of the parent model;

[0024] Based on the master weight of each component of the floating foundation model and according to the mapping relationship between the weight change and volume change of each component of the floating foundation model, the weight change of each component of the floating foundation model is determined to obtain candidate weight distribution parameters.

[0025] A further technical solution is that for any k-th component of the floating foundation model, according to the volume change ΔV of the k-th component k Determine the weight change of the kth component in, is the volume of the matrix of the kth component, is the parent weight of the kth component.

[0026] Its further technical solution is that the motion cycle data includes a column of natural cycle data and multiple columns of motion cycle data with different degrees of freedom, and the natural cycle of degree of freedom i is determined according to the motion cycle data. include:

[0027] The motion cycle data of degree of freedom i and the natural cycle data are subtracted column by column to obtain the difference data column, and the motion cycle data corresponding to the minimum value in the difference data column is selected as the natural cycle of degree of freedom i

[0028] A further technical solution is to determine that the comprehensive objective function reaches the optimum when the first objective function and the second objective function are both as small as possible and the function values ​​of the first objective function and the second objective function fluctuate within a predetermined range.

[0029] Its further technical solution is that the hydrodynamic analysis module adopts the aqwa-line module;

[0030] Parametric modeling methods include:

[0031] Based on the APDL command stream modeling method, the candidate principal scale parameters are constructed as a floating foundation model to obtain an .aqwa input file, which includes the model parameters of the floating foundation model. The candidate weight distribution parameters in the .aqwa input file are automatically modified by a Python script to generate a .dat calculation file suitable for the AQWA-Line module.

[0032] Calling the hydrodynamic analysis module includes:

[0033] Based on the Python script, batch commands are generated to automatically call the aqwa-line module to perform hydrodynamic calculations and generate a .lis hydrodynamic calculation result file. The hydrodynamic calculation result file includes the hydrostatic stiffness matrix, motion period data, and RAO motion response data of the floating foundation model.

[0034] The beneficial technical effects of this application are:

[0035] This application proposes a method for optimizing the principal dimensions of a floating foundation for a floating wind turbine. This method uses a parametric modeling method and batch processing commands to automatically call a hydrodynamic analysis module to obtain the hydrodynamic performance indicators of the floating foundation model. This method, combined with a multi-objective particle swarm optimization algorithm, automatically and iteratively optimizes the principal dimensions of the floating foundation based on the hydrodynamic performance indicators of the floating foundation model. Compared to the traditional method of manually adjusting parameters through modeling, this fully automated method for optimizing the principal dimensions of the floating foundation significantly improves the efficiency of the floating foundation principal dimensions optimization design process. Furthermore, by analyzing the hydrodynamic characteristics of the floating foundation model as the optimization target of the intelligent optimization algorithm, the rationality and effectiveness of the designed principal dimensions of the floating foundation are ensured.

[0036] The objective function for optimizing the main dimensions of the floating foundation is constructed by analyzing the short-term forecasts of wind-induced inclination angles and the amplitudes of motion of each degree of freedom, which are related to the stability of the floating wind turbine in the hydrodynamic performance indicators. By using the natural period related to the vibration characteristics of the floating foundation model in the hydrodynamic performance indicators to be greater than the spectral peak period of the target sea condition as a constraint, resonance between the floating foundation model and the wave fluctuations can be avoided, thereby further ensuring the stability of the floating wind turbine. At the same time, limiting the displacement of the floating foundation model as a constraint further ensures the rationality of the structure and performance of the designed floating foundation. By converting the actual operational requirements of the floating wind turbine into a mathematical model and solving it using an intelligent optimization algorithm, designers can quantitatively understand the design effects, thereby quickly and efficiently designing an optimized design scheme for the main dimensions of the floating foundation that conforms to the actual operational rules.

[0037] Considering that the accuracy of the weight distribution of each component of the floating foundation determines the accuracy of the hydrodynamic calculation results, this application automatically calculates the weight distribution parameters of each component corresponding to different main scale parameters by analyzing the mapping relationship between the change in the weight of each component relative to the weight of the floating foundation mother model and the change in the volume of each component relative to the volume of the floating foundation mother model. Compared with the conventional method of using default values, this method effectively improves the accuracy of the hydrodynamic calculation results, thereby ensuring the accuracy of the designed floating foundation main scale optimization design scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of the optimization design method of the main dimensions of the floating foundation.

[0039] Figure 2 The figure is a schematic diagram of iterative convergence of the objective function of an embodiment. DETAILED DESCRIPTION

[0040] The specific implementation of this application will be further described below with reference to the accompanying drawings.

[0041] This application proposes a floating foundation main scale optimization design method for floating wind turbines, please refer to Figure 1 The specific steps of the method are as follows:

[0042] Step 1: Determine the target sea state parameters, determine the main scale parameters and weight distribution parameters of the mother template of the floating foundation to be optimized, and determine the variation range of the main scale parameters of the floating foundation based on the main scale parameters of the mother template; the weight distribution parameters indicate the weight distribution of each component of the floating foundation.

[0043] The main scale optimization design of the floating foundation of a floating wind turbine needs to be based on the relevant parameters of the floating foundation prototype. By adjusting the main scale parameters of the floating foundation prototype, a floating foundation main scale design scheme with better performance than the floating foundation prototype is designed. According to the actual application requirements, the variation range of the main scale parameters relative to the main scale parameters of the prototype is set to obtain the variation range of the main scale parameters of the floating foundation. Among them, the floating foundation includes multiple components. This application is divided into four categories according to the professional fields to which each component belongs: structure, outfitting, turbine and electrical. The structure includes the strut structure, column structure, lower buoy structure, wind turbine structure, welding materials and structural construction allowance; the turbine structure includes the column turbine equipment, lower buoy turbine equipment and equipment allowance; the outfitting includes the column outfitting equipment, lower buoy outfitting equipment, coating materials, sacrificial anode materials and equipment allowance; the electrical includes the column electrical equipment and equipment allowance. The main scale parameters of each component include several main scale variables. For example, the column structure of the floating foundation includes the column side length and column height. Each main scale variable corresponds to a variation interval. When generating candidate main scale parameters, it is necessary to ensure that each main scale variable is within the corresponding variation interval.

[0044] Another type of parameter for the floating foundation master form is the master form weight distribution parameter. These parameters are used to calculate candidate weight distribution parameters for each set of candidate principal scale parameters. In other words, the weight distribution of each component of the master form is used to determine the weight distribution of each floating foundation component after the principal scale parameters are changed. These candidate weight distribution parameters serve as input for subsequent hydrodynamic calculations and have a significant impact on the accuracy of these results.

[0045] Step 2: randomly generate multiple sets of candidate main scale parameters within the variation range of the floating foundation main scale parameters, and construct multiple floating foundation models using a parametric modeling method.

[0046] The specific method for generating candidate principal scale parameters involves randomly generating principal scale variables based on random number theory. Based on the variation range of each principal scale variable, multiple candidate principal scale variables are selected within the variation range. These candidate principal scale variables are then combined to generate candidate principal scale parameters. Each set of candidate principal scale parameters includes platform radius, draft, lower buoyancy half-width, lower buoyancy height, column side length, column height, heave plate radius, and heave plate height. In the multi-objective particle swarm optimization algorithm, each set of candidate principal scale parameters is considered an individual, and the randomly generated sets of candidate principal scale parameters constitute the initial population.

[0047] Step 3: Determine the candidate weight distribution parameters corresponding to each group of candidate main scale parameters based on the main scale parameters and the weight distribution parameters of the mother model; call the hydrodynamic analysis module, use each group of candidate weight distribution parameters as input to perform hydrodynamic calculations on each floating foundation model, and obtain multiple groups of hydrodynamic performance indicators; the hydrodynamic performance indicators include wind-induced heel angle, natural period and short-term forecast, the wind-induced heel angle and short-term forecast represent the motion stability of the floating foundation model, and the natural period represents the inherent motion properties of the floating foundation model.

[0048] Traditional modeling methods use 3D modeling software to manually construct a floating foundation model, and then use hydrodynamic analysis software to calculate the hydrodynamic response of the floating foundation model. Conventional AQWA hydrodynamic analysis software is primarily used through interactive interfaces, requiring manual operation, which significantly impacts the efficiency of model construction and hydrodynamic calculations. To improve the efficiency of model construction and hydrodynamic calculations, this application utilizes parametric modeling methods and batch processing commands to automatically invoke the AQWA-Line module within the AQWA hydrodynamic analysis software to automate floating foundation model construction and hydrodynamic calculations.

[0049] In one embodiment, the specific method of parametric modeling is as follows: based on the APDL command stream modeling method, the candidate main scale parameters are constructed as a floating base model to obtain an .aqwa input file, which includes the model parameters of the floating base model; and the candidate weight distribution parameters in the .aqwa input file are automatically modified by a Python script to generate a .dat calculation file adapted for the AQWA-Line module.

[0050] The specific method for calling the hydrodynamic analysis module is to generate a batch command based on a Python script to automatically call the aqwa-line module to perform hydrodynamic calculations and generate a .lis hydrodynamic calculation result file. The hydrodynamic calculation result file includes the hydrostatic stiffness matrix, motion period data, and RAO motion response data of the floating foundation model.

[0051] According to the generated candidate main scale parameters, the floating foundation model is constructed using the parametric modeling method, which automatically generates an input file in the .aqwa format containing the model parameters of the floating foundation model. The input file is used as the input of the hydrodynamic analysis module. However, since some parameters in the .aqwa format input file automatically generated by the parametric modeling method are automatically created and do not necessarily meet the actual calculation requirements, appropriate modifications are required. At the same time, the aqwa-line module requires a calculation file in the .dat format as input, so the .aqwa format input file needs to be converted into a .dat format calculation file. Among them, for reading, modifying and writing parameters in the .aqwa format input file, this application uses the cmd command-based APDL modeling sub-thread developed by the Python program to call the ANSYS-APDL module of the ANSYS three-dimensional modeling software, and uses the script developed by the Python program to generate a batch command to call the AQWA-line module of the AQWA hydrodynamic analysis software. Through this method, it is possible to automatically calculate the hydrodynamic response results when the main scale parameters change.

[0052] The candidate weight distribution parameters of the floating foundation model in the automatically generated .aqwa input file are often inaccurate. The candidate weight distribution parameters are key parameters for hydrodynamic calculations. Therefore, the candidate weight distribution parameters in the .aqwa input file need to be modified accordingly after the main scale parameters are changed.

[0053] In one embodiment, the specific steps of determining the candidate weight distribution parameters corresponding to the candidate principal scale parameters are as follows:

[0054] (1) Determine the master weight of each component of the floating foundation model based on the master weight distribution parameters.

[0055] (2) Determine the volume change of each component of the floating foundation model based on the change of the candidate main scale parameters relative to the main scale parameters of the parent model.

[0056] According to the classification of each component, the components whose weight changes with the change of the main scale parameters and the components whose weight remains unchanged when the main scale parameters change are screened out. This application analyzes the influence of the structures or equipment in the four categories of structure, outfitting, turbine and electrical on the performance of the floating wind turbine, and takes into account the simplification of calculations to improve calculation efficiency, and only sets the weight of the key components to change with the main scale parameters. Specifically, the weight of the strut structure, column structure, lower floating body structure, wind turbine structure, column turbine equipment, lower floating body turbine equipment, column outfitting equipment, lower floating body outfitting equipment, and column electrical equipment is set to change with the change of the main scale parameters, and the weight of the remaining non-critical components is set to remain unchanged.

[0057] (3) Based on the master weight of each component of the floating foundation model, the weight change of each component of the floating foundation model is determined according to the mapping relationship between the weight change and volume change of each component of the floating foundation model, and the candidate weight distribution parameters are obtained.

[0058] For each component whose weight varies with the main scale parameter, the weight change of each component is different and is related to the volume of the component and the weight and volume of its parent model. In one embodiment, for any k-th component of the floating foundation model, according to the volume change ΔV of the k-th component k Determine the weight change of the kth component in, is the volume of the matrix of the kth component, is the parent weight of the kth component.

[0059] The volume calculation method of each component is also different. It is calculated according to the specific geometric structure shape according to the conventional calculation formula. Taking the strut as an example, the volume of the strut V bar The calculation formula is:

[0060]

[0061] Where r1 is the platform radius, r2 is the column side length, r3 is the half-width of the strut, and h is the strut height.

[0062] According to the mapping relationship corresponding to each component, the weight change of each component can be calculated, and the weight change of each component can be superimposed on the weight of the parent model to obtain the weight of each component corresponding to the candidate main scale parameter, and the candidate weight distribution parameter can be obtained.

[0063] In addition, the candidate weight distribution parameters also include the floating foundation center of gravity parameters, which are calculated based on the center of gravity position of each component. The calculation method of the center of gravity position of each component needs to refer to the setting of its parent coordinate system, and the center of gravity calculation formula is calculated according to its specific geometric structure. Taking the column as an example, its vertical center of gravity position is the coordinate along the Z axis of its parent coordinate system. The vertical center of gravity position z of the column is the coordinate along the Z axis of its parent coordinate system. new The calculation method is h old is the length of the female column, h new is the column length among the candidate main scale parameters, z oldis the vertical position of the center of gravity of the master model; the corresponding coordinates along the X and Y directions are calculated based on the corresponding geometric structure shape. After obtaining the center of gravity position and weight of each component, the mass moment of the floating foundation model is calculated based on the center of gravity position and weight of each component. The mass moment of the floating foundation model is divided by the total mass of the floating foundation model to obtain the floating foundation center of gravity parameter corresponding to the candidate principal scale parameter. The specific calculation of the mass moment is performed according to conventional calculation methods and will not be repeated in this application.

[0064] Step 4: Determine the comprehensive objective function for the optimization design of the main scale of the floating foundation based on the wind-induced inclination angle and the short-term forecast, and the comprehensive objective function characterizes the stability of the floating foundation model; determine the displacement of the floating foundation model based on the candidate weight distribution parameters, and determine the constraint conditions for the optimization design of the main scale of the floating foundation based on the condition that the displacement of the floating foundation model meets the design requirements and the relationship between the natural period and the spectral peak period of the target sea state parameters.

[0065] Because floating foundations are moored on the sea surface, where wind and waves fluctuate significantly, they are not always stable. Therefore, to ensure the performance of floating wind turbines on the sea surface, the goal of the primary dimension optimization design of the floating foundation is to ensure good stability. By analyzing the factors influencing the stability of the floating wind turbine foundation, the objective function of the multi-objective particle swarm optimization algorithm was determined.

[0066] The short-term prediction of the wind-induced inclination angle of a floating wind turbine and the amplitude of movement of the floating foundation under the action of wind force are critical to the stability of the floating wind turbine. Excessive wind-induced inclination angle and short-term prediction may cause the floating wind turbine to tilt or even roll over. In one embodiment, the comprehensive objective function for determining the optimal design of the main dimensions of the floating foundation includes a first objective function f that minimizes the wind-induced inclination angle. w (x), and the second objective function f that minimizes the short-term forecast r (x); determine the comprehensive objective function F(x) = (f w (x),f r (x)) T ;

[0067] Where x is the candidate main scale parameter vector and x=(x1,x2,...,x j ,...,x n ) T , x j is the jth principal scale variable in the candidate principal scale parameter vector x and x jmin ≤x j ≤x jmax , x jmin is the lower limit of the change interval of the jth main scale variable, x jmaxis the upper limit of the variation interval of the jth main scale variable. Each component of the floating basic model includes several main scale variables. n is the total number of main scale variables. The integer parameter n≥1, (·) T Represents matrix transpose.

[0068] The method for determining the first objective function is: performing hydrodynamic calculation on the floating foundation model to obtain the hydrostatic stiffness matrix k of the floating foundation model, and determining the wind-induced inclination angle according to the hydrostatic stiffness matrix k According to the wind-induced inclination angle g w Determine the first objective function f w (x) = g w ;

[0069] Where c is the dimensionless ultimate wind load, which is a constant, and [k] represents the hydrostatic stiffness value of the degree of freedom corresponding to the incoming wind direction in the hydrostatic stiffness matrix k of the floating foundation model.

[0070] Because the horizontal movement of a floating wind turbine on the sea surface does not affect its performance, only its vertical stability is considered. The values ​​on the diagonal of the calculated hydrostatic stiffness matrix k represent the hydrostatic stiffness values ​​for each degree of freedom. When calculating wind-induced heave, only the hydrostatic stiffness values ​​for roll, pitch, and heave, which are relevant to vertical stability, are selected. Because wind has a minimal effect on the heave direction of a floating wind turbine, the wind direction determines whether the hydrostatic stiffness value is selected for roll or pitch.

[0071] The method for determining the second objective function is: performing hydrodynamic calculation on the floating foundation model to obtain the RAO motion response data RAO of the floating foundation model at any degree of freedom i i (ω), according to the RAO motion response data RAO of degree of freedom i i (ω) Determine the short-term prediction of the floating foundation model in degree of freedom i by combining the target sea state parameters Short-term forecast based on degrees of freedom i Determine the second objective function Among them, S(ω) is the wave spectrum of the target sea state parameters, m is the total number of degrees of freedom, and the integer parameter m≥1. In this application, considering the stability in the vertical direction, the total number of degrees of freedom is 3, and the degrees of freedom i∈(1,3) are the roll direction, pitch direction, and heave direction, respectively. Since it is necessary to consider the RAO motion response in multiple directions, setting too many objective functions will seriously affect the computational efficiency of the optimization iteration. Therefore, in order to improve the computational efficiency, the short-term forecast of each degree of freedom is integrated into one objective function. Considering that the RAO change trend of each degree of freedom is the same, integrating the short-term forecast of each degree of freedom into one objective function does not affect the accuracy of the iterative optimization results.

[0072] In order to prevent the floating foundation model from resonating with the fluctuations of the waves and further ensure the stability of the floating wind turbine, the natural period related to the vibration characteristics of the floating foundation model is analyzed as a constraint condition. At the same time, the displacement of the floating foundation model is limited as a constraint condition to further ensure the rationality of the structure and performance of the designed floating foundation.

[0073] In one embodiment, the method for determining the constraint condition is: performing hydrodynamic calculation on the floating foundation model to obtain the motion period data of the floating foundation model, and determining the natural period of the degree of freedom i according to the motion period data.

[0074] Determine the constraints for the optimal design of the main dimensions of the floating foundation:

[0075]

[0076] Among them, Δ x is the displacement of the floating foundation model corresponding to the candidate main scale parameter vector x, and the displacement of the floating foundation model is calculated by summing the weights of the various components in the candidate weight distribution parameters. max It is the upper limit of the water discharge, which can be customized according to the actual application. p is the spectral peak period of the target sea state parameter. The constraint condition for the optimization design of the main dimensions of the floating foundation needs to ensure that the natural period of each degree of freedom is not less than the spectral peak period of the target sea state parameter.

[0077] The hydrostatic stiffness matrix, RAO motion response data, and motion period data are read from the .lis hydrodynamic calculation result file generated by the aqwa-line module. The motion period data includes a column of natural period data and multiple columns of motion period data for different degrees of freedom. The natural period of degree of freedom i is determined based on the motion period data. Including: subtracting the motion cycle data of degree of freedom i from the natural cycle data by column to obtain a difference data column, and selecting the motion cycle data corresponding to the minimum value in the difference data column as the natural cycle of degree of freedom i The natural period data is the theoretical value of the motion period corresponding to the frequency. The natural period is obtained by comparing the actual calculated motion period of each degree of freedom with the theoretical value. The motion period data characterizes the inherent properties of the floating foundation model and is expressed in seconds. Each column of motion period data has the same number of data as the natural period data, and each data point corresponds one-to-one with the frequency to be calculated. When calculating the motion response of the floating foundation model in the frequency domain, the frequency domain range required for calculation should theoretically be set to the entire frequency calculation domain. However, in reality, wave energy is primarily concentrated within a certain range. Therefore, for computational efficiency, a specific frequency domain range can be set as the frequency calculation domain. The specific frequency domain range can be customized according to actual needs.

[0078] Step 5: Under the constraints, a multi-objective particle swarm optimization algorithm is used to solve the main scale parameter set that makes the comprehensive objective function reach the optimal state, and the main scale optimization design scheme of the floating foundation is obtained.

[0079] The objective function in step 4 can be used to calculate the fitness of each individual in the population. Based on these individual fitness values, the non-dominated solution set of the multi-objective optimization problem is calculated according to the Pareto frontier solution set theory. Any non-dominated solution in the non-dominated solution set is randomly selected as the global optimal solution, guiding the generation of a new generation of populations, and continuing the algorithm's iterative update calculations. The specific calculation process of the multi-objective particle swarm optimization algorithm is state-of-the-art and will not be detailed in this application.

[0080] It should be noted that this application improves the particle search strategy of the multi-objective particle swarm optimization algorithm and introduces the idea of ​​nonlinear inertial flight, that is, the inertia weight of the particles changes nonlinearly with the position during the migration process. Its advantage is that when the swarm finds a better solution, it can accelerate the surrounding particles to approach it, improve the convergence speed, and the global search ability of the particles decreases as the iterative process proceeds, while the local search ability increases, which promotes the convergence speed and improves the optimization rate of the algorithm to a certain extent. The calculation formula for the update of the nonlinear inertia weight w is: w min is the preset minimum weight, w max is the preset weight maximum value, d is the distance between the particle and the optimal solution, T max is the maximum distance of the population.

[0081] Since it is difficult to obtain a unique solution that makes each objective function optimal in multi-objective optimization, the result of multi-objective optimization is a solution set consisting of multiple optimal solutions that weigh each objective function to make it optimal overall. In one embodiment, when the first objective function and the second objective function are both as small as possible, and the function values ​​of the first objective function and the second objective function fluctuate within a predetermined range, it is determined that the comprehensive objective function has reached the optimum. The function value of each objective function is as small as possible, which means that it is within the specified range of the minimum value of the objective function, and it does not necessarily have to be the minimum value. Among them, the predetermined range of the function values ​​of the first objective function and the second objective function is set according to the actual application situation, and the goal is to make the objective function values ​​calculated in each iterative process remain stable and not undergo large jumps.

[0082] by Figure 2 Taking the iteration process of the objective function value of the first objective function shown in the figure as an example, as the number of iterations increases, the value of the wind-induced tilt angle decreases and gradually approaches 0. The zero-order moment, standard deviation, and significant value of the wind-induced tilt angle all become increasingly stable. Therefore, it can be determined that the first objective function reaches its optimal value when the number of iterations is 30.

[0083] Each set of main scale parameters in the main scale parameter set calculated using the multi-objective particle swarm optimization algorithm is superior to the main scale parameters of the floating foundation prototype. Subsequent designers can screen the main scale parameter set to obtain the final floating foundation main scale optimization design scheme.

[0084] The above description is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and variations directly derived or imagined by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the scope of protection of the present application.

Claims

1. A method for optimizing the main dimensions of a floating foundation of a floating wind turbine, characterized in that: The floating foundation main dimension optimization design method includes: determining target sea state parameters, determining principal dimension parameters and weight distribution parameters of a master template of the floating foundation to be optimized, and determining a variation range of the principal dimension parameters of the floating foundation based on the principal dimension parameters of the master template; the weight distribution parameters indicating the weight distribution of various components of the floating foundation; Randomly generate multiple sets of candidate main scale parameters within the variation range of the floating foundation main scale parameters, and construct multiple floating foundation models using a parametric modeling method; Determining candidate weight distribution parameters corresponding to each set of candidate principal scale parameters based on the parent model principal scale parameters and the parent model weight distribution parameters; calling a hydrodynamic analysis module to perform hydrodynamic calculations on each floating foundation model using each set of candidate weight distribution parameters as input to obtain multiple sets of hydrodynamic performance indicators; the hydrodynamic performance indicators include wind-induced heel angle, natural period, and short-term forecast; The comprehensive objective function for optimizing the main dimensions of the floating foundation is determined based on wind-induced heel angles and short-term forecasts. The comprehensive objective function characterizes the stability of the floating foundation model. The displacement of the floating foundation model is determined based on candidate weight distribution parameters. The constraints for optimizing the main dimensions of the floating foundation are determined based on the relationship between the natural period and the spectral peak period of the target sea state parameters, based on the condition that the displacement of the floating foundation model meets the design requirements. Under the constraints, a multi-objective particle swarm optimization algorithm is used to solve the main scale parameter set that makes the comprehensive objective function reach the optimal value, and the main scale optimization design scheme of the floating foundation is obtained.

2. The floating foundation main dimension optimization design method according to claim 1, characterized in that: The comprehensive objective function includes a first objective function f that minimizes the wind-induced tilt angle. w (x), and the second objective function f that minimizes the short-term forecast r (x); determine the comprehensive objective function F(x) = (f w (x),f r (x)) T ; Where x is the candidate main scale parameter vector and x=(x1,x2,...,x j ,...,x n ) T , x j is the jth principal scale variable in the candidate principal scale parameter vector x and x jmin ≤x j ≤x jmax , x jmin is the lower limit of the change interval of the jth main scale variable, x jmax is the upper limit of the variation interval of the jth main scale variable. Each component of the floating basic model includes several main scale variables. n is the total number of main scale variables. The integer parameter n≥1, (·) T Represents matrix transpose.

3. The floating foundation main dimension optimization design method according to claim 2, characterized in that: The hydrodynamic calculation of the floating foundation model is performed to obtain the hydrostatic stiffness matrix k of the floating foundation model. The wind-induced inclination angle is determined according to the hydrostatic stiffness matrix k. According to the wind-induced inclination angle g w Determine the first objective function f w (x) = g w ; Where c is the dimensionless ultimate wind load, and [k] represents the hydrostatic stiffness value of the degree of freedom corresponding to the wind direction in the hydrostatic stiffness matrix k of the floating foundation model.

4. The floating foundation main dimension optimization design method according to claim 2, characterized in that: The hydrodynamic calculation of the floating foundation model is performed to obtain the RAO motion response data of the floating foundation model at any degree of freedom i. i (ω), according to the RAO motion response data RAO of degree of freedom i i (ω) Determine the short-term prediction of the floating foundation model in degree of freedom i by combining the target sea state parameters Short-term forecast based on degrees of freedom i Determine the second objective function Where S(ω) is the wave spectrum of the target sea state parameters, m is the total number of degrees of freedom, and the integer parameter m≥1.

5. The floating foundation main dimension optimization design method according to claim 1, characterized in that: The hydrodynamic calculation of the floating foundation model is performed to obtain the motion period data of the floating foundation model, and the natural period of the degree of freedom i is determined based on the motion period data. Determine the constraints for the optimal design of the main dimensions of the floating foundation: Among them, Δ x is the displacement of the floating foundation model corresponding to the candidate main scale parameter vector x, Δ max is the upper threshold of the water discharge, T p is the spectrum peak period of the target sea state parameter.

6. The floating foundation main dimension optimization design method according to claim 1, characterized in that: Determining the candidate weight distribution parameters corresponding to the candidate principal scale parameters includes: Determine the weight of the master form of each component of the floating foundation model according to the master form weight distribution parameters; Determine the volume change of each component of the floating foundation model according to the change of the candidate main scale parameters relative to the main scale parameters of the parent model; Based on the master weight of each component of the floating foundation model and according to the mapping relationship between the weight change and volume change of each component of the floating foundation model, the weight change of each component of the floating foundation model is determined to obtain candidate weight distribution parameters.

7. The floating foundation main dimension optimization design method according to claim 6, characterized in that: For any k-th component of the floating foundation model, according to the volume change ΔV of the k-th component k Determine the weight change of the kth component in, is the volume of the matrix of the kth component, is the parent weight of the kth component.

8. The floating foundation main dimension optimization design method according to claim 5, characterized in that: The motion cycle data includes a column of natural cycle data and multiple columns of motion cycle data with different degrees of freedom. The natural cycle of degree of freedom i is determined according to the motion cycle data. include: The motion cycle data of degree of freedom i and the natural cycle data are subtracted column by column to obtain the difference data column, and the motion cycle data corresponding to the minimum value in the difference data column is selected as the natural cycle of degree of freedom i 9. The floating foundation main dimension optimization design method according to claim 2, characterized in that: When the first objective function and the second objective function are both as small as possible and the function values ​​of the first objective function and the second objective function fluctuate within a predetermined range, it is determined that the comprehensive objective function reaches the optimum.

10. The floating foundation main dimension optimization design method according to claim 1, characterized in that: The hydrodynamic analysis module adopts the aqwa-line module; The parametric modeling method comprises: The candidate principal scale parameters are constructed as a floating foundation model based on the APDL command stream modeling method to obtain an .aqwa input file, which includes the model parameters of the floating foundation model; the candidate weight distribution parameters in the .aqwa input file are automatically modified by a Python script to generate a .dat calculation file adapted for the AQWA-Line module; The calling of the hydrodynamic analysis module includes: Based on the Python script, batch commands are generated to automatically call the aqwa-line module to perform hydrodynamic calculations and generate a .lis hydrodynamic calculation result file, which includes the hydrostatic stiffness matrix, motion period data, and RAO motion response data of the floating foundation model.

Images