Floating fan anchor chain optimization method and device based on genetic algorithm and medium

Abstract

The invention relates to the field of ocean energy development, and discloses a floating fan anchor chain optimization method and device based on a genetic algorithm and a medium, and the method comprises the steps: obtaining the installation data of a floating fan system; based on the installation data of the floating fan system, a full-coupling numerical model of the mooring system is established by adopting CFD software; a genetic algorithm is adopted to optimize the full-coupling numerical model, and mooring optimization parameters are obtained; the mooring optimization parameters are applied to actual installation of the floating type draught fan system; according to the method, the high-precision full-coupling numerical model is established through the advanced CFD technology, the performance of the floating fan system in the complex marine environment is comprehensively and accurately predicted and evaluated, meanwhile, the mooring system is optimized through the genetic algorithm, the optimal parameter combination is found, the stability and reliability of the system are greatly improved, and the method is suitable for popularization and application. Powerful support is provided for application and development of the floating fan technology under deep sea and severe sea conditions, and remarkable beneficial effects are achieved.

Description

Technical Field

[0001] The present invention relates to the field of ocean energy development, and particularly to an optimization method for the mooring chain of a floating wind turbine based on a genetic algorithm. Background Art

[0002] In the current state of technological development, the design and optimization of floating wind turbine systems face numerous challenges. On the one hand, it is difficult to accurately simulate the motion response and load conditions of the wind turbine system in a complex ocean environment. Traditional numerical models and simulation methods often struggle to comprehensively consider various influencing factors, resulting in deviations between predicted results and reality. On the other hand, the design and optimization of the mooring system lack efficient and accurate algorithm support, making it difficult to find the optimal balance among multiple parameters and restricting the performance improvement of the floating wind turbine system.

[0003] The existing technical deficiencies are mainly reflected in the accuracy of the numerical model and the limitations of the mooring system optimization method. Numerical models often simplify too much and cannot accurately reflect the complexity and dynamics of the actual ocean environment; while the optimization of the mooring system relies on empirical formulas or trial-and-error methods, which are inefficient and difficult to guarantee a globally optimal solution. These problems limit the application and development of floating wind turbine systems in deep waters and harsh sea conditions. Summary of the Invention

[0004] The purpose of the present invention is to propose an optimization method, device, and medium for the mooring chain of a floating wind turbine based on a genetic algorithm to solve the technical problems of system instability and low design efficiency existing in the current design process of floating wind turbine systems.

[0005] Specifically, an optimization method for the mooring chain of a floating wind turbine based on a genetic algorithm provided by the present invention includes the following steps:

[0006] S1. Obtain the data of the floating wind turbine system installation;

[0007] S2. Based on the data of the floating wind turbine system installation, establish a fully coupled numerical model of the mooring system using CFD software;

[0008] S3. Optimize the fully coupled numerical model using a genetic algorithm to obtain mooring optimization parameters;

[0009] S4. Apply the mooring optimization parameters to the actual installation of the floating wind turbine system.

[0010] A storage medium stores instructions and data for implementing an optimization method for the mooring chain of a floating wind turbine based on a genetic algorithm.

[0011] A floating wind turbine mooring chain optimization device based on a genetic algorithm, comprising: a processor and the storage medium; the processor loads and executes the instructions and data in the storage medium to implement a floating wind turbine mooring chain optimization method based on a genetic algorithm.

[0012] The beneficial effects provided by the present invention are as follows: by using advanced CFD technology to establish a high-precision fully coupled numerical model, the present invention comprehensively and accurately predicts and evaluates the performance of the floating wind turbine system in a complex marine environment. At the same time, the genetic algorithm is used to optimize the mooring system to find the best parameter combination, greatly improving the stability and reliability of the system, providing strong support for the application and development of floating wind turbine technology in deep sea and harsh sea conditions, and having significant beneficial effects. Brief Description of the Drawings

[0013] Figure 1 is a schematic diagram of the process of the method of the present invention;

[0014] Figure 2 is a top view and side view of the structure of a three-buoy floating wind turbine in an embodiment of the present invention;

[0015] Figure 3 is a schematic diagram of the analysis and calculation of the mooring chain of a floating wind turbine in an embodiment of the present invention;

[0016] Figure 4 is a step diagram of the genetic algorithm for optimizing the mooring chain of a floating wind turbine in an embodiment of the present invention;

[0017] Figure 5 is a flowchart of the genetic algorithm in an embodiment of the present invention;

[0018] Figure 6 is a schematic diagram of the operation of the hardware device in an embodiment of the present invention. Detailed Embodiments

[0019] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0020] Before formally elaborating on the present invention, the solution of the present invention will be generally described first for easy understanding.

[0021] Please refer to Figure 1 , a floating wind turbine mooring chain optimization method provided by the present invention includes:

[0022] S1. Obtain the data of the floating wind turbine system installation;

[0023] It should be noted that in step S1, the data of the floating wind turbine system installation includes: marine environmental condition data and floating wind turbine foundation data.

[0024] As an embodiment, the process of collecting ocean environmental condition data in the present invention is as follows: Set up multiple meteorological observation stations in the selected wind farm area. These stations should cover different geographical locations and altitude levels to obtain comprehensive meteorological data. Use professional meteorological measuring instruments, such as meteorological towers, anemometers, wind vanes, thermometers, hygrometers, etc., to monitor and record meteorological data in real time, including wind speed, wind direction, air temperature, air pressure, humidity, etc. The data collection frequency is set to once per minute to ensure the continuity and accuracy of the data. The collection time span should cover at least one complete seasonal cycle to obtain meteorological characteristics of different seasons. Conduct quality inspection and preprocessing on the collected data, and eliminate abnormal data and error data. Use statistical methods, such as mean filtering, median filtering, etc., to smooth the data and reduce noise interference.

[0025] Deploy professional measuring equipment such as current profilers and acoustic Doppler current profilers (ADCP) to measure the current velocity at different positions and depths in the wind farm area. The selection of measurement points should consider factors such as the topography and bathymetry changes in the wind farm area to ensure that the measurement results can represent the current situation of the entire wind farm area. The data collection frequency is consistent with the meteorological data, and data quality inspection and preprocessing are also carried out to ensure the accuracy and reliability of the data.

[0026] Use equipment such as multibeam echosounders and single-beam echosounders to measure the water depth in the wind farm area. The measurement route should cover the entire wind farm area to ensure no omission. For complex terrain areas, increase the measurement density to improve the measurement accuracy. Conduct post-processing on the measurement data to generate a bathymetric map, providing basic data for subsequent numerical simulations.

[0027] As an embodiment, the process of collecting basic data of floating wind turbines in the present invention is as follows:

[0028] Collect the design drawings and technical specifications of components such as wind turbines, towers, and floating platforms to obtain detailed information such as the dimensions, weights, and material properties of each component. Conduct actual measurements and weighings on key components to verify the accuracy of the design data. For example, use a laser rangefinder to measure the length and diameter of wind turbine blades, and use an electronic scale to weigh the towers and platforms. Conduct experimental tests on material properties, such as the strength, stiffness, density, etc. of the materials, to ensure the reliability of the material property data.

[0029] Collect information such as the model, specifications, length, diameter, and material of the mooring chain, as well as data such as the type, weight, and dimensions of the mooring anchor. Measure parameters such as the actual length, tension, and bending stiffness of the mooring chain to ensure the accuracy of the initial parameters. Investigate the installation method and connection details of the mooring system to provide accurate basic data for subsequent numerical simulations.

[0030] Specifically, the model can select the open-source OC4-DeepCwind semi-submersible three-buoy platform and the offshore 5MW wind turbine model of the National Renewable Energy Laboratory (NREL) (see Figure 2 ).

[0031] Parameters of the floating platform:

[0032] Total draft: 20.0m; Distance between the centerlines of the side buoys: 50.0m; Radius of the side buoy: 26.0m; Height of the lower buoy: 6.0m; Radius of the main buoy: 6.5m; Radius of the buoy: 12.0m; Diameter of the buoy: 24.0m; Diameter of the support: 1.6m; Total mass of the floating wind turbine platform: 1.35×10 7 kg.

[0033] Parameters of the wind turbine:

[0034] Rated power: 5MW; Upwind wind turbine; Number of blades: 3; Rotor diameter: 126m; Hub diameter: 3m; Hanging height (top height of the tower): 90m; Cut-in and cut-out wind speeds: 3m / s, 25m / s; Rated wind speed: 11.4kg / m; Cut-in and cut-out rotational speeds: 6.9rpm, 12.1rpm; Rated tip speed: 80m / s; Distance from the hanging point to the central axis of the tower: 5m.

[0035] Parameters of the mooring chain:

[0036] Number of mooring chains: 3; Angle between adjacent lines: 120°; Depth of the anchor point below SWL: 200m; Unstretched length: 835.5m; Cross-sectional diameter of the anchor chain: 0.0766m; Equivalent mooring mass density: 113.35kg / m; Equivalent mooring chain tensile stiffness: 7.536N; Equivalent mooring chain mass in water: 108.63kg / m; Normal drag coefficient: 2.00; Tangential drag coefficient: 0.40; Normal added mass coefficient: 0.80; Tangential added mass coefficient: 0.25.

[0037] S2. Based on the data of the floating wind turbine system installation, use CFD software to establish a fully coupled numerical model of the mooring system;

[0038] It should be noted that step S2 is specifically as follows:

[0039] S21. Based on the basic data of the floating wind turbine, use a three-dimensional modeling system to establish a geometric model of the floating wind turbine system and its mooring system;

[0040] As an embodiment, the present invention uses professional 3D modeling software (such as SolidWorks, CATIA, etc.) to establish an accurate geometric model based on the collected detailed basic data of the floating wind turbine system and its mooring system. A detailed geometric description of each component is made, including the shape of the wind turbine blades, the structural form of the tower, the geometric profile of the floating platform, and the connection method of the mooring chain. To ensure that the accuracy of the geometric model meets the requirements of numerical simulation, the key parts are locally enlarged and refined, such as the tip of the wind turbine blade, the connection node of the mooring chain, etc.

[0041] S22, importing the geometric model into CFD software for meshing;

[0042] As an embodiment, the present invention imports the established geometric model into CFD software (such as OpenFOAM), and uses a suitable meshing method (such as structured mesh, unstructured mesh or hybrid mesh) to mesh the computational domain. In areas where the flow field changes drastically, such as wind turbine blades and tower surfaces, high-density meshing is used to improve the calculation accuracy; in areas far away from the object, a relatively sparse mesh is used to reduce the amount of calculation. The mesh quality is checked and optimized to ensure the orthogonality, smoothness and continuity of the mesh, and to avoid problems such as mesh distortion and negative volume.

[0043] Specifically, the platform operates at a water depth of 200m, and the overall dimensions are set to -200m to 300m in the x-direction, -200m to 200m in the y-direction, and -180m to 180m in the z-direction.

[0044] Use snappyHexMeshDict, the meshing tool provided by OpenFOAM. Step 1: Create a hexahedral background mesh area.

[0045] Use the snappyHexMeshDict file for mesh refinement. The refinement process is as follows: Define the geometric region in the geometry dictionary. In this numerical simulation experiment, two three-dimensional geometric models (triSurfaceMesh) are defined, which store the meshes of the semi-submersible floating wind turbine. Encrypt the meshes of the geometric region in the refinmentRegions dictionary. Second step: Use the castellatedMeshControls sub-dictionary method for cell division. Ensure the effectiveness and uniformity of the division by controlling the maximum number of cells during refinement. Third step: Use the surfaceFeature tool to extract the feature edges of the stl files in the triSurface folder and generate the.eMesh file. The size of the stl rigid body can be scaled up or down proportionally during the generation process. Fourth step: Set castellatedMeshControls. There are three settings for the position of the rigid body encryption region in the snappyHexMeshDict, namely: inside, outside, and distance. Refine all the meshes within the geometric region, and the geometric region must be a closed three-dimensional structure, with 4-level refinement encryption.

[0046] S23. Set up a fully coupled numerical model in the CFD software. The fully coupled numerical model includes: an aerodynamic load calculation model, a hydrodynamic load calculation model, and a mooring load calculation model.

[0047] It should be noted that the calculation models of aerodynamic loads, hydrodynamic loads, and mooring loads are set in the CFD software.

[0048] Among them, for the aerodynamic load, a suitable turbulence model (such as the RANS k-ε model) and aerodynamic theory are used to calculate the aerodynamic forces on the wind turbine blades at different wind speeds and wind directions.

[0049] As an example, the aerodynamic load calculates the lift and drag through the blade element momentum theory, combines the driving line model to convert the blade forces into the volume forces of the flow field, and uses the Gaussian weight function to achieve the spatial distribution of the forces. To simplify the aerodynamic load, the "one command, one movement" method is used to calculate the aerodynamic load. The process is as follows:

[0050] 1. Time Discretization and Module Initialization. In this module, the time list is checked. This verification process ensures that the program time list corresponds precisely to the numerical simulation time step, thus maintaining the integrity and accuracy of the simulation timeline. The line of action method is used to obtain the wind load. Once the wind load is obtained, it is dispersed at fixed time steps. This dispersion process is carefully calibrated to ensure that the wind load is applied evenly and accurately over time, mimicking real-world wind conditions as closely as possible. The initialization process includes several key tasks, including background grid construction, grid surface feature extraction, STL model grid generation, automatic grid fitting and refinement, environmental field setting, and determination of the initial wind turbine position. Each task is carefully executed to ensure that the simulation environment is correctly set up and the wind turbine is accurately placed in the simulation space.

[0051] 2. Data Input. The data input module is designed to handle the input of various data sources required for the simulation. The simulation module reads the output file according to a predefined fixed format. This fixed format ensures the consistency and compatibility of the data generated by different components of the simulation system. After one simulation time step, the simulation module regenerates the output file for the current time. This regeneration process is crucial as it allows for the continuous update and tracking of simulation results over time. The data input file consists of two main parts. The first part includes the wind load calculated by an external program at different time periods according to the line of action theory. This wind load data is generated independently but integrated into the simulation process to accurately represent the wind force acting on the structure. The second part includes the data calculated by the numerical simulation at the last time step. By incorporating the data from the previous time step, the simulation can continue based on the previously obtained results, ensuring the continuity and accuracy of the simulation.

[0052] 3. Data Output. The data output module is responsible for recording and storing the key data generated during the simulation. This module plays a crucial role in recording the simulation results and enabling further analysis and verification. During the simulation, various physical parameters are recorded in the output data file. These parameters provide a comprehensive description of the system's behavior over time. During the rigid body motion simulation, due to the applied forces, the grid deforms. However, the grid topology remains unchanged, ensuring the structural integrity of the simulation model. Therefore, the grid ID remains constant throughout the simulation. By using the position coordinates of the same grid ID number at different times, the position of the wind turbine center coordinates can be accurately determined at each time step. This method ensures the correct use of virtual forces to describe the forces on the wind turbine at each time step, maintaining the accuracy and reliability of the simulation results.

[0053] 4. Numerical Simulation. The time step for the numerical simulation is set to 0.02 seconds, which provides a fine resolution for the simulation process. This relatively small time step allows for a more accurate capture of the dynamic behavior of the FOWT. The time step for the additional wind load is set to 0.5 seconds as one simulation step. This specific wind load time step is chosen based on a balance between computational efficiency and accuracy. It ensures that the wind load is applied at appropriate time intervals without sacrificing the overall accuracy of the simulation. The output data file is used to record the physical quantities during the simulation. This file serves as a repository for the simulation results and contains valuable information about the system's state at different time steps. At each time step, the mooring restraint force, the magnitude and direction of the additional aerodynamic load, and the acting position of the aerodynamic load are read using the interface provided by the data input module. This data retrieval process is crucial as it ensures that the simulation has access to the latest information about the external forces acting on the FOWT. Meanwhile, the data output module is used to input the numerical simulation data into the numerical simulation of the next time step. This seamless data transfer between time steps ensures the continuity and consistency of the simulation process, allowing for an accurate and efficient calculation of the FOWT's response over time.

[0054] For hydrodynamic loads, considering the effects of factors such as waves and ocean currents, potential flow theory, boundary element method, or computational fluid dynamics method is used to calculate the motion response and force conditions of the floating platform under hydrodynamic action. Hydrodynamic loads cover wave excitation forces. The Pierson-Moscowitz spectrum is used to generate irregular waves, and a random wave field is simulated through linear wave superposition. The velocity inlet boundary condition defines the axial, tangential velocity, and pressure distribution of the waves, and the interaction between the fluid and the structure is analyzed in combination with the turbulence model.

[0055] For mooring loads, according to the mechanical properties and connection methods of the mooring chain, a catenary model or other suitable models are used to calculate the tension and deformation of the mooring chain, and then the mooring force is determined.

[0056] The motion of the floating wind turbine (FOWT) is calculated by a rigid body motion solver, considering the relative motion relationship of translation and rotation. The dynamic modeling of the anchor chain is based on the strain-tension relationship, and its spatial position and force state are analyzed through a discretization method. The external forces include added mass force (fluid inertia), gravity and buoyancy, seabed contact force, and drag force, where the drag force is calculated by the Morison equation, and the contact force considers dynamic friction and stiffness characteristics. The calculation method for the rigid body motion of the wind turbine is as follows:

[0057]

[0058] Among them, F M represents the mooring force, r cs and r cmrespectively represent the distance vectors from the centroid to the mooring connection point and the center of each surface panel. The rigid body motion solver includes a body-fixed coordinate system and a fixed left coordinate system, and simulates the translation and rotation of the blade through the relative motion relationship between these two coordinate systems.

[0059] The constitutive model of the mooring chain is based on the strain-tension relationship, as Figure 3 shown. For a mooring chain with length L c , the global coordinate position vector of the mooring chain is expressed as r = [r1(s), r2(s), r3(s)] T , where s is the local mooring chain length (s ∈ [0, L c ). The motion equation of the anchor chain can be written as:

[0060]

[0061] where m0 is the mass per unit length of the mooring chain, T is the tension of the mooring chain, τ is the tangential unit vector of the mooring chain, and f represents all external forces. According to the position r of the mooring chain, the spatial discrete position The elongation (ε) of the anchor chain is calculated as:

[0062]

[0063] The calculation formula for the external force is:

[0064] f = f a + f b + f c + f d

[0065] where f a is the added mass force and the Froude-Krylov force, f b is the resultant force of gravity and buoyancy, f c is the contact force between the anchor chain and the seabed, f d is the drag force.

[0066] The added mass force includes hydrodynamic force and inertial force, and is calculated by the Morison equation.

[0067]

[0068] where C Mn and C Mt are the vertical and parallel coefficients; v f is the fluid velocity; A c is the projected area of the anchor chain. The buoyancy is written as:

[0069] f b = g(ρ c - ρ f ) / ρc m0

[0070] where g is the gravitational acceleration vector; ρ c is the density of the anchor chain. The contact force (f c ) is written as:

[0071]

[0072] where xy and z are the projections in the horizontal and vertical directions; r z is the vertical coordinate at the position of the mooring chain; z G is the vertical position, K G is the stiffness, ξ G is the critical damping ratio; μ is the dynamic friction coefficient; v μ is the velocity corresponding to the maximum friction coefficient. The viscous drag (f d ) can be expressed as:

[0073]

[0074] where, C dn and C dt are the vertical and parallel drag coefficients; v d is the relative velocity between the fluid and the anchor chain.

[0075] In addition, the present invention also establishes a three-dimensional numerical model for the pool environment where the wind turbine is located, and the specific process is as follows:

[0076] In the numerical pool, based on the incompressible Navier-Stokes (N-S) equations and the continuity equation, the flow characteristics of the ocean environment are described. Using the finite volume method or other suitable numerical discretization methods, the partial differential equations are transformed into algebraic equations, and the physical quantities such as the velocity and pressure of the flow field are obtained by iterative solution. Appropriate boundary conditions, such as the no-slip boundary condition, the free surface boundary condition, etc., are adopted to simulate the boundary conditions of the real ocean environment.

[0077] The volume of fluid (VOF) method is used to simulate the free liquid surface of the wave. By tracking the interface between different phases (water and air), the undulation and propagation process of the wave are accurately described. According to the actual sea conditions, the parameters of the wave, such as wave height, wavelength, wave period, wave direction, etc., are set to generate waves that conform to the actual situation. The wave simulation results are verified and calibrated. By comparing and analyzing with experimental data or actual observation data, the wave simulation parameters are adjusted to improve the simulation accuracy.

[0078] Specifically, based on the OpenFOAM program, the incompressible Reynolds-averaged Navier-Stokes (RANS) equations are used to simulate gas-liquid two-phase flow. The governing equations include the continuity equation, the momentum equation, and the volume-of-fluid transport equation for viscous fluids, with a focus on analyzing the free surface boundary conditions. The velocity and pressure fields are coupled through the PIMPLE algorithm, and the k-ε turbulence model is used to capture the characteristics of fluid flow. The volume transport parameter (F value) is introduced in the simulation to distinguish the water domain, the air domain, and the free surface, in order to study the influence of complex hydrodynamic environments such as waves and tides on the flow field. The governing equations include the continuity equation, the momentum equation, and the volume-of-fluid transport equation for incompressible viscous fluids, as follows:

[0079]

[0080] where t represents time, Ω(t) represents the fluid domain at time t, and the superscript f indicates that the variable belongs to the fluid domain. σ f , ρ f and f f represent the Cauchy stress, density, and physical tensor, respectively. Meanwhile, u f represents the time-averaged velocity in the flow field. F refers to the volume transport parameter, F = 1 represents the water domain, F = 0 represents the air region, and 0 < F < 1 represents the free surface boundary between the water domain and the air region.

[0081] On the free surface, the boundary conditions for velocity and pressure are expressed as follows:

[0082]

[0083] where τ and n represent the tangential and normal directions, respectively, and represent the tangential and normal velocities at the boundary, respectively. p f and p0 represent the pressure parameters in the water domain and the air region, respectively; μ f is the dynamic viscosity coefficient in the water domain. In this study, the PIMPLE algorithm is used to couple velocity and pressure. This algorithm accelerates the simulation speed and achieves a good convergence rate for the under-relaxation time between time steps. The κ-ε turbulence model is used in the simulation. The motion states and characteristic parameters of the wind turbine under different variables of the mooring chain are extracted through numerical simulation.

[0084] S3. Optimize the fully coupled numerical model using the genetic algorithm to obtain the optimized mooring parameters;

[0085] It should be noted that step S3 is specifically as follows:

[0086] S31. Taking the mooring chain length, the wet weight of the mooring chain, and the hanging height of the mooring chain from the platform as optimization parameters, construct the solution space and set the training set range;

[0087] Specifically, the present invention first defines the objective function, sets the constraint conditions to define the scope, selects a suitable coding method, initially constructs the population according to the rules, and initializes the parameters.

[0088] Maximize the suppression of the motions of the surge, pitch, and heave degrees of freedom of the floating wind turbine system, with the mooring chain length L, the wet weight W of the mooring chain, and the suspension height H between the mooring chain and the platform as the optimization parameters. According to the actual working environment and design requirements of the floating wind turbine, determine the value ranges of the respective optimization parameters.

[0089] For example, the value range of the mooring chain length is 80 - 200 m, the value range of the wet weight of the mooring chain is 15 - 35 kg / m, and the value range of the suspension height between the mooring chain and the platform is 5 - 15 m. Real number coding is adopted, and each individual is represented as a chromosome L, W, H containing three genes. Randomly generate N individuals in the solution space to form the initial population P(0). For example, for a population size N = 20, the following initial population can be generated:

[0090]

[0091] The initial population P(0), which contains 20 individuals, and each individual consists of three genes (i.e., the optimization parameters). L i (Mooring chain length): i = 1, 2, …, 20, representing the mooring chain length parameter of the i-th individual. This parameter is randomly generated in the solution space, and its value range is usually determined according to the background knowledge or prior conditions of the actual problem. For example, if the reasonable range of the mooring chain length (L min, L max ) is known, then the value here is any real number within this interval, which represents the specific length value of the mooring chain and is one of the important factors affecting the surge, pitch, and heave motions of the floating wind turbine. W i (Wet weight of the mooring chain): Similarly, i = 1, 2, …, 20, corresponding to the wet weight parameter of the mooring chain of the i-th individual. This is also a real number randomly generated within the given value range (W min, W max ). The wet weight of the mooring chain affects the mass and inertial characteristics of the entire mooring system, and thus has an impact on the motion state of the wind turbine. A larger wet weight may make the pitch and heave motions of the wind turbine relatively stable, but at the same time, other factors such as increasing the load and cost of the system need to be considered. H i (Suspension height between the mooring chain and the platform): For the i-th individual, H i represents the parameter of the suspension height between the mooring chain and the platform, and its value is within the range (H min, H max) is randomly generated within this range. This parameter is directly related to the position and attitude of the floating wind turbine in water, and a reasonable suspension height helps reduce the adverse movement of the wind turbine. For example, a higher suspension height may cause the wind turbine to be subjected to a smaller wave impact force in certain sea conditions, thereby reducing the amplitudes of surge, pitch, and heave motions. The crossover probability P c is taken as 0.8, and the mutation probability P m is taken as 0.05.

[0092] S32. According to the range of the training set, conduct numerical simulations of the floating wind turbine motion to obtain the motion state characteristic values and quantification results of the surge, pitch, and heave of the floating wind turbine system under different parameters;

[0093] According to the training set range (i.e., the range of mooring chain length, mooring chain wet weight, and the suspension height between the mooring chain and the platform), using the established numerical model of the floating wind turbine, considering various forces acting on the wind turbine such as wind force, wave force, and current force, apply kinematic and dynamic equations to solve the surge, pitch, and heave responses of the wind turbine under different parameters, and obtain the wind turbine motion data.

[0094] Take the quantification index for suppressing the surge, pitch, and heave motions of the floating wind turbine system as the fitness function. A common form is to take the reciprocal of the weighted sum of the surge, pitch, and heave standard deviations as the fitness function value. The formula is as follows:

[0095]

[0096] Among them, f(x) is the fitness function value of individual x; w1, w2, and w3 are the weights of surge, pitch, and heave respectively, which can be adjusted according to actual needs; σ s , σ p , σ h are the standard deviations of surge, pitch, and heave respectively.

[0097] After numerical simulation, the surge displacement data s of the i-th individual within M time steps is obtained i1 , s i2 , …, s iM , the pitch angle data θ i1 , θ i2 , …, θ iM , and the heave displacement data h i1 , h i2 , …, iM . Then the formula for calculating the surge standard deviation σ si is:

[0098]

[0099] The pitch standard deviation σ θi and the heave standard deviation σ hiThe calculation method is similar.

[0100] Furthermore, through the roulette wheel selection method, simulating the gambling roulette mechanism, the probability of an individual being selected is determined according to the individual fitness proportion, realizing the screening and reproduction of the population.

[0101] Calculate the sum of the fitness function values of each individual and the proportion of the individual fitness value in the total fitness value Determine the probability of each individual being selected according to the proportion, and then use the roulette wheel selection method to select the parental individuals. For example, generate a random number r ∈ [0, 1]. If it satisfies p1 + p2 + … + p i < r < p1 + p2 + … + p j , then select the j-th individual as one of the parental individuals. Repeat this process until the required number of parental individuals is selected.

[0102] S33. Conduct iterative search within the solution space, continuously evolving the population through selection, crossover, and mutation operations to approach the optimal solution; in each iteration, evaluate the fitness function of the genetic algorithm according to the quantization result of the fan motion response to ensure the correctness of the optimization direction.

[0103] The present invention uses the uniform crossover method to randomly select gene segments of the parental chromosomes with equal probability and combine them into new offspring chromosomes to introduce gene diversity and improve the global search ability.

[0104] According to the crossover probability P c = 0.8, select two parental individuals x1 = L1, W1, H1 and x2 = L2, W2, H2, and perform crossover operations on each gene locus. For the k-th gene locus (k = 1, 2, 3, corresponding to L, W, H respectively), generate a random number α ∈ [0, 1]. If α < P c , then exchange the values of the two parental individuals at this gene locus according to the following formula:

[0105]

[0106] The resulting offspring individuals after crossover are x′1 = L′1, W′1, H′1 and x′2 = L′2, W′2, H′2.

[0107] Furthermore, through the real number mutation method, slightly perturb the real number genes of the individuals within a certain range to introduce mutation to increase the diversity of the population.

[0108] According to the mutation probability P mCheck each gene locus of each offspring individual. For the k-th gene locus of the i-th offspring individual (k = 1, 2, 3, corresponding to L, W, H respectively), generate a random number β ∈ [0, 1]. If β < P m , then perform a small random mutation within the feasible region of this gene locus. For example, the mutated gene value can be calculated according to the following formula:

[0109] y k = x i ' k + η·N(0, 1)

[0110] where x′ ik is the original value of the i-th offspring individual at the k-th gene locus, η is a small mutation step size (which can be adjusted according to the scale of the problem), and N(0, 1) represents a standard normal distribution random number with a mean of 0 and a variance of 1.

[0111] Finally, through iterative search and optimization, continuously verify the direction and improve the quality of the solution to approximate the global optimal solution.

[0112] 1. Iteratively update the population: Use the new population generated after selection, crossover, and mutation operations as the next generation population P(t + 1), and repeat S402 to S405 to continuously perform iterative search to approximate the optimal solution.

[0113] 2. Judge the termination condition: Set the maximum number of iterations T, and stop the algorithm when the maximum number of iterations is reached. Or, if the best fitness values of consecutive generations of the population no longer change significantly (the change amount is less than a certain threshold), it is also considered that the algorithm converges and stops the iteration.

[0114] 3. Verify the optimization direction: In each iteration, evaluate the fitness function of the genetic algorithm according to the quantization result of the wind turbine motion response to ensure the correctness of the optimization direction. If it is found that the optimization direction deviates from the expectation, the weight parameters of the fitness function can be adjusted or the genetic operation process can be re-examined to guide the algorithm to continue evolving towards the goal of reducing the surge, pitch, and heave motions of the floating wind turbine. Through the genetic algorithm, the optimized mooring system parameters are obtained, and the flowchart is as Figure 5 shown.

[0115] S4. Apply the mooring optimization parameters to the actual installation of the floating wind turbine system;

[0116] It should be noted that after being optimized by the genetic algorithm, the obtained mooring optimization parameters are applied to the actual installation and commissioning of the floating wind turbine system. During the actual operation process, the motion responses of the floating wind turbine system are monitored and evaluated in real time, and data is collected and analyzed to verify the effectiveness of the optimized mooring system in actual applications. If any situation that does not meet the expectations is found, the mooring system is fine-tuned according to the feedback in actual applications to ensure that it achieves the best motion suppression performance.

[0117] After being optimized by the genetic algorithm, the obtained mooring parameters. According to the design drawings, the fixed points of the mooring device are installed at the predetermined positions on the offshore wind turbine platform. For the anchoring foundation, if a gravity anchor is used, it is necessary to ensure that it accurately falls into the predetermined seabed position and the verticality deviation is controlled within a very small range (for example, the verticality deviation is less than α degrees).

[0118] The mooring line length parameter L obtained according to the genetic algorithm optimization opt (assuming L opt = 200m), the wet weight W of the anchor chain opt (assuming W opt = 25 Kg / m), the suspension height H of the anchor chain opt (assuming H opt = 25m), the mooring line is laid by a professional mooring line laying device. During the laying process, a measuring instrument is used to monitor the length of the mooring line in real time to ensure that the error between the actual laying length and the design length is within the allowable range (such as the error does not exceed ±ε1L opt , where ε1 = 0.5%), that is, the error is controlled within ±1m.

[0119] Acceleration sensors installed at key parts of the wind turbine platform (such as the center of gravity position, blade root, etc.) are used to collect the acceleration signals of the platform in three-dimensional space in real time. Let the components of the acceleration in the x, y, and z directions be a x 、a y 、a z , and the sampling frequency is f a (such as f a = 100Hz), then a set of acceleration data is collected every time interval Δt a = 1 / f a = 0.01s to form a time series data set. By analyzing the collected acceleration signals, the acceleration change situation of the platform at different times can be obtained, and then the vibration characteristics of the wind turbine platform can be evaluated.

[0120] Displacement sensors are installed at the connection between the mooring line and the platform and near the fixed points of the seabed foundation to measure the displacement of the platform in three-dimensional space. Similarly, taking the x, y, and z directions as an example, let the components of the displacement in the three directions be x, y, and z, and the sampling frequency is f d (such as fd = 50 Hz), and displacement data is collected every Δt d = 1 / f d = 0.02 s. By analyzing the displacement data, the position change trajectory of the wind turbine platform can be obtained to understand its movement trend under environmental loads.

[0121] An attitude sensor (such as a combination of a gyroscope and an inclinometer) is used to measure the changes in the roll angle θ(t), pitch angle φ(t), and yaw angle ψ(t) of the wind turbine platform over time. The sampling frequency is f p (such as f p = 20 Hz), and a set of attitude angle data is obtained every Δt p = 1 / f p = 0.02 s. The attitude data can help analyze the spatial attitude changes of the wind turbine platform and is of great significance for evaluating the stability control of the mooring system on the platform.

[0122] Devices such as an anemometer, a wind vane, and a wave height sensor are installed around the floating wind turbine system to record environmental parameters such as wind speed, wind direction, and wave height in real time. The measurement range of the anemometer for wind speed is v min -v max (such as 0 - 40 m / s), and the accuracy reaches ±∈ v v (assuming ∈ v = 5%); the measurement accuracy of the wind vane is ±∈ w degrees (such as ∈ w = 3°); the measurement range of the wave height sensor is h min- h max (such as 0 - 20 m), and the resolution is not less than Δh (such as Δh = 0.1 m). These environmental parameter data will be used as important reference factors for analyzing the motion response of the floating wind turbine system.

[0123] The collected data such as acceleration, displacement, and attitude is first filtered to eliminate noise interference. For example, a low-pass filter is used to process the acceleration signal, and the cut-off frequency is set to f c (such as f c = 30 Hz). Let the acceleration signal before filtering be a(t), and the acceleration signal after filtering be a(t), then its frequency domain expression is:

[0124]

[0125] where R is the resistance value, C is the capacitance value, and ω = 2πf. By adjusting the filter parameters, the filtered data can more truly reflect the actual motion state of the wind turbine system.

[0126] Calculate various characteristic indexes of the fan movement based on the filtered data. For example, calculate the average speed v of the fan platform within a certain period (such as one hour). avg 、the maximum speed v max 、the average acceleration amplitude a vag 、the maximum acceleration amplitude a max 、the movement period T motion and so on. Suppose the speed sample sequence collected within this period is {v1, v2, …, v n}, then the calculation formula for the average speed is:

[0127]

[0128] The maximum speed is the maximum value in the sample sequence:

[0129] v max =max{v1, v2, …, v n}

[0130] Similarly, relevant indexes of the acceleration amplitude can be calculated. For the calculation of the movement period, it can be determined by analyzing the periodicity of the displacement signal. For example, perform autocorrelation analysis on the displacement time series {x(t1), x(t2), …, x(t m ), and the time interval corresponding to the first zero point of the autocorrelation function obtained is the estimated value T estimated of the movement period. Then calculate the average value and standard deviation of the movement period through statistical analysis methods to evaluate the periodic stability of the movement.

[0131] Compare and analyze the actually monitored movement characteristic indexes with the preset expected values (determined according to the design and safety requirements of the fan system). For example, the expected movement amplitude envelope is within a certain range (suppose the maximum movement amplitude does not exceed A expected ). If the actually monitored maximum movement amplitude v max exceeds A threshold and the exceeding amplitude is greater than the set threshold (such as ΔA thresholdIf it is equal to 1m, it indicates that the motion suppression performance of the mooring system under this condition does not meet the expected requirements; or by comparing the probability distribution of the motion amplitudes before and after optimization (such as by calculating the probability density function), if there are significant differences between the two (such as the Kullback-Leibler divergence exceeding the set value), it also shows that the actual effect may not meet the expectations. At the same time, analyze the variation laws of the motion characteristics of the wind turbine system under different environmental conditions (such as different wind speed ranges and wave height ranges) to provide a basis for further fine-tuning the mooring system. During the actual operation process, finely adjust the length parameter of the mooring line and monitor it in real time through professional measuring instruments. At the same time, corresponding optimizations are also made for the wet weight and suspension height of the anchor chain. Based on accurate calculations and on-site actual situations, reasonably distribute the anchor chain load to ensure the stability, safety, and efficiency of the entire mooring system.

[0132] Please refer to Figure 6 , Figure 6 is a schematic diagram of the operation of the hardware device according to an embodiment of the present invention. The hardware device specifically includes: a floating wind turbine anchor chain optimization device 401 based on a genetic algorithm, a processor 402, and a storage medium 403.

[0133] A floating wind turbine anchor chain optimization device 401 based on a genetic algorithm: The floating wind turbine anchor chain optimization device 401 based on a genetic algorithm implements the floating wind turbine anchor chain optimization method based on a genetic algorithm.

[0134] Processor 402: The processor 402 loads and executes the instructions and data in the storage medium 403 to implement the floating wind turbine anchor chain optimization method based on a genetic algorithm.

[0135] Storage medium 403: The storage medium 403 stores instructions and data; the storage medium 403 is used to implement the floating wind turbine anchor chain optimization method based on a genetic algorithm.

[0136] The beneficial effects of the present invention are: The present invention establishes a high-precision fully coupled numerical model through advanced CFD technology to comprehensively and accurately predict and evaluate the performance of the floating wind turbine system in a complex marine environment. At the same time, the genetic algorithm is used to optimize the mooring system to find the best parameter combination, which greatly improves the stability and reliability of the system, provides strong support for the application and development of floating wind turbine technology in deep sea and harsh sea conditions, and has significant beneficial effects.

[0137] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A floating wind turbine mooring chain optimization method based on genetic algorithm, characterized in that: It includes the following steps: S1. Obtain the data for the installation of the floating wind turbine system; S2. Based on the data for the installation of the floating wind turbine system, use CFD software to establish a fully coupled numerical model of the mooring system; S3. Use the genetic algorithm to optimize the fully coupled numerical model to obtain mooring optimization parameters; S4. Apply the mooring optimization parameters to the actual installation of the floating wind turbine system.

2. The floating wind turbine anchor chain optimization method based on genetic algorithm according to claim 1, wherein: In step S1, the data for the installation of the floating wind turbine system includes: marine environmental condition data and floating wind turbine foundation data.

3. The floating wind turbine anchor chain optimization method based on genetic algorithm according to claim 2, characterized in that: Step S2 is specifically as follows: S21. Based on the floating wind turbine foundation data, use a 3D modeling system to establish a geometric model of the floating wind turbine system and its mooring system; S22. Import the geometric model into CFD software for mesh generation; S23. Set up a fully coupled numerical model in CFD software, and the fully coupled numerical model includes: an aerodynamic load calculation model, a hydrodynamic load calculation model, and a mooring load calculation model.

4. The floating wind turbine anchor chain optimization method based on genetic algorithm according to claim 3, characterized in that: In step S23, for the aerodynamic load calculation model, use a turbulence model and aerodynamic theory to calculate the aerodynamic forces on the wind turbine blades at different wind speeds and wind directions.

5. The optimization method for the mooring chain of a floating wind turbine based on a genetic algorithm according to claim 3, wherein: In step S23, for the hydrodynamic load calculation model, considering the influence of wave and current factors, use potential flow theory, boundary element method or computational fluid dynamics method to calculate the motion response and force conditions of the floating platform under hydrodynamic action.

6. The floating wind turbine mooring chain optimization method based on genetic algorithm according to claim 3, wherein: In the mooring load calculation model described in step S23, according to the mechanical characteristics and connection methods of the mooring chain, use a catenary model or other suitable models to calculate the tension and deformation of the mooring chain, and then determine the mooring force.

7. The floating wind turbine anchor chain optimization method based on genetic algorithm according to claim 1, characterized in that: Step S3 is specifically as follows: S31. Take the mooring chain length, mooring chain wet weight, and the suspension height of the mooring chain from the platform as optimization parameters, construct a solution space, and set the training set range; S32. According to the range of the training set, conduct numerical simulations of the floating wind turbine motion to obtain the motion state characteristic values and quantization results of the floating wind turbine system in the surge, pitch, and heave degrees of freedom under different parameters; S33. Conduct iterative search within the solution space, continuously evolve the population through selection, crossover, and mutation operations, and approach the optimal solution; In each iteration, evaluate the fitness function of the genetic algorithm according to the quantization results of the wind turbine motion response to ensure the correctness of the optimization direction.

8. The floating wind turbine anchor chain optimization method based on genetic algorithm according to claim 7, characterized in that: In step S33, the fitness function specifically uses the quantization index for suppressing the surge, pitch, and heave motions of the floating wind turbine system as the fitness function, as shown in the following formula: where f(x) is the fitness function value of individual x; w1, w2, and w3 are the weights of surge, pitch, and heave respectively; σ s , σ p , σ h are the standard deviations of surge, pitch, and heave respectively.

9. A storage medium, characterized in that: The storage medium stores instructions and data for implementing a floating wind turbine mooring chain optimization method according to any one of claims 1 to 8.

10. A floating wind turbine mooring chain optimization device based on a genetic algorithm, characterized in that: It includes: A processor and a storage medium; the processor loads and executes the instructions and data in the storage medium for implementing a floating wind turbine mooring chain optimization method according to any one of claims 1 to 8.

Images