A design method and system for structural lightweight of a wind turbine generator

Abstract

The application belongs to the field of wind power generation, and discloses a design method and system for lightweight structure of a wind turbine, which comprises the following steps: obtaining meteorological environment parameters and an initial three-dimensional structure model, and calculating initial aerodynamic load; taking the initial aerodynamic load as a boundary condition, sequentially performing multi-component lightweight parameter optimization, and generating lightweight structure parameters; mapping the lightweight structure parameters to generate a lightweight three-dimensional structure model and superimposing structure deformation data to obtain a model after load deformation; inputting the model after load deformation into an aerodynamic surrogate model in combination with a displacement relaxation factor to obtain updated aerodynamic load; and when a load deviation value is less than a preset convergence threshold, outputting the lightweight structure parameters as final design parameters. The application can solve the technical problem that in the existing lightweight design of a wind turbine, the global weight reduction effect is limited due to isolated optimization of each component.

Description

Technical Field

[0001] This application belongs to the field of wind power generation, specifically relating to a design method and system for lightweight wind turbine structures. Background Technology

[0002] In the pursuit of large capacity and long blade diameters, lightweighting has become a core aspect of controlling manufacturing costs for large wind turbines. Currently, conventional wind turbine structural design primarily employs an engineering approach combining discrete component optimization and static boundary loading. In the specific implementation process, the initial aerodynamic load of the entire turbine is typically calculated beforehand based on standard meteorological parameters, and this fixed static load is used as the external boundary condition. After obtaining this preset load boundary, for each independent load-bearing component such as the blades, nacelle, and tower, conventional mechanical simulation software is used to independently optimize lightweighting parameters such as dimensional reduction or material layout. This aims to meet the basic strength requirements of the turbine under extreme or fatigue conditions while minimizing the overall weight.

[0003] However, the aforementioned conventional lightweight design process has significant limitations in practical engineering applications. First, there are strong load transfer and coupling relationships between the various components of a wind turbine. Optimizing each component in isolation severs the dynamic load influence between upstream and downstream components, making it difficult to achieve the true lightweight limit of the entire turbine through the physical stacking of such localized lightweighting results. Second, for large, highly flexible wind turbines, changes in structural parameters due to lightweighting will inevitably lead to more significant deformation under load. This deformation directly alters the actual aerodynamic shape of the turbine during operation, resulting in a redistribution of the actual aerodynamic load. Existing unidirectional decoupling designs often have significant deviations in output design parameters because they lack a load re-estimation mechanism after deformation. However, if a high-precision fluid dynamics solver is called for fully coupled aerodynamic and structural calculations in each iteration of massive parameter optimization, the computational cost becomes excessively high, making the overall design process extremely difficult to converge and time-consuming. Summary of the Invention

[0004] This application addresses the technical problem that the overall weight reduction effect is limited due to the isolated optimization of each component in the existing lightweight design of wind turbine units, and provides a design method and system for lightweight wind turbine unit structure.

[0005] To achieve the above objectives, this application adopts the following technical solution: Firstly, a design method for lightweight wind turbine structures includes: Obtain the meteorological environment parameters of the target wind farm and the initial three-dimensional structural model of the wind turbine, and calculate the initial aerodynamic load of the wind turbine based on the meteorological environment parameters; Using the initial aerodynamic load as a boundary condition, the initial three-dimensional structural model is sequentially optimized for multiple lightweight components to generate lightweight structural parameters. The lightweight structural parameters are mapped to the initial three-dimensional structural model to generate a lightweight three-dimensional structural model. The structural deformation data of the lightweight three-dimensional structural model is calculated based on the boundary conditions. The structural deformation data is superimposed on the lightweight three-dimensional structural model to obtain a lightweight three-dimensional structural model after load deformation. The lightweight three-dimensional structural model after load deformation is input into the pre-constructed aerodynamic proxy model along with a preset displacement relaxation factor to obtain the updated aerodynamic load. The load deviation value between the updated aerodynamic load and the initial aerodynamic load is calculated. When the load deviation value is less than the preset convergence threshold, the current lightweight structural parameters are output as the final design parameters, and the lightweight structural design of the wind turbine is realized based on the final design parameters.

[0006] In some implementations, the step of sequentially optimizing the lightweight parameters of multiple components in the initial three-dimensional structural model includes: The lightweighting parameters are optimized sequentially from the blade components, nacelle components to the tower components. When the optimization is transferred between adjacent components, the component with the earlier transfer sequence outputs a dynamic response load, which is converted into an equivalent fatigue load. The equivalent fatigue load is then applied as an input boundary condition to the component with the later transfer sequence.

[0007] In some embodiments, the step of optimizing lightweight parameters sequentially from blade components, nacelle components to tower components includes: Using the maximization of stiffness and minimization of weight as multi-objective functions, a multi-objective genetic algorithm is used to optimize the layup sequence and fiber volume fraction of the blade components. The optimized layup sequence and fiber volume fraction are used as lightweight structural parameters of the blade, and the dynamic response load of the optimized blade components is extracted and converted into the corresponding equivalent fatigue load. Using the equivalent fatigue load of the optimized blade component as the input boundary condition, the topology configuration and material distribution of the nacelle component are optimized by a topology optimization algorithm. The optimized topology configuration and material distribution are used as the lightweight structural parameters of the nacelle. The dynamic response load of the optimized nacelle component is extracted and converted into the corresponding equivalent fatigue load. Using the equivalent fatigue load of the optimized nacelle component as the input boundary condition and satisfying buckling stability as the constraint condition, the gradient optimization algorithm is used to optimize the segment height, taper and wall thickness of the tower component, and the optimized segment height, taper and wall thickness are used as the lightweight structural parameters of the tower. The lightweight structural parameters of the blade, the lightweight structural parameters of the nacelle, and the lightweight structural parameters of the tower together constitute the lightweight structural parameters.

[0008] In some implementations, during the process of sequentially optimizing the lightweight parameters of multiple components of the initial three-dimensional structural model, the first-order natural frequency of the overall structure of the wind turbine avoids a preset excitation frequency range as a dynamic constraint condition, including: When generating the lightweight structural parameters in each iteration, the mass matrix and stiffness matrix of the initial three-dimensional structural model or the lightweight three-dimensional structural model in the current iteration are extracted for modal analysis, and the first-order natural frequency of the overall structure of the wind turbine is calculated. Determine whether the first-order natural frequency of the overall structure of the current wind turbine falls within the preset excitation frequency range. If it does, add the first-order natural frequency of the overall structure of the current wind turbine as a penalty variable to the objective function of the multi-component lightweighting parameter optimization.

[0009] In some implementations, the pneumatic proxy model is pre-built through the following steps: A flexible beam aeroelastic coupling model of the wind turbine blades is constructed, and a training sample set is obtained, which includes multiple sets of aerodynamic shape parameters and blade structural deformation parameters. The training sample set is input into the flexible beam aeroelastic coupling model and solved using a computational fluid dynamics algorithm to calculate the aerodynamic load corresponding to each set of aerodynamic shape parameters and blade structure deformation parameters. The aerodynamic surrogate model is generated by using the aerodynamic shape parameters and the blade structural deformation parameters as input matrices and the corresponding aerodynamic loads as output matrices for machine learning training.

[0010] In some embodiments, constructing the flexible beam aeroelastic coupling model of the wind turbine blades includes: Based on the fluid parameters of the preset low air density operating condition, the pressure distribution of the preset basic airfoil is iterated to obtain the target aerodynamic shape parameters corresponding to the maximum lift-to-drag ratio. The target aerodynamic shape parameters are input into the structural mechanics equations that include blade geometric nonlinearity, and the flapping and swaying coupling parameters of the wind turbine blades are extracted as blade structural deformation parameters. The flexible beam aeroelastic coupling model is constructed based on the target aerodynamic shape parameters and the blade structural deformation parameters.

[0011] Secondly, a lightweight design system for wind turbine structures includes: The initial load calculation module is used to obtain the meteorological environment parameters of the target wind farm and the initial three-dimensional structural model of the wind turbine, and calculate the initial aerodynamic load of the wind turbine based on the meteorological environment parameters. The parameter optimization module is used to take the initial aerodynamic load as a boundary condition and perform multi-component lightweight parameter optimization on the initial three-dimensional structural model in sequence to generate lightweight structural parameters. The deformation reconstruction module is used to map the lightweight structural parameters to the initial three-dimensional structural model, generate a lightweight three-dimensional structural model, calculate the structural deformation data of the lightweight three-dimensional structural model based on the boundary conditions, and superimpose the structural deformation data onto the lightweight three-dimensional structural model to obtain a lightweight three-dimensional structural model after load deformation. The aerodynamic update and evaluation module is used to input the lightweight three-dimensional structural model after load deformation, combined with a preset displacement relaxation factor, into a pre-constructed aerodynamic proxy model to obtain the updated aerodynamic load, and calculate the load deviation value between the updated aerodynamic load and the initial aerodynamic load. The convergence output module is used to output the current lightweight structural parameters as the final design parameters when the load deviation value is less than the preset convergence threshold, and to realize the lightweight structural design of the wind turbine based on the final design parameters.

[0012] Thirdly, a computer device includes: a processor and a computer-readable storage medium; A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, implements the lightweight design method for wind turbine structures as described above.

[0013] Fourthly, a computer-readable storage medium storing a computer program adapted to be loaded by a processor and executed as described in the design method for lightweight wind turbine structures.

[0014] Fifthly, a computer program product comprising a computer program that, when executed by a processor, implements the design method for lightweight wind turbine structure as described above.

[0015] Compared with the prior art, this application has the following beneficial effects: This application provides a lightweight design method for wind turbine structures. By using the initial aerodynamic load as boundary conditions, lightweight structural parameters are generated by sequentially optimizing the lightweight parameters of multiple components on the initial three-dimensional structural model. These lightweight structural parameters are then mapped back to the initial three-dimensional structural model. The reconstructed lightweight three-dimensional structural model, obtained by superimposing load deformations, is then combined with a preset displacement relaxation factor and input into a pre-constructed aerodynamic proxy model to obtain updated aerodynamic loads. Finally, design parameters are output when the load deviation is less than a preset convergence threshold. This scheme uses a pre-trained aerodynamic proxy model to replace the full-scale hydrodynamic calculation process in conventional methods. While introducing load deformation data for aeroelastic coupling closed-loop iteration, it reduces the enormous computational overhead to an engineering-acceptable range. This design directly overcomes the limitation of global weight reduction caused by isolated optimization of individual components in existing technologies and effectively resolves the technical contradiction between aerodynamic load prediction distortion caused by structural load deformation and low computational efficiency.

[0016] Furthermore, by optimizing lightweight parameters sequentially from blade components, nacelle components to tower components, and outputting dynamic response loads as adjacent components are transferred and converted into equivalent fatigue loads, the equivalent fatigue loads are then applied as input boundary conditions to the components transferred later in the sequence. This breaks down the data barriers between the upstream and downstream load-bearing structures of the wind turbine, ensuring that the optimized boundary conditions of downstream components are based on the actual mechanical state of the upstream components after weight reduction. This completely eliminates the local optimum trap caused by traditional static load independent optimization and ensures the physical reliability of the multi-component collaborative weight reduction design.

[0017] Furthermore, a multi-objective genetic algorithm is used to optimize the layup sequence and fiber volume fraction of the blade components. A topology optimization algorithm is then used to optimize the topology and material distribution of the nacelle components, using the equivalent fatigue load of the optimized blade components as the input boundary condition. Finally, a gradient optimization algorithm is used to optimize the segment height, taper, and wall thickness of the tower components, using the equivalent fatigue load of the optimized nacelle components as the input boundary condition and combining it with buckling stability constraints. This process precisely transfers macroscopic loads to the unique material and geometric constraints of each component, matching appropriate algorithms and engineering design parameters. This results in lightweight structural parameters with high manufacturing feasibility and structural mechanical performance. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a lightweight design method for wind turbine structures provided in this application embodiment; Figure 2 A structural diagram of a lightweight wind turbine design system provided in this application embodiment; Figure 3 A computer device diagram provided for an embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] In one embodiment of this application, such as Figure 1 As shown, a lightweight design method for wind turbine structures is provided, including: S1, Obtain the meteorological environment parameters of the target wind farm and the initial three-dimensional structural model of the wind turbine, and calculate the initial aerodynamic load of the wind turbine based on the meteorological environment parameters; Specifically, four meteorological environmental parameters—annual average wind speed, turbulence intensity, wind shear index, and air density—are obtained from an on-site wind measurement tower for the target high-altitude wind farm. Simultaneously, a pre-constructed initial three-dimensional structural model of a 5MW wind turbine is imported, which includes the baseline geometric parameters and material properties of the blades, nacelle, and tower. After obtaining the above basic data, the three-dimensional flow field distribution under rated operating conditions of the target wind farm is solved using blade element momentum theory (BEM), thereby calculating the initial thrust and torque on the swept surface of the wind turbine rotor, i.e., the initial aerodynamic load of the wind turbine. This initial aerodynamic load provides the basic boundary input for subsequent static and optimization solutions.

[0022] S2, using the initial aerodynamic load as a boundary condition, perform multi-component lightweight parameter optimization on the initial three-dimensional structural model in sequence to generate lightweight structural parameters; In the parameter optimization process of this step, to follow the actual physical law of load transfer from top to bottom in wind turbine units, the transfer is carried out sequentially from blade components, nacelle components to tower components. When optimizing the transfer between adjacent components, the rainflow counting method and Miner's linear fatigue cumulative damage theory are used to convert the time-series dynamic response load output by the component with the earlier transfer sequence into the corresponding equivalent fatigue load (DEL), and the equivalent fatigue load is applied as an input boundary condition to the component with the later transfer sequence.

[0023] The specific execution process of component-level optimization is as follows: First, for the blade component, the NSGA-II multi-objective genetic algorithm is used to optimize the layup sequence and fiber volume fraction of the main beam and skin regions of the blade component, with the goal of maximizing stiffness and minimizing weight. After convergence, the optimized layup sequence and fiber volume fraction are used as lightweight structural parameters of the blade, and the dynamic response load of the blade component at the hub center is extracted and converted into the corresponding equivalent fatigue load.

[0024] Secondly, for the nacelle components, the equivalent fatigue load of the optimized blade components is used as the input boundary condition and transformed into the equivalent static load condition of the topology solver. The SIMP topology optimization algorithm based on the variable density method is used to optimize the topology configuration and material distribution of the nacelle base components, and redundant materials in low-stress areas are eliminated. The optimized topology configuration and material distribution are used as the lightweight structural parameters of the nacelle. The dynamic response load of the optimized nacelle components at the yaw bearing is extracted and converted into the corresponding equivalent fatigue load.

[0025] Finally, for the tower component, using the equivalent fatigue load of the optimized nacelle component as the input boundary condition and satisfying the preset fatigue strength and fatigue life as constraints, the sequential quadratic programming (SQP) gradient optimization algorithm is used to optimize the segment height, taper, and wall thickness of each segment of the tower component. The optimized segment height, taper, and wall thickness are then used as the lightweight structural parameters of the tower. The lightweight structural parameters of the blades, the lightweight structural parameters of the nacelle, and the lightweight structural parameters of the tower together constitute the lightweight structural parameters of the entire aircraft.

[0026] Furthermore, during each iteration of generating the lightweight structural parameters, the overall mass matrix and stiffness matrix of the initial three-dimensional structural model or the lightweight three-dimensional structural model in the current iteration are extracted for finite element modal analysis to calculate the first-order natural frequency of the current wind turbine's overall structure. It is then determined whether this natural frequency falls within the preset excitation frequency range of the first harmonic (1P) or third harmonic (3P) of the wind turbine's rotor rotation during normal operation. If it falls within this range, the extreme value penalty function method is used, adding the current first-order natural frequency of the wind turbine's overall structure as a penalty variable to the objective function of the aforementioned multi-component lightweight parameter optimization. This forces the solver to adjust the material distribution and geometric dimensions in the next round of calculation, thereby causing the first-order natural frequency of the overall structure to deviate from the preset excitation frequency range, achieving dynamic vibration avoidance constraints.

[0027] S3, map the lightweight structural parameters to the initial three-dimensional structural model to generate a lightweight three-dimensional structural model, calculate the structural deformation data of the lightweight three-dimensional structural model based on the boundary conditions, and superimpose the structural deformation data onto the lightweight three-dimensional structural model to obtain a lightweight three-dimensional structural model after load deformation. Specifically, the ply thickness, nacelle topology, and tower wall thickness parameters obtained through multi-component collaborative optimization in S2 are updated in the finite element analysis software to reconstruct the mesh file of the lightweight three-dimensional structural model. Subsequently, under the action of initial aerodynamic loads, nonlinear statics are solved to extract the three-dimensional spatial displacement vectors of each node, i.e., the structural deformation data. Using mesh deformation technology, the node displacement vectors are directly superimposed onto the mesh node coordinates of the original lightweight three-dimensional structural model to update the spatial geometry, thereby obtaining the true flexible physical morphology of the wind turbine under lightweight and actual loading conditions, i.e., the lightweight three-dimensional structural model after loading and deformation.

[0028] S4, The lightweight three-dimensional structural model after load deformation is input into the pre-constructed aerodynamic proxy model along with a preset displacement relaxation factor to obtain the updated aerodynamic load, and the load deviation value between the updated aerodynamic load and the initial aerodynamic load is calculated. Specifically, to avoid numerical oscillations during the fluid-structure interaction iterative calculation, a preset displacement relaxation factor of 0.5 is introduced to under-relax the input structural deformation data. The relaxed deformed shape is then input into a pre-built aerodynamic surrogate model for rapid simulation, directly outputting the updated aerodynamic loads redistributed due to deformation. Subsequently, the L2 norm tool is used to calculate the load deviation between the updated aerodynamic load vector and the initial aerodynamic load vector.

[0029] In this step, the aerodynamic proxy model is pre-built in the early design phase through the following data processing flow: For the low air density conditions of the target high-altitude wind farm, local characteristic fluid parameters are acquired; the pressure distribution of the preset basic airfoil is iteratively solved to determine the target aerodynamic shape parameters that satisfy the maximum lift-to-drag ratio. Subsequently, these target aerodynamic shape parameters are input into the structural mechanics equations that include blade geometric nonlinearity; the flapping and flaring coupled displacement data of the customized blade under thrust are extracted and used as blade structural deformation parameters.

[0030] Based on the aforementioned target aerodynamic shape parameters and blade structural deformation parameters, a parameterized aeroelastic coupling model of the flexible beam for the wind turbine blade is constructed. Using this flexible beam aeroelastic coupling model, a training sample set containing multiple sets of aerodynamic shape parameters and blade structural deformation parameters is generated using the Latin hypercube sampling algorithm. This training sample set is input into the model, and a computational fluid dynamics solver is invoked for high-precision batch solving to calculate the corresponding aerodynamic load matrix. Finally, using the aerodynamic shape parameters and blade structural deformation parameters as input matrices and the corresponding aerodynamic loads as output matrices, a Kriging machine learning model is trained to generate the aerodynamic surrogate model that balances computational accuracy and inference speed.

[0031] S5, when the load deviation value is less than the preset convergence threshold, output the current lightweight structural parameters as the final design parameters, and realize the lightweight structural design of the wind turbine based on the final design parameters.

[0032] Specifically, a preset convergence threshold for aerodynamic and structural coupling is set as a load relative error of less than 1%. When the load deviation value calculated in S4 is greater than or equal to this preset convergence threshold, the system will replace the initial aerodynamic load with the updated aerodynamic load and return to S2 to re-optimize the lightweight parameters of multiple components and calculate the load deformation, starting a new round of fluid-structure interaction iteration. When the calculated load deviation value is less than this preset convergence threshold, it indicates that the current lightweight structure has reached a stable aeroelastic equilibrium state between the applied load and its own deformation. At this time, the iteration loop is exited, and the lightweight structural parameters output in the current round are exported as the final design parameters for subsequent guidance in mold development, drawing, and prototype manufacturing, thus implementing the lightweight structural design of the wind turbine in a physical entity.

[0033] In one embodiment of this application, such as Figure 2 As shown, a lightweight design system for wind turbine structures is provided, including: The initial load calculation module is used to obtain the meteorological environment parameters of the target wind farm and the initial three-dimensional structural model of the wind turbine, and calculate the initial aerodynamic load of the wind turbine based on the meteorological environment parameters. The parameter optimization module is used to take the initial aerodynamic load as a boundary condition and perform multi-component lightweight parameter optimization on the initial three-dimensional structural model in sequence to generate lightweight structural parameters. The deformation reconstruction module is used to map the lightweight structural parameters to the initial three-dimensional structural model, generate a lightweight three-dimensional structural model, calculate the structural deformation data of the lightweight three-dimensional structural model based on the boundary conditions, and superimpose the structural deformation data onto the lightweight three-dimensional structural model to obtain a lightweight three-dimensional structural model after load deformation. The aerodynamic update and evaluation module is used to input the lightweight three-dimensional structural model after load deformation, combined with a preset displacement relaxation factor, into a pre-constructed aerodynamic proxy model to obtain the updated aerodynamic load, and to calculate the load deviation value between the updated aerodynamic load and the initial aerodynamic load. The convergence output module is used to output the current lightweight structural parameters as the final design parameters when the load deviation value is less than the preset convergence threshold, and to realize the lightweight structural design of the wind turbine based on the final design parameters.

[0034] Specific limitations regarding the lightweight design system for wind turbine structures can be found in the above-described limitations of the lightweight design method for wind turbine structures; the corresponding technical effects are equivalent and will not be repeated here. Each module in the aforementioned lightweight design system for wind turbine structures can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0035] Figure 3 An internal structural diagram of a computer device is shown in one embodiment. This computer device may specifically be a terminal or a server. Figure 3As shown, the computer device includes a processor, memory, network interface, display, camera, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a lightweight design method for wind turbine structures. The display screen can be an LCD screen or an e-ink display screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0036] As will be understood by those skilled in the art, computer equipment Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computing device may include more or fewer components than those shown in the figure, or combine certain components, or have the same component arrangement.

[0037] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0038] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0039] In summary, the lightweight design method, system, computer equipment, and storage medium for wind turbine structures provided in this application break through the technical barriers of isolated optimization of individual components in traditional wind turbines. It constructs a multi-component collaborative optimization mechanism based on the cascaded transmission of fatigue loads and innovatively introduces structural deformation under load into a pre-trained aerodynamic surrogate model to achieve closed-loop load updates. This completely overcomes the load prediction distortion defects caused by unidirectional decoupling in the lightweight design of large flexible turbines. While significantly reducing the computational cost of full fluid-structure interaction calculations and ensuring the overall dynamic stability and fatigue life of the turbine, it achieves a highly reliable globally optimal lightweight design in a physically and mechanically sense.

[0040] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0041] The above-described embodiments are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A design method for lightweight wind turbine structure, characterized in that, include: Obtain the meteorological environment parameters of the target wind farm and the initial three-dimensional structural model of the wind turbine, and calculate the initial aerodynamic load of the wind turbine based on the meteorological environment parameters; Using the initial aerodynamic load as a boundary condition, the initial three-dimensional structural model is sequentially optimized for multiple lightweight components to generate lightweight structural parameters. The lightweight structural parameters are mapped to the initial three-dimensional structural model to generate a lightweight three-dimensional structural model. The structural deformation data of the lightweight three-dimensional structural model is calculated based on the boundary conditions. The structural deformation data is superimposed on the lightweight three-dimensional structural model to obtain a lightweight three-dimensional structural model after load deformation. The lightweight three-dimensional structural model after load deformation is input into the pre-constructed aerodynamic proxy model along with a preset displacement relaxation factor to obtain the updated aerodynamic load. The load deviation value between the updated aerodynamic load and the initial aerodynamic load is calculated. When the load deviation value is less than the preset convergence threshold, the current lightweight structural parameters are output as the final design parameters, and the lightweight structural design of the wind turbine is realized based on the final design parameters.

2. The lightweight design method for wind turbine structures according to claim 1, characterized in that, The step of sequentially optimizing the lightweight parameters of multiple components in the initial three-dimensional structural model includes: The lightweighting parameters are optimized sequentially from the blade components, nacelle components to the tower components. When the optimization is transferred between adjacent components, the component with the earlier transfer sequence outputs a dynamic response load, which is converted into an equivalent fatigue load. The equivalent fatigue load is then applied as an input boundary condition to the component with the later transfer sequence.

3. The lightweight design method for wind turbine structures according to claim 2, characterized in that, The process of optimizing lightweight parameters sequentially from blade components, nacelle components to tower components includes: Using the maximization of stiffness and minimization of weight as multi-objective functions, a multi-objective genetic algorithm is used to optimize the layup sequence and fiber volume fraction of the blade components. The optimized layup sequence and fiber volume fraction are used as lightweight structural parameters of the blade, and the dynamic response load of the optimized blade components is extracted and converted into the corresponding equivalent fatigue load. Using the equivalent fatigue load of the optimized blade component as the input boundary condition, the topology configuration and material distribution of the nacelle component are optimized by a topology optimization algorithm. The optimized topology configuration and material distribution are used as the lightweight structural parameters of the nacelle. The dynamic response load of the optimized nacelle component is extracted and converted into the corresponding equivalent fatigue load. Using the equivalent fatigue load of the optimized nacelle component as the input boundary condition and satisfying buckling stability as the constraint condition, the gradient optimization algorithm is used to optimize the segment height, taper and wall thickness of the tower component, and the optimized segment height, taper and wall thickness are used as the lightweight structural parameters of the tower. The lightweight structural parameters of the blade, the lightweight structural parameters of the nacelle, and the lightweight structural parameters of the tower together constitute the lightweight structural parameters.

4. The lightweight design method for wind turbine structures according to claim 1, characterized in that, During the process of sequentially optimizing the lightweight parameters of multiple components of the initial three-dimensional structural model, the first-order natural frequency of the overall structure of the wind turbine avoids the preset excitation frequency range as a dynamic constraint condition, including: When generating the lightweight structural parameters in each iteration, the mass matrix and stiffness matrix of the initial three-dimensional structural model or the lightweight three-dimensional structural model in the current iteration are extracted for modal analysis, and the first-order natural frequency of the overall structure of the wind turbine is calculated. Determine whether the first-order natural frequency of the overall structure of the current wind turbine falls within the preset excitation frequency range. If it does, add the first-order natural frequency of the overall structure of the current wind turbine as a penalty variable to the objective function of the multi-component lightweighting parameter optimization.

5. The lightweight design method for wind turbine structures according to claim 1, characterized in that, The pneumatic proxy model is pre-built through the following steps: A flexible beam aeroelastic coupling model of the wind turbine blades is constructed, and a training sample set is obtained, which includes multiple sets of aerodynamic shape parameters and blade structural deformation parameters. The training sample set is input into the flexible beam aeroelastic coupling model and solved using a computational fluid dynamics algorithm to calculate the aerodynamic load corresponding to each set of aerodynamic shape parameters and blade structure deformation parameters. The aerodynamic surrogate model is generated by using the aerodynamic shape parameters and the blade structural deformation parameters as input matrices and the corresponding aerodynamic loads as output matrices for machine learning training.

6. The lightweight design method for wind turbine structures according to claim 5, characterized in that, The construction of the flexible beam aeroelastic coupling model for the wind turbine blades includes: Based on the fluid parameters of the preset low air density operating condition, the pressure distribution of the preset basic airfoil is iterated to obtain the target aerodynamic shape parameters corresponding to the maximum lift-to-drag ratio. The target aerodynamic shape parameters are input into the structural mechanics equations that include blade geometric nonlinearity, and the flapping and swaying coupling parameters of the wind turbine blades are extracted as blade structural deformation parameters. The flexible beam aeroelastic coupling model is constructed based on the target aerodynamic shape parameters and the blade structural deformation parameters.

7. A lightweight design system for wind turbine structures, characterized in that, include: The initial load calculation module is used to obtain the meteorological environment parameters of the target wind farm and the initial three-dimensional structural model of the wind turbine, and calculate the initial aerodynamic load of the wind turbine based on the meteorological environment parameters. The parameter optimization module is used to take the initial aerodynamic load as a boundary condition and perform multi-component lightweight parameter optimization on the initial three-dimensional structural model in sequence to generate lightweight structural parameters. The deformation reconstruction module is used to map the lightweight structural parameters to the initial three-dimensional structural model, generate a lightweight three-dimensional structural model, calculate the structural deformation data of the lightweight three-dimensional structural model based on the boundary conditions, and superimpose the structural deformation data onto the lightweight three-dimensional structural model to obtain a lightweight three-dimensional structural model after load deformation. The aerodynamic update and evaluation module is used to input the lightweight three-dimensional structural model after load deformation, combined with a preset displacement relaxation factor, into a pre-constructed aerodynamic proxy model to obtain the updated aerodynamic load, and to calculate the load deviation value between the updated aerodynamic load and the initial aerodynamic load. The convergence output module is used to output the current lightweight structural parameters as the final design parameters when the load deviation value is less than the preset convergence threshold, and to realize the lightweight structural design of the wind turbine based on the final design parameters.

8. A computer device, characterized in that, include: Processor and computer-readable storage media; A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, implements a design method for lightweight wind turbine structure as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded by a processor and executed as described in any one of claims 1 to 6, a design method for lightweight wind turbine structures.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements a design method for lightweight wind turbine structure as described in any one of claims 1 to 6.

Images