Method and system for designing rack of wind generating set

Abstract

The embodiment of the invention provides a wind generating set rack design method and system, and belongs to the technical field of wind power. The method comprises the following steps: S10) carrying out topological optimization on a topological region of a rack model, and checking the rack model after topological optimization; s20) identifying and constructing a transfer matrix of stress and load of each key part; s30) executing a genetic algorithm based on a load updating result and the transfer matrix to obtain optimized geometric parameters of each key part; s40) on the basis of the optimized set parameters of each key part, updating a topological region optimization result of the rack, and checking the rack model after updating topological optimization; and (S50) repeating the step (S20) to the step (S40) until the geometric parameters, meeting the design requirements, of all the key parts are obtained, and outputting a corresponding rack design plane.According to the scheme, the lightweight design effect of the rack is improved, and the safety and economical efficiency of the rack are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a wind turbine generator set frame design method and a wind turbine generator set frame design system. Background Art

[0002] With the rapid development of the wind power industry, the design of wind turbines has gradually shifted from onshore to offshore, and the diameter of wind rotors has continued to increase. As an important component supporting the transmission chain, the frame is under increasing pressure to withstand steady-state and non-steady-state loads. This change not only leads to a significant increase in the weight of the frame, but also brings cost issues. Especially in the manufacturing, transportation and installation processes, how to achieve lightweight and optimized design of the frame while ensuring safety has become a key issue that needs to be solved urgently.

[0003] At present, the frames of wind turbines are mainly regarded as rigid components, and their design methods often adopt a high safety redundancy to cope with potential load fluctuations. However, this design concept leads to overly conservative and unreasonable material distribution, and the overall structural weight of the frame is relatively large. Overweight frames not only increase material costs, but also increase the difficulty of transportation and installation, which affects the economic feasibility of the entire wind power project. In addition, the high redundancy design of the frame makes it unable to fully exert its performance in actual work, reduces the utilization rate of materials, and causes waste of resources.

[0004] In existing design methods, the frame usually lacks dynamic response analysis for load changes and fails to fully consider the actual stress conditions of wind turbines under different working conditions. This design scheme lacking flexibility and adaptability cannot meet the dual requirements of modern wind turbines for structural lightweight and strength. Especially in the face of ever-changing offshore wind conditions, the design of the frame needs to have higher dynamic adaptability. Therefore, there is an urgent need for a new type of frame optimization design scheme that can achieve lightweight design of the frame while ensuring safety and reliability based on accurate load analysis and advanced material distribution optimization methods. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a wind turbine generator frame design method and system to at least solve the problem that existing solutions cannot meet the dual requirements of modern wind turbines for structural lightness and strength.

[0006] In order to achieve the above-mentioned purpose, the first aspect of the present invention provides a wind turbine generator frame design method, the method comprising: S10) based on the basic design information of the frame, constructing a frame model with the frame lightweight index, topologically optimizing the topological area of ​​the frame model, and verifying the frame model after topological optimization; S20) based on the verification data of the frame model, identifying and constructing the transfer matrix of stress and load of each key part; S30) performing load update on the verified frame model, and executing a genetic algorithm based on the load update result and the transfer matrix to obtain the optimized geometric parameters of each key part; S40) when the geometric parameters of each key part after optimization do not meet the design requirements, based on the set parameters of each key part after optimization, updating the optimization result of the topological area of ​​the frame, and verifying the updated frame model after topological optimization; S50) repeating steps S20) to S40) until the geometric parameters of each key part that meet the design requirements are obtained, and the corresponding frame design scheme is output.

[0007] Optionally, the rack lightweight index is the rack weight; the rack model includes a rack structure model and a rack weight optimization model; the rack weight optimization model is a model constructed in the rack structure model with the rack weight and the rack stiffness as optimization variables; the rack basic design information includes: location information and functional information of each design component.

[0008] Optionally, the topological optimization of the topological area of ​​the rack model includes: classifying each component based on the position information and function information of each component to obtain a set of accessories for multiple design areas; wherein each design area includes: an assembly area, a topological area, and a non-optimized area; constructing a structural sub-model of the corresponding design area based on the accessory set of each design area; determining and updating the basic geometric parameters of each structural sub-model based on the basic design information, and assembling each structural sub-model to obtain an updated rack structure model; performing topological area optimization on the updated rack structure model to obtain a topologically optimized rack structure model, and obtaining a corresponding topologically optimized rack model based on the optimized topologically optimized rack structure model.

[0009] Optionally, the topological area optimization is performed on the updated rack structure model, including: taking the minimum geometric volume of the topological area as the optimization target, constructing an optimization problem of the topological area; adding geometric constraints, stress constraints and boundary condition constraints to the optimization problem based on design principles; solving the optimization problem based on the density method to obtain the topological area optimization result in the current updated rack structure model.

[0010] Optionally, the verifying the rack model after topology optimization includes: performing geometric repair and smoothing on the rack model after topology optimization, performing strength verification on the rack model after the processing, and adjusting local shapes and dimensions of the rack model based on the verification results until a rack model that meets the strength verification is obtained.

[0011] Optionally, based on the calibration data of the frame model, a transfer matrix of stress and load at each key part is identified and constructed, including: preprocessing the calibration data, and performing principal component analysis on the preprocessed calibration data to screen out the top N parts with the greatest influence on the relationship between stress and load as key parts; constructing a corresponding price reduction model based on the screened key parts, and constructing a transfer matrix of stress and load at each key part based on the price reduction model.

[0012] Optionally, the load update is performed on the verified frame model, and a genetic algorithm is executed based on the load update result and the transfer matrix to obtain the optimized geometric parameters of each key part, including: establishing a dynamic model of the frame and the transmission chain based on the verified frame model, and updating the load at the connection position of the frame model based on the dynamic model; setting minimization of stress concentration as the optimization goal, defining the optimization range of the geometric parameters of each key part, and generating initial geometric parameters; iteratively searching, crossovering and mutating each geometric parameter through a genetic algorithm, and evaluating the influence of each generation of parameter combination on the optimization goal; outputting a set of geometric parameters corresponding to the current minimized stress concentration as the optimized geometric parameters of each key part.

[0013] Optionally, the method also includes: judging whether the geometric parameters of each key part after current optimization meet the design requirements, including: reconstructing the rack model based on the geometric parameters of each key part after optimization, and verifying the reconstructed rack model based on preset design limits to judge whether it meets each design limit at the same time; if so, the geometric parameters of each key part after current optimization meet the design requirements; otherwise, the geometric parameters of each key part after current optimization do not meet the design requirements; wherein, the design limits include: any one or more of rack shape and size limits, rack weight limits and rack strength limits.

[0014] Optionally, until the geometric parameters of each key part that meet the design requirements are obtained, the corresponding rack design scheme is output, including: reconstructing the rack model based on the geometric parameters of each key part that meet the design requirements to obtain the final rack model; performing geometric repair and smoothing on the final rack model, and performing finite element analysis on the final rack model after the processing to construct a corresponding mesh model; applying loads and boundary constraints under actual working conditions of the installation location to the mesh model, and evaluating whether the strength of the final rack model meets environmental requirements, and outputting the corresponding rack design scheme when it meets environmental requirements.

[0015] The second aspect of the present invention provides a wind turbine generator frame design system, the system comprising: a pre-construction unit, which is used to construct a frame model based on the basic design information of the frame and the frame lightweight index, perform topological optimization on the topological area of ​​the frame model, and perform verification on the frame model after topological optimization; a matrix construction unit, which is used to identify and construct the transfer matrix of stress and load of each key part based on the verification data of the frame model; an optimization unit, which is used to perform load update on the verified frame model, and execute a genetic algorithm based on the load update result and the transfer matrix to obtain the geometric parameters of each key part after optimization; a verification unit, which is used to update the topological area optimization result of the frame based on the set parameters of each key part after optimization when the geometric parameters of each key part after optimization do not meet the design requirements, and perform verification on the updated frame model after topological optimization; an iteration unit, which is used to repeatedly perform the geometric parameter optimization of each key part until the geometric parameters of each key part that meet the design requirements are obtained, and the corresponding frame design scheme is output.

[0016] On the other hand, the present invention provides a computer-readable storage medium, on which instructions are stored, which, when executed on a computer, enable the computer to execute the above-mentioned wind turbine generator frame design method.

[0017] Through the above technical solution, based on the basic design information of the rack, the rack model was constructed and topology optimization was performed, which significantly reduced the weight of the rack while maintaining its structural strength and stiffness. This process ensures that the rack can adapt to larger loads while reducing the amount of material used. Secondly, by checking the data, the stress and load transfer matrix of each key part was successfully identified and constructed, making the force analysis of the rack more accurate. This transfer matrix provides an important basis for subsequent load updates and optimization. Furthermore, after the load update is performed, the genetic algorithm is used to optimize the geometric parameters of the key parts to ensure that these parameters further improve the performance of the rack while meeting the design requirements. This method continuously adjusts the design through intelligent algorithms, so that the adaptability and reliability of the rack are enhanced. In the case of design dissatisfaction, the topology optimization results are updated based on the set parameters, and repeatedly checked to ensure that each optimization is moving in the direction of meeting the design requirements. Through this iterative method, the optimal design of the geometric parameters of each key part of the rack is finally achieved. In summary, this technical solution effectively combines topology optimization, stress analysis and intelligent algorithms, which not only improves the lightweight design effect of the frame, but also ensures its safety and economy, making an important contribution to the overall performance improvement of the wind turbine.

[0018] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings: Figure 1 is a flow chart of steps of a wind turbine generator set frame design method provided by one embodiment of the present invention; Figure 2 It is a system structure diagram of a wind turbine generator set frame design system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0020] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0021] Figure 1 FIG. 1 is a flow chart of a method for designing a wind turbine generator frame according to an embodiment of the present invention. Figure 1 As shown, an embodiment of the present invention provides a method for designing a wind turbine generator frame, the method comprising: Step S10: Based on the basic design information of the rack, a rack model is constructed with a rack lightweight index, topology optimization is performed on the topology area of ​​the rack model, and the rack model after topology optimization is verified.

[0022] Specifically, the rack lightweight index is the rack weight; the rack model includes a rack structure model and a rack weight optimization model; the rack weight optimization model is a model constructed in the rack structure model with the rack weight and the rack stiffness as optimization variables; the rack basic design information includes: location information and function information of each design component.

[0023] In an embodiment of the present invention, the rack lightweight index focuses on the weight of the rack, and its reduction is achieved through precise calculation and optimization. The rack model includes not only the structural model of the rack, but also the rack weight optimization model, which is constructed on the basis of the structural model with the weight and stiffness of the rack as optimization variables. This optimization process aims to balance the relationship between the load-bearing capacity and lightweight of the rack, so as to ensure that the required stiffness and strength are maintained while reducing the weight. In addition, the basic design information of the rack covers the position information and functional information of each design component, which provides necessary support for the establishment and optimization of the model. By taking the weight and stiffness of the rack as optimization variables, designers can minimize the weight of the rack while meeting the performance requirements. This method significantly improves the efficiency and scientificity of the rack design, not only reduces the manufacturing cost and transportation difficulty, but also improves the economy and market competitiveness of the overall equipment. Therefore, this technical solution effectively promotes the lightweight and performance optimization of the wind turbine rack, and provides important technical support for the development of the wind power industry.

[0024] Furthermore, the topological optimization of the topological area of ​​the rack model includes: classifying each component based on the position information and function information of each component to obtain a set of accessories for multiple design areas; wherein each design area includes: an assembly area, a topological area, and a non-optimized area; constructing a structural sub-model of the corresponding design area based on the accessory set of each design area; determining and updating the basic geometric parameters of each structural sub-model based on the basic design information, and assembling each structural sub-model to obtain an updated rack structure model; performing topological area optimization on the updated rack structure model to obtain a topologically optimized rack structure model, and obtaining a corresponding topologically optimized rack model based on the optimized topologically optimized rack structure model.

[0025] In an embodiment of the present invention, the topological optimization of the topological area of ​​the rack model first involves classifying the components based on the location information and functional information of each component, thereby obtaining a set of accessories for multiple design areas. These design areas specifically include assembly areas, topological areas, and non-optimized areas. The assembly area is used to connect with other components of the unit to ensure the stability and function of the rack; the topological area focuses on realizing the functional connection topology of the rack and is the core area of ​​optimization; and the non-optimized area is a fixed part of the design, which limits the scope of optimization.

[0026] Furthermore, the structural sub-models of the corresponding design areas are constructed based on the accessory sets of each design area. In this process, designers can understand the functions and roles of each part more clearly, so as to develop a more reasonable optimization strategy. Based on the basic design information, the basic geometric parameters of each structural sub-model are determined and updated, and these sub-models are assembled together to obtain an updated rack structure model. Subsequently, the updated rack structure model is optimized in the topological area. This process is achieved through advanced topological optimization algorithms, which aims to minimize the use of materials while ensuring the stiffness and strength of the structure. Finally, the corresponding topologically optimized rack model is obtained by processing the optimized topological structure.

[0027] Based on the solution of the present invention, the weight of the frame is effectively reduced, reducing the manufacturing and transportation costs of the overall equipment. Secondly, through the reasonable division of the design area and the construction of the structural sub-model, the flexibility and adjustability of the frame design are improved, making the design process more efficient. In addition, the optimized frame structure model is also enhanced in load-bearing capacity and stability, ensuring the safe operation of the unit under various working conditions. This comprehensive optimization solution provides a solid technical foundation for improving the performance of wind turbines and promotes the development of renewable energy.

[0028] Preferably, the topological area optimization is performed on the updated rack structure model, including: taking the minimum geometric volume of the topological area as the optimization target, constructing an optimization problem of the topological area; adding geometric constraints, stress constraints and boundary condition constraints to the optimization problem based on design principles; solving the optimization problem based on the density method to obtain the topological area optimization result in the current updated rack structure model.

[0029] In the embodiment of the present invention, the process of performing topological area optimization on the updated rack structure model first takes the minimum geometric volume of the topological area as the optimization target, and constructs an optimization problem of the topological area. This goal is to ensure that the rack achieves effective use of materials, reduces unnecessary redundancy, and improves the economy of the overall design while meeting the structural strength and stiffness requirements.

[0030] Furthermore, in the process of constructing the optimization problem, multiple constraints were added based on the design principles. These constraints include geometric constraints, stress constraints, and boundary condition constraints. Geometric constraints ensure that the shape and structure of the rack meet the design requirements and do not cause excessive deformation or instability; stress constraints are to ensure that the stress levels of each component of the rack are within a safe range when subjected to external loads to avoid premature failure of the material; boundary condition constraints ensure that the model can accurately reflect the actual working environment and stress conditions during actual installation. On this basis, the density method is used to solve the optimization problem. The density method is a technology widely used in topology optimization, which achieves the optimal design by adjusting the material density of each unit in the design area. During the solution process, the algorithm will continue to iterate, and by removing unnecessary materials, a lighter and more efficient structural form will eventually be formed.

[0031] Based on the solution of the present invention, the geometric volume of the frame is effectively reduced, which directly leads to a reduction in the use of materials, thereby reducing manufacturing and transportation costs. Secondly, the optimized frame performs well in strength and stiffness, can better withstand external loads, and ensures the safety and stability of the wind turbine generator set in harsh environments. In addition, after topological optimization, the overall structure of the frame is more reasonable, and the mechanical properties are improved, which helps to improve the service life and operating efficiency of the equipment. In summary, this optimization method not only improves the scientificity and rationality of the frame design, but also promotes the sustainable development of the wind power industry.

[0032] Preferably, the verification of the rack model after topology optimization includes: performing geometric repair and smoothing processing on the rack model after topology optimization, performing strength verification on the rack model after processing, and adjusting local shapes and dimensions of the rack model based on the verification results until a rack model that meets the strength verification is obtained.

[0033] In an embodiment of the present invention, the process of checking the rack model after topology optimization first includes geometric repair and smoothing of the optimized rack model. Since topology optimization may cause irregularities in the geometric shape of the model, or even sharp edges or local defects, these may cause problems in subsequent manufacturing and practical applications. Therefore, through geometric repair, the designer can ensure that the surface of the rack model is smooth and meets the manufacturing standards, reduce the difficulty of processing, and improve production efficiency. After completing the geometric repair, the strength check is then performed on the processed rack model. This check process is to comprehensively evaluate the load-bearing capacity of the rack under different working conditions through methods such as finite element analysis to ensure that the rack can work safely and stably in practical applications. If the check results show that the strength of certain parts fails to meet the requirements, the local shapes and sizes of the rack model will be adjusted based on these results. This iterative process will continue until the strength check of each key part meets the design standards.

[0034] Step S20: Based on the verification data of the frame model, the transfer matrix of stress and load of each key part is identified and constructed.

[0035] Specifically, preprocessing is performed on the verification data, and principal component analysis is performed on the preprocessed verification data to screen out the top N parts that have the greatest influence on the relationship between stress and load as key parts; a corresponding price reduction model is constructed based on the screened key parts, and a transfer matrix of stress and load of each key part is constructed based on the price reduction model.

[0036] In an embodiment of the present invention, identifying and constructing the transfer matrix of stress and load of each key part based on the verification data of the rack model is an important step to improve the scientificity and reliability of the rack design. First, for the verification data of the rack model, the original data needs to be preprocessed. The purpose of this step is to remove noise and outliers to ensure the accuracy of subsequent analysis. After preprocessing, the verification data is cleaned and sorted to better reflect the stress state and load distribution of the rack under actual working conditions. Principal component analysis (PCA) is performed on the preprocessed verification data. PCA is a statistical method that can filter out the most representative features in the data through dimensionality reduction technology. In this process, principal component analysis helps identify the top N key parts that affect the relationship between stress and load. These parts are important nodes for stress transfer and distribution. By analyzing the status of these key parts, we can focus on the core areas that affect the performance of the rack, thereby improving the pertinence and effectiveness of the design.

[0037] Furthermore, based on the selected key parts, the corresponding reduced-order model is constructed. The reduced-order model is a model that retains important features by simplifying the calculation, so that it can still be analyzed quickly and accurately when facing complex loads and stress states. Through the reduced-order model, designers can more efficiently understand the changes in the stress state of each key part under different loads. Subsequently, based on the reduced-order model, a transfer matrix of stress and load of each key part is constructed. This matrix intuitively reflects how the load is transferred and transformed between different parts, allowing designers to quickly evaluate the stress state of key parts. When faced with complex load conditions, the transfer matrix can be used to quickly identify which parts may be at risk of high stress concentration, so that design optimization measures can be taken in time to avoid potential structural failure.

[0038] Preferably, data organization and cleaning are crucial in the whole process. By using principal component analysis, designers can minimize the scale of analysis while maintaining key features. This method not only improves the efficiency of analysis, but also ensures the reliability and accuracy of the results, making design decisions more scientific.

[0039] Based on the scheme of the present invention, using the transfer matrix, designers can quickly evaluate the rack structure under different working conditions, enhancing the dynamic understanding and response capabilities of the rack. Secondly, the establishment of a reduced-order model enables efficient management and analysis of stress states in complex situations, improving the flexibility and adaptability of the design. By focusing on the analysis of key parts, the uncertainty in the design process can be effectively reduced, ensuring the safety and reliability of the rack in actual applications. The combination of this series of technical means ultimately promotes the rack design of wind turbines to higher performance and lower costs, providing solid technical support for the sustainable development of the industry.

[0040] Step S30: performing load update on the verified frame model, and executing a genetic algorithm based on the load update result and the transfer matrix to obtain optimized geometric parameters of each key part.

[0041] Specifically, a dynamic model of the frame and the transmission chain is established based on the verified frame model, and the load at the connection position of the frame model is updated based on the dynamic model; minimizing stress concentration is set as the optimization goal, the optimization range of the geometric parameters of each key part is defined, and the initial generation of geometric parameters is generated; through the iterative search of the genetic algorithm, the geometric parameters are crossed and mutated, and the influence of each generation of parameter combination on the optimization goal is evaluated; the geometric parameter set corresponding to the current minimized stress concentration is output as the geometric parameters of each key part after optimization.

[0042] In an embodiment of the present invention, performing load update on the verified rack model is an important step in the optimization design process. In this process, firstly, a dynamic model of the rack and the transmission chain is established based on the verified rack model. This dynamic model can accurately simulate the interaction between the rack and the transmission chain during operation, and fully reflect the load changes at the rack connection position under different working conditions. By analyzing the dynamic model, the designer can identify the actual loads borne by each connection point of the rack under different working conditions, thereby dynamically updating the rack model to ensure that it better adapts to the actual application scenario.

[0043] Furthermore, setting the minimization of stress concentration as the optimization goal is the core idea of ​​the design process. Stress concentration can cause excessive fatigue of the material and reduce the reliability of the rack, so this problem must be taken seriously during the optimization process. Define the optimization range of the geometric parameters of each key part, such as thickness, shape and size, and generate the initial geometric parameters to provide basic data for the iteration of the genetic algorithm.

[0044] Furthermore, through the iterative search of the genetic algorithm, the designers will cross and mutate the initial generation of geometric parameters. This method imitates the natural selection process and continuously optimizes the combination of geometric parameters in each generation. The advantage of the genetic algorithm is that it can efficiently explore the huge parameter space to find the optimal solution. In each generation, the influence of each parameter combination on the optimization target is evaluated to ensure that the solution found is effective in minimizing stress concentration. The set of geometric parameters corresponding to the current minimized stress concentration is output as the geometric parameters of each key part after optimization. This process not only improves the performance and safety of the frame, but also provides precise guidance for subsequent manufacturing and assembly.

[0045] In one possible implementation, after verification, it was found that the frame of a wind turbine generator set had a large stress concentration phenomenon at the connection part under certain working conditions. In order to deal with this problem, firstly, based on the verified frame model, a dynamic model of the frame and the transmission chain was established, taking into account the actual operation under strong wind and high load conditions. Through this model, the loads at each connection position of the frame are updated to ensure that the real working conditions are reflected. After setting the optimization goal to minimize the stress concentration at key parts, the optimization range of geometric parameters such as thickness, length, and aperture is defined, and the first generation of geometric parameters is generated, for example: the thickness of the connection part is adjusted from 10 mm to a range of 8 to 12 mm. Then, the genetic algorithm is used to iteratively search these parameters. Through crossover and mutation operations, the team generated multiple new parameter combinations and evaluated the stress concentration of each generation of combinations. After several rounds of iterations, the optimized geometric parameter set was finally obtained, such as the optimal thickness of the connection part is 9.5 mm and the length is 30 mm. These parameters not only significantly reduce stress concentration, but also meet the strength and stability requirements of the frame under various working conditions. Through this embodiment, the design of the frame is more scientific and reasonable, ensuring the safe and reliable operation of the wind turbine generator set in harsh environments, while reducing material usage and subsequent maintenance costs.

[0046] Step S40: When the geometric parameters of the optimized key parts do not meet the design requirements, the topological area optimization result of the rack is updated based on the optimized set parameters of the key parts, and the rack model after the updated topology optimization is verified.

[0047] Specifically, judging whether the geometric parameters of each key part after the current optimization meet the design requirements includes: reconstructing the rack model based on the geometric parameters of each key part after the optimization, and verifying the reconstructed rack model based on preset design limits to judge whether it meets the design limits at the same time. If yes, the geometric parameters of each key part after the current optimization meet the design requirements; otherwise, the geometric parameters of each key part after the current optimization do not meet the design requirements; wherein, the design limits include: any one or more of rack shape and size limits, rack weight limits and rack strength limits.

[0048] In an embodiment of the present invention, when the optimized geometric parameters of each key part do not meet the design requirements, further measures must be taken to ensure the overall performance and safety of the rack. First, based on these optimized geometric parameters, the rack model is reconstructed. This step is crucial because it provides an accurate basis for subsequent analysis and verification. Next, the reconstructed rack model is verified to ensure that it meets the preset design limits. These design limits may include shape and size limits, weight limits, and strength limits of the rack, or even a combination of multiple conditions. By comparing the current optimized geometric parameters with these limits, it can be determined whether the rack model meets the actual design requirements.

[0049] Specifically, the verification process usually involves the following steps: First, a new rack structure is generated by calculating the optimized geometric parameters. If the shape, size and weight of the new model are within the preset range, its strength is further evaluated to see if it meets the requirements. Strength verification generally uses numerical calculation methods such as finite element analysis (FEA), which can effectively evaluate the stress distribution of the rack under different load conditions. If the reconstructed rack model meets all design limits at the same time, it means that the current geometric parameters are valid and the rack is reliable in performance. On the contrary, if it is not met, it is necessary to update the optimization results of the topological area based on the current geometric parameters. At this point, the topology optimization strategy of the rack will be re-evaluated, which may involve adjusting the material distribution, modifying the geometric parameters of key parts, or even changing the design concept to achieve a more reasonable structural form.

[0050] Based on the solution of the present invention, this iterative optimization process greatly enhances the flexibility and adaptability of the design. Through continuous verification and updating, designers can more accurately respond to changes and challenges in the project and ensure the safety and effectiveness of the rack in actual applications. In addition, this method also effectively reduces resource waste because it can avoid investing too much time and cost in designs that do not meet the requirements during the design phase.

[0051] Step S50: Repeat steps S20 to S40 until the geometric parameters of each key part that meet the design requirements are obtained, and the corresponding rack design solution is output.

[0052] Specifically, the rack model is reconstructed based on the geometric parameters of each key part that meets the design requirements to obtain the final rack model; geometric repair and smoothing are performed on the final rack model, and finite element analysis is performed on the final rack model after the processing to construct a corresponding mesh model; loads and boundary constraints under actual working conditions of the installation location are applied to the mesh model, and the strength of the final rack model is evaluated to see if it meets the environmental requirements, and the corresponding rack design plan is output when it meets the environmental requirements.

[0053] In an embodiment of the present invention, repeating the process of steps S20 to S40 until the geometric parameters of each key part that meet the design requirements are obtained is a key link in achieving efficient and reliable rack design. During this process, parameter optimization and verification will be continuously performed to ensure that each key part can meet the preset performance standards. Through this iterative process, designers can continuously adjust and improve the design of the rack to ensure that the final model is not only feasible in theory, but also efficient and safe in practical applications. After confirming that the geometric parameters of all key parts meet the design requirements, the next step is to reconstruct the rack model based on these parameters. This reconstruction process will integrate all optimized geometric data to generate a comprehensive final rack model. This model will serve as the basis for subsequent analysis and evaluation.

[0054] Furthermore, after the reconstruction of the frame model is completed, the final frame model will be subjected to geometric repair and smoothing. This processing step is crucial because it can eliminate any geometric defects that may affect subsequent manufacturing and performance, such as sharp edges or discontinuous surfaces. After smoothing, the model is not only more visually beautiful, but also reduces the difficulty of processing during the manufacturing process and improves production efficiency.

[0055] Next, finite element analysis (FEA) is performed on the final rack model after processing. This analysis technology can evaluate the strength and stiffness performance of the rack under different working conditions in detail. In this process, the corresponding mesh model needs to be constructed first. The quality of mesh division directly affects the accuracy of the analysis results. Therefore, designers need to select the appropriate mesh size and type according to the complexity of the model. After the mesh model is built, the next step is to apply loads and boundary constraints under actual working conditions. These conditions should be based on the expected environment of the rack in actual applications, such as wind, vibration, temperature changes, etc. By applying these loads, the performance of the rack in working conditions can be more realistically simulated.

[0056] Furthermore, the strength of the final rack model is evaluated to ensure that it meets environmental requirements. If the model shows good performance in the analysis and all strength indicators are within the safe range, the corresponding rack design plan will be output. This plan not only includes the detailed design parameters of the rack, but may also include manufacturing processes, material selection, and assembly recommendations.

[0057] Figure 2 1 is a system structure diagram of a wind turbine generator rack design system provided by an embodiment of the present invention. Figure 2As shown, an embodiment of the present invention provides a wind turbine generator frame design system, the system comprising: a pre-construction unit, for constructing a frame model based on the frame basic design information and the frame lightweight index, topologically optimizing the topological area of ​​the frame model, and verifying the frame model after topological optimization; a matrix construction unit, for identifying and constructing the transfer matrix of stress and load of each key part based on the verification data of the frame model; an optimization unit, for performing load update on the verified frame model, and executing a genetic algorithm based on the load update result and the transfer matrix to obtain the geometric parameters of each key part after optimization; a verification unit, for updating the topological area optimization result of the frame based on the set parameters of each key part after optimization when the geometric parameters of each key part after optimization do not meet the design requirements, and verifying the updated frame model after topological optimization; an iteration unit, for repeatedly performing the geometric parameter optimization of each key part until the geometric parameters of each key part meeting the design requirements are obtained, and the corresponding frame design scheme is output.

[0058] The embodiment of the present invention further provides a computer-readable storage medium, on which instructions are stored, which, when executed on a computer, enable the computer to execute the above-mentioned wind turbine generator frame design method.

[0059] Those skilled in the art will understand that all or part of the steps in the method for implementing the above-mentioned embodiments can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including several instructions for making a single-chip microcomputer, a chip or a processor execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0060] The optional embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above embodiments. Within the technical concept of the embodiments of the present invention, the technical scheme of the embodiments of the present invention can be subjected to a variety of simple modifications, and these simple modifications all belong to the protection scope of the embodiments of the present invention. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the embodiments of the present invention will not further describe various possible combinations.

[0061] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the embodiments of the present invention, they should also be regarded as the contents disclosed in the embodiments of the present invention.

Claims

1. A wind turbine generator frame design method, characterized in that: The method comprises: S10) Based on the basic design information of the rack, a rack model is constructed with a rack lightweight index, topology optimization is performed on a topology area of ​​the rack model, and verification is performed on the rack model after topology optimization; S20) Based on the verification data of the frame model, identify and construct the transfer matrix of stress and load of each key part; S30) performing load update on the verified frame model, and executing a genetic algorithm based on the load update result and the transfer matrix to obtain optimized geometric parameters of each key part; S40) when the geometric parameters of the optimized key parts do not meet the design requirements, based on the optimized set parameters of the key parts, the topological area optimization result of the rack is updated, and the rack model after the updated topological optimization is verified; S50) Repeat steps S20) to S40) until the geometric parameters of each key part that meet the design requirements are obtained, and the corresponding rack design solution is output.

2. The method according to claim 1, characterized in that The rack lightweight index is the rack weight; The rack model includes a rack structure model and a rack weight optimization model; The rack weight optimization model is a model constructed in the rack structure model with the rack weight and the rack stiffness as optimization variables; The rack foundation design information includes: The location and function information of each design component.

3. The method according to claim 2, characterized in that The topological optimization of the topological area of ​​the rack model includes: Based on the location information and function information of each component, each component is classified to obtain a set of accessories for multiple design areas; wherein, The design areas include: Assembly region, topological region and non-optimized region; Constructing structural sub-models of corresponding design areas based on the accessory sets of each design area; Determine and update basic geometric parameters of each structural sub-model based on basic design information, and assemble each structural sub-model to obtain an updated rack structure model; A topology area optimization is performed on the updated rack structure model to obtain a topology optimized rack structure model, and a corresponding topology optimized rack model is obtained based on the optimized topology optimized rack structure model.

4. The method according to claim 3, characterized in that The performing topology area optimization on the updated rack structure model includes: Taking the minimum geometric volume of the topological region as the optimization goal, the optimization problem of the topological region is constructed; Adding geometric constraints, stress constraints and boundary condition constraints to the optimization problem based on design principles; The optimization problem is solved based on the density method to obtain the topology area optimization result in the currently updated rack structure model.

5. The method according to claim 4, characterized in that The checking of the frame model after topology optimization includes: The frame model after topology optimization is subjected to geometric repair and smoothing, and the frame model after processing is subjected to strength verification. Based on the verification results, the local shapes and dimensions of the frame model are adjusted until a frame model that meets the strength verification is obtained.

6. The method according to claim 1, characterized in that Based on the verification data of the frame model, the transfer matrix of stress and load of each key part is identified and constructed, including: Preprocessing the calibration data, and performing principal component analysis on the preprocessed calibration data to select the top N parts that have the greatest influence on the relationship between stress and load as key parts; A corresponding price reduction model is constructed based on the selected key parts, and a transfer matrix of stress and load of each key part is constructed based on the price reduction model.

7. The method according to claim 1, characterized in that The load update is performed on the verified frame model, and a genetic algorithm is performed based on the load update result and the transfer matrix to obtain the optimized geometric parameters of each key part, including: Establishing a dynamic model of the frame and the transmission chain based on the verified frame model, and updating the load at the connection position of the frame model based on the dynamic model; Set minimizing stress concentration as the optimization goal, define the optimization range of geometric parameters of each key part, and generate the initial geometric parameters; Through the iterative search of genetic algorithms, crossover and mutation of various geometric parameters, and evaluation of the influence of each generation of parameter combination on the optimization target; The geometric parameter set corresponding to the current minimized stress concentration is output as the geometric parameters of each key part after optimization.

8. The method according to claim 1, characterized in that The method further comprises: Determine whether the geometric parameters of the key parts after optimization meet the design requirements, including: The rack model is reconstructed based on the optimized geometric parameters of each key part, and the reconstructed rack model is verified based on the preset design limits to determine whether it satisfies the design limits at the same time. If so, the geometric parameters of each key part after the current optimization meet the design requirements. Otherwise, the geometric parameters of each key part after the current optimization do not meet the design requirements. The design limits include: Any one or more of rack shape and size limits, rack weight limits, and rack strength limits.

9. The method according to claim 1, characterized in that: Until the geometric parameters of each key part that meet the design requirements are obtained, the corresponding rack design scheme is output, including: Reconstruct the rack model based on the geometric parameters of each key part that meets the design requirements to obtain the final rack model; Performing geometric repair and smoothing processing on the final rack model, and performing finite element analysis on the final rack model after processing to construct a corresponding mesh model; Apply loads and boundary constraints under actual working conditions of the installation location to the grid model, evaluate whether the strength of the final rack model meets environmental requirements, and output the corresponding rack design plan if it meets environmental requirements.

10. A wind turbine generator frame design system, characterized in that: The system comprises: A pre-construction unit is used to construct a rack model based on the basic design information of the rack and the lightweight index of the rack, perform topological optimization on the topological area of ​​the rack model, and perform verification on the rack model after the topological optimization; A matrix construction unit is used to identify and construct the transfer matrix of stress and load of each key part based on the verification data of the frame model; An optimization unit, used for performing load update on the verified frame model, and executing a genetic algorithm based on the load update result and the transfer matrix to obtain optimized geometric parameters of each key part; A verification unit is used to update the topological area optimization result of the rack based on the set parameters of the optimized key parts when the geometric parameters of the optimized key parts do not meet the design requirements, and to verify the rack model after the updated topological optimization; The iteration unit is used to repeatedly optimize the geometric parameters of each key part until the geometric parameters of each key part that meet the design requirements are obtained, and the corresponding rack design solution is output.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the wind turbine generator frame design method described in any one of claims 1-9.

Images