Quick optimization design method and device for anti-flutter structural parameters of wind turbine blade

Abstract

The invention discloses a quick optimization design method and device for anti-flutter structural parameters of a wind turbine blade, and relates to the technical field of wind turbine blades, and the method comprises the following steps: constructing a blade aeroelastic characteristic equation to respectively determine the blade critical flutter speed of a plurality of to-be-optimized structural parameter combinations of a to-be-designed blade; carrying out standard orthogonal polynomial approximate expansion on the initial parameter-critical flutter velocity high-dimensional expansion model, converting coefficient solution into a 1-norm minimization problem, and carrying out solution based on each to-be-optimized structure parameter combination and each blade critical flutter velocity to determine a target parameter-critical flutter velocity high-dimensional expansion model; calculating the sensitivity index of each to-be-optimized structure parameter to determine a key to-be-optimized structure parameter; taking maximization of the critical flutter speed of the blade as a target, adjusting the parameter gradient of each key to-be-optimized structure to solve a target parameter-critical flutter speed high-dimensional expansion model, and determining a target optimization structure parameter combination. According to the scheme, the optimal anti-flutter structure parameter design can be quickly determined.

Description

Technical Field

[0001] This invention relates to the field of wind turbine blade technology, and in particular to a method and apparatus for rapid optimization design of structural parameters for wind turbine blades to resist flutter. Background Technology

[0002] Wind power is crucial as a clean and renewable energy source in the global energy structure transformation. Among them, wind turbine blades, as the core components that convert wind energy into mechanical energy, directly affect the power generation efficiency and operational safety of the entire machine through their aerodynamic performance and structural strength. The development trend of wind turbines towards larger single-unit capacity and larger blade size has significantly increased blade flexibility, leading to increasingly prominent aeroelastic problems such as flutter and instability, which places higher demands on blade structural design.

[0003] Flutter, a typical fluid-structure interaction dynamic phenomenon, can cause catastrophic structural damage in a short period of time due to the coupling instability between aerodynamic forces and structural dynamics. It has become one of the key bottlenecks restricting the lightweight and large-scale development of blades. Existing blade structural parameter design methods mainly rely on iterative optimization based on the coupling of finite element simulation and optimization algorithms when dealing with flutter problems. However, each round of parameter iteration requires re-establishing the flutter boundary, which has limitations such as complex overall process and high computational cost, resulting in low efficiency in the design of flutter-resistant structural parameters for wind turbine blades. Summary of the Invention

[0004] This invention provides a method and apparatus for rapid optimization design of structural parameters for wind turbine blade flutter resistance, which solves the technical problem that existing blade structural parameter design methods have limitations such as complex overall process and large computational overhead when dealing with flutter resistance, resulting in low design efficiency of structural parameters for wind turbine blade flutter resistance.

[0005] The first aspect of this invention provides a rapid optimization design method for structural parameters of wind turbine blades to resist flutter, comprising:

[0006] The aeroelastic characteristic equation of the blade was constructed using Reissner strain theory and blade element momentum theory.

[0007] Based on the aeroelastic characteristic equation of the blade, the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade to be designed is determined respectively.

[0008] The initial parameter-critical flutter velocity high-dimensional expansion model is approximated by standard orthogonal polynomials and the coefficient solution is transformed into a L1 norm minimization problem. Based on the combination of structural parameters to be optimized and the critical flutter velocity of each blade, the target parameter-critical flutter velocity high-dimensional expansion model is determined.

[0009] The sensitivity index of each structural parameter to be optimized is calculated based on the target parameter-critical flutter velocity high-dimensional expansion model, and the key structural parameters to be optimized are determined based on each sensitivity index.

[0010] With the goal of maximizing the critical flutter velocity of the blade, the key structural parameters to be optimized are adjusted within the feasible domain of parameter design. The gradient solution of the high-dimensional expansion model of the target parameter-critical flutter velocity is used to determine the combination of target optimized structural parameters.

[0011] Furthermore, the aeroelastic characteristic equation of the blade, constructed based on Reissner strain theory and blade element momentum theory, includes:

[0012] The dynamic equations of the blade structure were constructed based on Reissner strain theory;

[0013] Aerodynamic linearization was performed by combining blade element momentum theory with the BL dynamic stall model, and the blade aerodynamic linearization equation was constructed.

[0014] By combining the blade structural dynamics equations and the blade aerodynamic linearization equations, the aeroelastic characteristic equations of the blade are determined.

[0015] Furthermore, the step of determining the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade under design, based on the blade aeroelastic characteristic equation, includes:

[0016] For the blade to be designed with a preset tip speed ratio, the blade rotation speed is adjusted under multiple combinations of structural parameters to be optimized, and the corresponding aeroelastic damping ratio is calculated by solving the aeroelastic characteristic equation of the blade using an orthogonal triangular decomposition algorithm.

[0017] The combined structural parameters to be optimized are used at the blade rotation speed where the aeroelastic damping ratio is zero as the corresponding critical flutter speed of the blade.

[0018] Furthermore, the high-dimensional expansion model of the initial parameter-critical flutter velocity is approximately expanded using standard orthogonal polynomials, and the coefficient solution is transformed into a L1 norm minimization problem. Based on the combination of structural parameters to be optimized and the critical flutter velocity of each blade, the target parameter-critical flutter velocity high-dimensional expansion model is determined, including:

[0019] A second-order high-dimensional model expansion was used to construct an initial parameter-critical flutter velocity high-dimensional expansion model;

[0020] The initial parameter-critical flutter velocity high-dimensional expansion model is approximately expanded using standard orthogonal polynomials to determine multi-layer component functions.

[0021] The problem of finding the coefficients of each component function is transformed into a problem of minimizing the 1-norm.

[0022] The associated combination of structural parameters to be optimized and the critical flutter velocity of the blade are used as sample points to divide the training set and the validation set, and training observation matrix and validation observation matrix are constructed based on the training set and the validation set, respectively.

[0023] The initial sparse coefficients are determined by solving the L1 norm minimization problem using the training observation matrix, and the convergence of the error is determined by substituting the verification observation matrix into the L1 norm minimization problem.

[0024] If convergence is not achieved, the sample points of the validation set are updated to the training set to obtain a new training set. After determining the new validation set, a new training observation matrix and a new validation observation matrix are constructed until the error converges. The corresponding initial sparse coefficients are then used as the target sparse coefficients.

[0025] After substituting the target sparse coefficients into the corresponding component functions, the component functions are added together to obtain the target parameter-critical flutter velocity high-dimensional expansion model.

[0026] Furthermore, the sensitivity index includes the first-order sensitivity index and / or the total sensitivity index.

[0027] Furthermore, the error is the root mean square error.

[0028] The second aspect of this invention provides a rapid optimization design device for structural parameters of wind turbine blades to resist flutter, comprising:

[0029] The aeroelastic equation construction module is used to construct the aeroelastic characteristic equations of the blade using Reissner strain theory and blade element momentum theory.

[0030] The flutter velocity determination module is used to determine the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade to be designed, based on the blade aeroelastic characteristic equation.

[0031] The model training and optimization module is used to approximate the initial parameter-critical flutter velocity high-dimensional expansion model by standard orthogonal polynomials and transform the coefficient solution into a L1 norm minimization problem. Based on the combination of structural parameters to be optimized and the critical flutter velocity of each blade, the module determines the target parameter-critical flutter velocity high-dimensional expansion model.

[0032] The key parameter identification module is used to calculate the sensitivity index of each structural parameter to be optimized according to the target parameter-critical flutter velocity high-dimensional expansion model, and to determine the key structural parameters to be optimized based on each sensitivity index;

[0033] The parameter optimization design module is used to adjust the key structural parameters to be optimized within the feasible domain of parameter design with the goal of maximizing the critical flutter velocity of the blade, and to determine the combination of target optimized structural parameters by solving the gradient of the high-dimensional expansion model of the target parameter-critical flutter velocity.

[0034] A computer device provided in a third aspect of the present invention includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the rapid optimization design method for structural parameters of wind turbine blades to resist flutter as described in any of the preceding claims.

[0035] The fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed, it implements the rapid optimization design method for structural parameters of wind turbine blades to resist flutter as described in any of the preceding claims.

[0036] The fifth aspect of the present invention provides a computer program product, comprising a computer program / instruction, wherein when the computer program / instruction is executed by a processor, it implements the rapid optimization design method for structural parameters of wind turbine blades to resist flutter as described in any of the preceding claims.

[0037] As can be seen from the above technical solutions, the present invention has the following advantages:

[0038] The above-mentioned solution of the present invention provides a rapid optimization design method for structural parameters of wind turbine blades to resist flutter, including: constructing the aeroelastic characteristic equation of the blade using Reissner strain theory and blade element momentum theory; determining the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade based on the aeroelastic characteristic equation of the blade; approximating the initial parameter-critical flutter velocity high-dimensional expansion model with standard orthogonal polynomials and transforming the coefficient solution into a norm 1 minimization problem, solving the problem based on each combination of structural parameters to be optimized and each blade critical flutter velocity, and determining the target parameter-critical flutter velocity high-dimensional expansion model; calculating the sensitivity index of each structural parameter to be optimized according to the target parameter-critical flutter velocity high-dimensional expansion model, and determining the key structural parameters to be optimized based on each sensitivity index; adjusting each key structural parameter to be optimized within the feasible domain of parameter design with the goal of maximizing the blade critical flutter velocity, and determining the target optimized structural parameter combination by solving the gradient of the target parameter-critical flutter velocity high-dimensional expansion model. Based on the above scheme, the flutter velocity of the nonlinearly deformed blade can be estimated relatively accurately by constructing the blade aeroelastic characteristic equation. By building a parameter-critical flutter velocity high-dimensional expansion model as a global proxy model between blade structural parameters and flutter performance, the key structural parameters to be optimized that affect the critical flutter velocity of the blade can be identified. This can significantly reduce the number of simulation calls and accelerate the design process. Compared with the method of directly performing simulation, the optimal anti-flutter structural parameter design can be determined more quickly. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating the steps of a rapid optimization design method for structural parameters of wind turbine blades to resist flutter, as provided in Embodiment 1 of the present invention.

[0041] Figure 2 This is an execution architecture diagram of a rapid optimization design method for structural parameters of wind turbine blades to resist flutter, provided in Embodiment 1 of the present invention.

[0042] Figure 3 This is a structural block diagram of a rapid optimization design device for anti-flutter structural parameters of wind turbine blades provided in Embodiment 2 of the present invention. Detailed Implementation

[0043] This invention provides a method and apparatus for rapid optimization design of structural parameters for wind turbine blade flutter resistance, which addresses the limitations of existing blade structural parameter design methods in dealing with flutter resistance, such as complex overall process and high computational overhead, resulting in low design efficiency of structural parameters for wind turbine blade flutter resistance.

[0044] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0045] Please see Figure 1 and Figure 2 Embodiment 1 of the present invention provides a rapid optimization design method for structural parameters of wind turbine blades to resist flutter, including:

[0046] Step 101: Construct the aeroelastic characteristic equation of the blade using Reissner strain theory and blade element momentum theory.

[0047] It should be noted that Reissner strain theory is a type of beam theory, while blade element momentum theory is a theory for analyzing the aerodynamic performance of wind turbine blades. For details, please refer to existing technologies. In this embodiment, Reissner strain theory can be used to describe the deformation and strain relationship of the blade structure, while blade element momentum theory can be used to analyze the aerodynamic characteristics of the blade. By combining the two, a blade aeroelastic characteristic equation describing the aeroelasticity of the blade under the coupling of structural elastic deformation and airflow excitation can be constructed.

[0048] In one specific embodiment of this example, step 101 includes the following sub-steps:

[0049] S11. Construct the blade structure dynamics equations based on Reissner strain theory;

[0050] S12. Aerodynamic linearization is performed by combining blade element momentum theory with the BL dynamic stall model, and the blade aerodynamic linearization equation is constructed.

[0051] S13. Combine the blade structure dynamics equations and the blade aerodynamic linearization equations to determine the blade aeroelastic characteristic equations.

[0052] It should be noted that, based on Reinssner strain theory, the deformation of the blade can be described by rigid body displacement and deformation displacement, and its strain vector... It can be represented as:

[0053]

[0054] In the formula, This is the axial strain vector. Torsion vector, Let be a rotation matrix. It is a transpose operator. Let be the position vector of the line connecting the midpoint of the beam in space. This is the derivative of the reference blade section along the Z-axis coordinate. is the unit vector in the global coordinate system. Operators for extracting antisymmetric tensors into vectors. The skew-symmetric matrix corresponding to the rotation matrix;

[0055] The relationship between stress and strain can be expressed as:

[0056]

[0057] In the formula, For cross-sectional shear force and axial force, For the bending moment and torque of the cross section, The material stiffness matrix is ​​6×6. The strain vector;

[0058] Using Hamilton's principle, and considering kinetic energy, external forces, and aerodynamic loads, the virtual work balance equation is obtained: In the formula, For variational notation, Internal energy is a form of empty skill. It is an external effort that is ultimately ineffective. This is an external, unreal effort; among them, , , In the formula, For the blade length, To integrate along the blade axis, Let be the position vector of the line connecting the midpoint of the beam in space. It is a transpose operator. To distribute external forces, For distributed torque, This is an imaginary angular displacement. For rotational inertia, Angular acceleration;

[0059] Discrete the blade into several nodes, differentiate, integrate, and linearize to obtain the discrete equilibrium equations of the equivalent nodal internal forces, external forces, and inertial forces of the blade, i.e., the blade structural dynamic equations are: In the formula, The structural mass matrix, Here is the structural damping matrix. Here is the structural stiffness matrix. These are the Gauss-Lobatto integral coefficients. To locate the matrix for differential integration, This represents the virtual displacement of the nodes. Virtual velocity, It is a virtual acceleration. For weak forms, calculate the product length coefficient. This is the variation of external force with respect to virtual displacement;

[0060] When performing blade flutter analysis, external forces The aerodynamic force can be calculated using the blade element momentum theory combined with the BL dynamic stall model. Therefore, the linearized aerodynamic force, i.e., the linearized equation of the blade aerodynamic force, can be expressed as: In the formula, The mass matrix is ​​obtained by linearizing the aerodynamic components. This is the damping matrix obtained by linearizing the aerodynamic components. This is the aerodynamic stiffness matrix obtained by linearizing the aerodynamic components. For state variable coefficients, The cross-section is used to describe the state variables of the gas flow and vortex shedding on the blade. For variational notation;

[0061] By combining the above blade structural dynamics equations and the linearized blade aerodynamic equations, we can obtain the blade aeroelastic characteristic equations. The aeroelastic characteristic equation of a blade considering blade deformation can be written as:

[0062]

[0063]

[0064] In the formula, The system's generalized stiffness matrix. For eigenvalues, For the system's generalized mass matrix, For feature vectors, It is the identity matrix. For the mass of the gas bullet, For damping, Let be the stiffness matrix, where , , , , and The mass matrix, damping matrix, and stiffness matrix generated by linearizing the state variables satisfy the following relationship: In the formula, The time rate of change of the state variable characterizing the effect of vortex shedding and flow separation on blade performance. This is the dynamic stall delay parameter (used to quantify the delay characteristics of unsteady aerodynamic response).

[0065] Step 102: Based on the aeroelastic characteristic equation of the blade, determine the critical flutter velocity of the blade corresponding to the multiple combinations of structural parameters to be optimized for the blade to be designed.

[0066] The combination of structural parameters to be optimized refers to different combinations of adjustable structural parameters that affect the flutter performance of the blade. These parameters affect the aerodynamic performance and structural dynamics of the blade. For example, ply angle, twist angle, chord length, material density, blade stiffness, and center of mass can be considered. In addition, due to the slender structure of the blade itself, these parameters should also take into account the local differences in the axial position of the blade. The same parameter will have different effects on flutter in different parts. For example, the parameters at the blade root, blade middle, and blade tip may have completely opposite effects on flutter.

[0067] The critical flutter speed of a blade refers to the speed at which a blade becomes unstable and flutters.

[0068] It should be noted that for different combinations of structural parameters to be optimized to form different designs of blades, the critical velocity at which the blade will flutter under each design of the structural parameters to be optimized is solved by the blade aeroelastic characteristic equation.

[0069] In one specific embodiment of this example, step 102 includes the following sub-steps:

[0070] For the blade to be designed with a preset tip speed ratio, the blade rotation speed is adjusted under multiple combinations of structural parameters to be optimized, and the corresponding aeroelastic damping ratio is calculated by solving the aeroelastic characteristic equation of the blade using an orthogonal triangular decomposition algorithm.

[0071] The combined structural parameters to be optimized are used at the blade rotation speed where the aeroelastic damping ratio is zero as the corresponding critical flutter speed of the blade.

[0072] Tip speed ratio refers to the ratio of the linear velocity at the tip of a wind turbine blade to the speed of the incoming wind. The tip speed is related to the blade rotation speed.

[0073] It should be noted that when the blade operates under a constant tip speed ratio, it means that the relationship between the blade speed and the wind speed is proportional. Increasing the blade speed also means increasing the wind speed. Therefore, by adjusting a series of different blade speeds, the aeroelastic damping ratio of the blade can be obtained by solving the aeroelastic characteristic equation of the blade using the orthogonal triangular decomposition algorithm. If the aeroelastic damping ratio of zero cannot be obtained directly, the corresponding critical flutter velocity of the blade can be obtained by interpolating multiple aeroelastic damping ratios obtained by combining various structural parameters to be optimized.

[0074] When solving the aeroelastic characteristic equation of a blade using the orthogonal trigonometric decomposition algorithm, eigenvalues ​​can be calculated, and the aeroelastic damping ratio can be determined accordingly.

[0075]

[0076] In the formula, For the frequency of the aeroelastic force, The imaginary part of the eigenvalues. For the first 1 eigenvalue, The aeroelastic damping ratio, The real part of the eigenvalue. The index of the feature value.

[0077] Step 103: The initial parameter-critical flutter velocity high-dimensional expansion model is approximately expanded using standard orthogonal polynomials, and the coefficient solution is transformed into a L1 norm minimization problem. Based on the combination of structural parameters to be optimized and the critical flutter velocity of each blade, the target parameter-critical flutter velocity high-dimensional expansion model is determined.

[0078] In one specific embodiment of this example, step 103 includes the following sub-steps:

[0079] A second-order high-dimensional model expansion was used to construct an initial parameter-critical flutter velocity high-dimensional expansion model;

[0080] The initial parameter-critical flutter velocity high-dimensional expansion model is approximately expanded using standard orthogonal polynomials to determine multi-layer component functions.

[0081] The problem of finding the coefficients of each component function is transformed into a problem of minimizing the 1-norm.

[0082] The associated combination of structural parameters to be optimized and the critical flutter velocity of the blade are used as sample points to divide the training set and the validation set, and training observation matrix and validation observation matrix are constructed based on the training set and the validation set, respectively.

[0083] The initial sparse coefficients are determined by solving the L1 norm minimization problem using the training observation matrix, and the convergence of the error is determined by substituting the verification observation matrix into the L1 norm minimization problem.

[0084] If convergence is not achieved, the sample points of the validation set are updated to the training set to obtain a new training set. After determining the new validation set, a new training observation matrix and a new validation observation matrix are constructed until the error converges. The corresponding initial sparse coefficients are then used as the target sparse coefficients.

[0085] After substituting the target sparse coefficients into the corresponding component functions, the component functions are added together to obtain the target parameter-critical flutter velocity high-dimensional expansion model.

[0086] The parameter-critical flutter velocity high-dimensional expansion model refers to a model that uses high-dimensional expansion to characterize the mapping relationship between the structural parameters to be optimized and the critical flutter velocity of the blade.

[0087] It should be noted that during the structural parameter optimization design of wind turbine blades, changes in structural parameters will directly affect the critical flutter velocity. At this point, including... One structural parameter to be optimized The aeroelastic characteristic equation can be expressed as: , This refers to the stable deformation position of a blade when it is subjected to a series of forces during operation and the internal and external forces are in balance. This directly changes the modal characteristics of the blade. In practice, the displacement response of each node of the wind turbine blade can be observed with the blade root centroid as the origin of the coordinate system. After the time domain response stabilizes, the initial unstable data is filtered out and the time domain average of each node position is performed to obtain the equilibrium deformation position of the blade.

[0088] To reduce the number of simulation calls and accelerate the design process, the relationship between the input structural parameters to be optimized and the critical flutter velocity of the blade can be approximated using high-dimensional model expansion (HDMR) as follows:

[0089]

[0090] In the formula, This is the critical flutter velocity of the blade. For the structural parameters to be optimized, , and Index of structural parameters to be optimized , and , The number of structural parameters to be optimized. These are zero-order terms and constant terms, typically the average or median response of the physical model within its domain. This is a first-order term used to evaluate the independent contribution of each input parameter to the response. This is a second-order term used to evaluate the coupling effect between the two parameters after eliminating the independent influence of the input parameters; This is a third-order term used to evaluate the coupling effect among the three parameters after excluding the independent influence of the input parameters and the two-parameter coupling effect; and so on.

[0091] In fact, the higher the order of the terms, the smaller the response of the model. To reduce computational costs, a second-order high-dimensional model expansion is used to construct an initial parameter-critical flutter velocity high-dimensional expansion model, whose terms can be expressed as:

[0092]

[0093]

[0094]

[0095] In the formula, for A 3D hypercube space, for Not included Item, for Not included The terms thus decompose a high-dimensional problem into a series of low-dimensional subproblems;

[0096] For each layer of component functions, an approximate expansion can be performed using orthogonal polynomials, which strictly correspond to the distribution type of the input variables. For example, if the parameters follow a standard normal distribution, Hermite polynomials are used; if they follow a uniform distribution, Legendre polynomials are used. Thus, the second-order HDMR can be approximated by orthogonal polynomials as follows:

[0097]

[0098] In the formula, For about The orthogonal polynomial of order 1 The coefficients are the corresponding polynomial basis. It is the highest order of the polynomial. index for orthogonal polynomials , index for orthogonal polynomials , index for orthogonal polynomials ;

[0099] To improve the component coefficient of each layer To improve computational efficiency, we can transform the problem into a 1-norm minimization problem: In the formula, It is a sparse coefficient vector. This is the critical flutter velocity response vector of the blade. For error tolerance, For the observation matrix, for 3D real vector space, It is a norm of 1. It is a 2-norm; for the observation matrix An adaptive random sampling method can be used to obtain a smaller number of samples. The specific operation is as follows: using the associated combination of structural parameters to be optimized and the critical flutter velocity of the blade as sample points, initial random sampling is performed. Using 10 sample points as the training set, construct the initial training observation matrix. Substituting these values ​​into the above equation yields the sparsity coefficients, which can then be further calculated by adding random sampling. Using a set of 100 sample points as the validation set, when the calculated error is less than the convergence accuracy... If the construction of the component is successful, then the component is considered to have been successfully constructed; otherwise, this component is considered to have been successfully constructed. Each sample point is added to the training set, and a new training observation matrix is ​​constructed. Solve for the coefficients and resample. The accuracy is verified using individual sample points until convergence is achieved. Finally, the component functions of each layer are added together to obtain an approximate relationship between the input quantity and the critical flutter velocity, which is the target parameter-critical flutter velocity high-dimensional expansion model.

[0100] In one specific embodiment of this example, the error is the root mean square error. It is understood that other error metrics can also be used to calculate the error, and this embodiment does not impose any limitations on this.

[0101] It is worth mentioning that this embodiment introduces a 1-norm minimization method, which can be used to calculate the main coefficients when the sample size is less than the unknown coefficients, improving computational efficiency and providing high noise resistance. This embodiment employs a dynamic sampling strategy, which is particularly suitable for large-scale parameter optimization problems. In actual modeling, different parameters have varying degrees of influence on the response, leading to differences in the model's convergence speed and sampling requirements for each parameter. By combining dynamic sampling with cross-validation using the root mean square error index, the approximation accuracy can be continuously checked during model construction until the convergence criterion is met, effectively avoiding unnecessary sample waste. Dimensionality reduction techniques are used, supporting simultaneous optimization design of multiple parameters. Furthermore, addressing the redundant modeling problem of traditional HDMR methods in high-dimensional problems, this embodiment introduces a preliminary sensitivity screening mechanism during the dynamic construction of first-order terms. Since the orthogonal polynomials have been standardized in the preprocessing stage, the first-order terms of each parameter have a unified weight benchmark. Therefore, the influence of parameters on the response can be initially judged by their coefficient magnitude. For variables that have no significant effect on the flutter response, their first-order terms are selected based on the initial sample size. The parameter has basically converged, and its corresponding coefficient is close to zero, indicating that its impact on the system output is extremely weak and can be accurately identified in the early stage. At this time, the parameter can be removed from the parameter space, and its corresponding sensitivity index can be directly assigned to zero to avoid the invalid construction of subsequent higher-order terms, thereby effectively saving computing resources and improving modeling efficiency.

[0102] Step 104: Calculate the sensitivity index of each structural parameter to be optimized according to the target parameter-critical flutter velocity high-dimensional expansion model, and determine the key structural parameters to be optimized based on each sensitivity index.

[0103] The sensitivity index refers to the relative importance of each structural parameter to be optimized on the critical flutter velocity of the blade.

[0104] Key structural parameters to be optimized refer to structural parameters that have a significant impact on the critical flutter velocity of the blades.

[0105] It should be noted that, due to the characteristics of orthogonal polynomials, the sensitivity index can be determined by the coefficients of HDMR to identify key design parameters affecting flutter, which can then be prioritized for adjustment during optimization.

[0106] In one specific embodiment of this example, the sensitivity index includes the first-order sensitivity index and / or the total sensitivity index.

[0107] It should be noted that the total variance of the high-dimensional expansion model of the target parameter-critical flutter velocity can be expressed by its coefficients as follows:

[0108]

[0109] In the formula, For the total variance, For variance calculation;

[0110] For any one of the structural parameters to be optimized, the first-order sensitivity exponent and the total sensitivity exponent can be expressed by coefficients as follows:

[0111]

[0112]

[0113] In the formula, The first-order sensitivity index, The overall sensitivity index; The larger the value, the greater the independent influence of this parameter on flutter. Much larger This indicates that the interaction effect of this parameter is relatively large;

[0114] In some feasible implementations, the structural parameters to be optimized with a first-order sensitivity index greater than the first-order sensitivity threshold can be identified as key structural parameters to be optimized. Alternatively, the structural parameters to be optimized with a difference between the total sensitivity index and the first-order sensitivity index greater than the sensitivity difference threshold can be identified as key structural parameters to be optimized. This embodiment does not impose any restrictions on this.

[0115] Step 105: With the goal of maximizing the critical flutter velocity of the blade, adjust each of the key structural parameters to be optimized within the feasible domain of parameter design, and determine the combination of target optimized structural parameters by solving the gradient of the high-dimensional expansion model of the target parameter-critical flutter velocity.

[0116] It should be noted that after identifying the key structural parameters that significantly affect flutter response, the goal of the optimization design is to find the optimal combination of these parameters that maximizes flutter performance.

[0117]

[0118] In the formula, Optimize the combination of structural parameters to achieve the target. Design a feasible region for the parameters. For the structural parameters to be optimized, This is the critical flutter velocity of the blade;

[0119] Since HDMR is a model with explicit representation, continuous differentiability, and low computational cost, it can be directly solved using the gradient method. The gradient at point can be expressed as:

[0120]

[0121] Given an initial point Update using the following iterative method: ,in For the first The step size of the next iteration, when the convergence threshold is met. At that time, that is When the iteration is complete, then the iteration is finished. This is the optimal parameter combination.

[0122] In this embodiment of the invention, the flutter velocity of a nonlinearly deformed blade can be estimated relatively accurately by constructing the aeroelastic characteristic equation of the blade. By building a high-dimensional expansion model of parameter-critical flutter velocity as a global proxy model between blade structural parameters and flutter performance, the number of simulation calls can be significantly reduced, the design process can be accelerated, and key structural parameters that affect the critical flutter velocity of the blade can be identified. As a result, designers can clearly know which parameters determine flutter performance and how to adjust these parameters to improve the blade's flutter resistance. Compared with the method of directly performing simulation, the optimal flutter resistance structural parameter design can be determined more quickly.

[0123] Please see Figure 3 The present invention provides a rapid optimization design device for structural parameters of wind turbine blades to resist flutter, comprising:

[0124] The aeroelastic equation construction module 301 is used to construct the aeroelastic characteristic equation of the blade using Reissner strain theory and blade element momentum theory.

[0125] The flutter velocity determination module 302 is used to determine the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade to be designed, based on the blade aeroelastic characteristic equation.

[0126] The model training and optimization module 303 is used to approximate the initial parameter-critical flutter velocity high-dimensional expansion model by standard orthogonal polynomials and transform the coefficient solution into a L1 norm minimization problem. Based on the combination of structural parameters to be optimized and the critical flutter velocity of each blade, the module solves the problem to determine the target parameter-critical flutter velocity high-dimensional expansion model.

[0127] The key parameter identification module 304 is used to calculate the sensitivity index of each structural parameter to be optimized according to the target parameter-critical flutter velocity high-dimensional expansion model, and to determine the key structural parameters to be optimized based on each sensitivity index.

[0128] The parameter optimization design module 305 is used to adjust the key structural parameters to be optimized within the feasible domain of parameter design with the goal of maximizing the critical flutter velocity of the blade, and to determine the combination of target optimized structural parameters by solving the gradient of the high-dimensional expansion model of the target parameter-critical flutter velocity.

[0129] Furthermore, the gas elasticity equation construction module 301 is specifically used for:

[0130] The dynamic equations of the blade structure were constructed based on Reissner strain theory;

[0131] Aerodynamic linearization was performed by combining blade element momentum theory with the BL dynamic stall model, and the blade aerodynamic linearization equation was constructed.

[0132] By combining the blade structural dynamics equations and the blade aerodynamic linearization equations, the aeroelastic characteristic equations of the blade are determined.

[0133] Furthermore, the flutter velocity determination module 302 is specifically used for:

[0134] For the blade to be designed with a preset tip speed ratio, the blade rotation speed is adjusted under multiple combinations of structural parameters to be optimized, and the corresponding aeroelastic damping ratio is calculated by solving the aeroelastic characteristic equation of the blade using an orthogonal triangular decomposition algorithm.

[0135] The combined structural parameters to be optimized are used at the blade rotation speed where the aeroelastic damping ratio is zero as the corresponding critical flutter speed of the blade.

[0136] Furthermore, the model training optimization module 303 is specifically used for:

[0137] A second-order high-dimensional model expansion was used to construct an initial parameter-critical flutter velocity high-dimensional expansion model;

[0138] The initial parameter-critical flutter velocity high-dimensional expansion model is approximately expanded using standard orthogonal polynomials to determine multi-layer component functions.

[0139] The problem of finding the coefficients of each component function is transformed into a problem of minimizing the 1-norm.

[0140] The associated combination of structural parameters to be optimized and the critical flutter velocity of the blade are used as sample points to divide the training set and the validation set, and training observation matrix and validation observation matrix are constructed based on the training set and the validation set, respectively.

[0141] The initial sparse coefficients are determined by solving the L1 norm minimization problem using the training observation matrix, and the convergence of the error is determined by substituting the verification observation matrix into the L1 norm minimization problem.

[0142] If convergence is not achieved, the sample points of the validation set are updated to the training set to obtain a new training set. After determining the new validation set, a new training observation matrix and a new validation observation matrix are constructed until the error converges. The corresponding initial sparse coefficients are then used as the target sparse coefficients.

[0143] After substituting the target sparse coefficients into the corresponding component functions, the component functions are added together to obtain the target parameter-critical flutter velocity high-dimensional expansion model.

[0144] Furthermore, the sensitivity index includes the first-order sensitivity index and / or the total sensitivity index.

[0145] Furthermore, the error is the root mean square error.

[0146] This invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program; when the computer program is executed by the processor, the processor performs the steps of the rapid optimization design method for the structural parameters of wind turbine blades to resist flutter as described in any of the above embodiments.

[0147] This invention also provides a computer-readable storage medium storing a computer program / instruction thereon, which, when executed by a processor, implements the steps of the rapid optimization design method for the structural parameters of wind turbine blades to resist flutter as described in any of the above embodiments.

[0148] This invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the rapid optimization design method for the structural parameters of wind turbine blades to resist flutter as described in any of the above embodiments.

[0149] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0151] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0153] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0154] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rapid optimization design method for structural parameters of wind turbine blades to resist flutter, characterized in that, include: The aeroelastic characteristic equation of the blade was constructed using Reissner strain theory and blade element momentum theory. Based on the aeroelastic characteristic equation of the blade, the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade to be designed is determined respectively. The initial parameter-critical flutter velocity high-dimensional expansion model is approximated by standard orthogonal polynomials and the coefficient solution is transformed into a L1 norm minimization problem. Based on the combination of structural parameters to be optimized and the critical flutter velocity of each blade, the target parameter-critical flutter velocity high-dimensional expansion model is determined. The sensitivity index of each structural parameter to be optimized is calculated based on the target parameter-critical flutter velocity high-dimensional expansion model, and the key structural parameters to be optimized are determined based on each sensitivity index. With the goal of maximizing the critical flutter velocity of the blade, the key structural parameters to be optimized are adjusted within the feasible domain of parameter design. The gradient solution of the high-dimensional expansion model of the target parameter-critical flutter velocity is used to determine the combination of target optimized structural parameters.

2. The rapid optimization design method for structural parameters of wind turbine blades to resist flutter according to claim 1, characterized in that, The aeroelastic characteristic equations for blades, constructed based on Reissner strain theory and blade element momentum theory, include: The dynamic equations of the blade structure were constructed based on Reissner strain theory; Aerodynamic linearization was performed by combining blade element momentum theory with the BL dynamic stall model, and the blade aerodynamic linearization equation was constructed. By combining the blade structural dynamics equations and the blade aerodynamic linearization equations, the aeroelastic characteristic equations of the blade are determined.

3. The rapid optimization design method for structural parameters of wind turbine blades to resist flutter according to claim 1, characterized in that, The step of determining the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade under design, based on the blade aeroelastic characteristic equation, includes: For the blade to be designed with a preset tip speed ratio, the blade rotation speed is adjusted under multiple combinations of structural parameters to be optimized, and the corresponding aeroelastic damping ratio is calculated by solving the aeroelastic characteristic equation of the blade using an orthogonal triangular decomposition algorithm. The combined structural parameters to be optimized are used at the blade rotation speed where the aeroelastic damping ratio is zero as the corresponding critical flutter speed of the blade.

4. The rapid optimization design method for structural parameters of wind turbine blades to resist flutter according to claim 1, characterized in that, The high-dimensional expansion model of the initial parameter-critical flutter velocity is approximated by standard orthogonal polynomial expansion and the coefficient solution is transformed into a L1 norm minimization problem. Based on the combinations of structural parameters to be optimized and the critical flutter velocities of the blades, the target parameter-critical flutter velocity high-dimensional expansion model is determined, including: A second-order high-dimensional model expansion was used to construct an initial parameter-critical flutter velocity high-dimensional expansion model; The initial parameter-critical flutter velocity high-dimensional expansion model is approximately expanded using standard orthogonal polynomials to determine multi-layer component functions. The problem of finding the coefficients of each component function is transformed into a problem of minimizing the 1-norm. The associated combination of structural parameters to be optimized and the critical flutter velocity of the blade are used as sample points to divide the training set and the validation set, and training observation matrix and validation observation matrix are constructed based on the training set and the validation set, respectively. The initial sparse coefficients are determined by solving the L1 norm minimization problem using the training observation matrix, and the convergence of the error is determined by substituting the verification observation matrix into the L1 norm minimization problem. If convergence is not achieved, the sample points of the validation set are updated to the training set to obtain a new training set. After determining the new validation set, a new training observation matrix and a new validation observation matrix are constructed until the error converges. The corresponding initial sparse coefficients are then used as the target sparse coefficients. After substituting the target sparse coefficients into the corresponding component functions, the component functions are added together to obtain the target parameter-critical flutter velocity high-dimensional expansion model.

5. The rapid optimization design method for structural parameters of wind turbine blades to resist flutter according to claim 1, characterized in that, The sensitivity index includes the first-order sensitivity index and / or the total sensitivity index.

6. The rapid optimization design method for structural parameters of wind turbine blades to resist flutter according to claim 4, characterized in that, The error is the root mean square error.

7. A device for rapid optimization design of structural parameters for wind turbine blade flutter resistance, characterized in that, include: The aeroelastic equation construction module is used to construct the aeroelastic characteristic equations of the blade using Reissner strain theory and blade element momentum theory. The flutter velocity determination module is used to determine the critical flutter velocity of the blade corresponding to multiple combinations of structural parameters to be optimized for the blade to be designed, based on the blade aeroelastic characteristic equation. The model training and optimization module is used to approximate the initial parameter-critical flutter velocity high-dimensional expansion model by standard orthogonal polynomials and transform the coefficient solution into a L1 norm minimization problem. Based on the combination of structural parameters to be optimized and the critical flutter velocity of each blade, the module determines the target parameter-critical flutter velocity high-dimensional expansion model. The key parameter identification module is used to calculate the sensitivity index of each structural parameter to be optimized according to the target parameter-critical flutter velocity high-dimensional expansion model, and to determine the key structural parameters to be optimized based on each sensitivity index; The parameter optimization design module is used to adjust the key structural parameters to be optimized within the feasible domain of parameter design with the goal of maximizing the critical flutter velocity of the blade, and to determine the combination of target optimized structural parameters by solving the gradient of the high-dimensional expansion model of the target parameter-critical flutter velocity.

8. A computer device, characterized in that, The device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the rapid optimization design method for structural parameters of wind turbine blades to resist flutter as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the rapid optimization design method for structural parameters of wind turbine blades to resist flutter as described in any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the rapid optimization design method for structural parameters of wind turbine blades to resist flutter as described in any one of claims 1-6.

Images