Hybrid aerodynamic solving method and system based on wake interference and dynamic inflow time

Abstract

The application discloses a hybrid aerodynamic solving method and system based on wake interference and dynamic inflow time, and belongs to the technical field of wind turbine aerodynamic performance calculation. The method comprises the following steps: obtaining the operating condition parameters of a wind turbine; calculating a wake interference intensity index and a dynamic inflow time scale based on the operating condition parameters; calculating a comprehensive switching parameter based on the wake interference intensity index and the dynamic inflow time scale; comparing the comprehensive switching parameter with preset first and second threshold values; and when the comprehensive switching parameter is between the first and second threshold values, selecting a weighted mixing of a blade element momentum theory method and a generalized dynamic wake theory method for aerodynamic solving. Through the adaptive switching criterion based on the two parameters, the optimal balance between the calculation efficiency and the calculation accuracy is realized, and the calculation accuracy and numerical stability of the aerodynamic load under complex conditions are significantly improved.

Description

Technical Field

[0001] This invention belongs to the field of wind turbine aerodynamic performance calculation technology, and relates to a hybrid aerodynamic solution method and system based on wake interference and dynamic inflow time. Background Technology

[0002] Aerodynamic performance calculation of wind turbines is a core component of wind turbine design and simulation, and its accuracy directly affects the reliability of load assessment, control strategy design, and wind farm optimization layout. With the continuous increase in the capacity of individual wind turbines and the ongoing expansion of wind farm scale, the operating conditions faced by wind turbines are becoming increasingly complex and variable, placing higher demands on the accuracy, robustness, and computational efficiency of aerodynamic calculation methods.

[0003] Currently, the mainstream aerodynamic calculation methods mainly fall into two categories: Blade Element Momentum (BEM) theory and Generalized Dynamic Wake (GDW) theory. Blade Element Momentum (BEM) theory is the most widely used aerodynamic calculation method in engineering. Its basic principle is to divide the blade along its span into several blade elements, apply both momentum theory and blade element theory to each blade element to establish equilibrium equations, and then iteratively solve for the axial and tangential induction factors to calculate the aerodynamic loads at each section of the blade. However, the Blade Element Momentum (BEM) theory has several inherent limitations: it is based on the static equilibrium assumption and cannot accurately describe dynamic inflow effects, meaning that the transient response of aerodynamic loads exhibits hysteresis when wind speed changes rapidly; the Prandtl correction in BEM theory only considers tip losses and lacks sufficient modeling capability for complex wake structures (such as partial wake obstruction, wake deflection, and wake expansion); while the Glauret correction considers high induction ratio conditions, it is highly empirical and has limited applicability. Generalized dynamic wake theory is an aerodynamic calculation method based on unsteady potential flow theory. It describes the rotor wake dynamics by solving the partial differential equations of the potential function. It can naturally capture dynamic inflow effects and wake distortion and has higher calculation accuracy under complex conditions. However, its calculation cost is significantly higher than that of blade element momentum theory methods, and there is a lack of unified guidelines for the selection of the modal truncation order.

[0004] However, in practical applications, existing hybrid schemes combining blade element momentum theory and generalized dynamic wake theory have significant shortcomings. First, the switching criteria in existing hybrid schemes are mostly based on fixed thresholds (such as fixed induction factor thresholds), lacking comprehensive consideration of wake interference intensity and dynamic inflow timescales. This leads to unreasonable switching timing and difficulty in adapting to complex and variable operating conditions. Second, there is a lack of effective transfer mechanisms between the state variables of blade element momentum theory and generalized dynamic wake theory, resulting in discontinuous aerodynamic load changes during method switching, severely affecting computational stability. Furthermore, existing schemes lack cross-validation mechanisms to evaluate the reliability of the hybrid solution results, failing to guarantee that the computational results under all operating conditions are superior to those of a single method. Summary of the Invention

[0005] The purpose of this invention is to provide a hybrid aerodynamic solution method and system based on wake interference and dynamic inflow time, so as to solve the problems of unreasonable switching timing and poor adaptability in the existing technology.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a hybrid aerodynamic solution method based on wake interference and dynamic inflow time, comprising the following steps: Obtain the operating parameters of the wind turbine, and calculate the wake interference intensity index and dynamic inflow time scale based on the operating parameters; Based on the wake interference intensity index and the dynamic inflow time scale, the comprehensive switching parameters are calculated, and the comprehensive switching parameters are compared with a preset first threshold and a second threshold, wherein the first threshold is less than the second threshold. When the overall switching parameters are less than the first threshold, the blade element momentum theory method is selected for aerodynamic solution; when the overall switching parameters are greater than the second threshold, the generalized dynamic wake theory method is selected for aerodynamic solution. When the integrated switching parameters are between the first and second thresholds, a weighted hybrid of the leaf element momentum theory method and the generalized dynamic wake theory method is selected for aerodynamic solution.

[0007] Furthermore, the step of acquiring the operating condition parameters of the wind turbine and calculating the wake interference intensity index based on the operating condition parameters includes: The wake interference intensity index is calculated as follows:

[0008] in, The velocity loss is caused by the wake. The impeller swept area is used; for cases with partial wake obstruction, a weighted calculation is performed based on the obstruction area ratio.

[0009] in The impeller area affected by the wake.

[0010] Furthermore, the acquisition of the wind turbine's operating condition parameters, and the calculation of the dynamic inflow time scale based on these parameters, includes: Dynamic inflow timescale Definition:

[0011] in For wind speed change rate; for yaw transient conditions:

[0012] in The rate of change of yaw angle. This represents the impeller speed.

[0013] Furthermore, the calculation of integrated switching parameters based on the wake interference intensity index and the dynamic inflow time scale includes: Comprehensive switching parameters :

[0014] in, and For the weighting coefficients, satisfying , The indicator of wake interference intensity. For reference wake interference intensity, For dynamic inflow timescale, For reference, the dynamic inflow timescale.

[0015] Furthermore, when the integrated switching parameters are between the first and second thresholds, a weighted hybrid of the blade element momentum theory method and the generalized dynamic wake theory method is selected for aerodynamic solution, including: Solving using a weighted mixture method:

[0016] The weight function Use a Sigmoid-type smooth transition function:

[0017] in As a transition center, For transition bandwidth; this function ensures that during... and Smooth transition nearby; Weighting coefficient and Adaptively adjust based on current operating conditions:

[0018] When wake interference dominates Increase when the dynamic inflow effect dominates Increase.

[0019] Furthermore, when switching from the leaf element momentum theory method to the generalized dynamic wake theory method, a positive mapping of state variables is performed; Axial induction factor distribution of leaf element momentum theory BEM The modal coefficients are mapped to the generalized dynamic wake theory (GDW) via Legendre polynomial expansion:

[0020] Written in matrix form:

[0021] Where the mapping matrix The elements are:

[0022] This mapping is essentially a Galerkin projection, which projects the radially discrete induced distribution of BEM onto the modal space of GDW; When switching from the generalized dynamic wake theory method to the leaf element momentum theory method, the inverse mapping of state variables is performed; The potential function modal coefficients of the generalized dynamic wake theory (GDW) are reconstructed to obtain the induction factor distribution of the leaf element momentum theory (BEM):

[0023] Written in matrix form:

[0024] The inverse mapping matrix is ​​the generalized inverse of the forward mapping:

[0025] Introducing a relaxation smoothing factor :

[0026] in This is the state vector after the switch. The mapped state vector, The state vector of the previous time step; relaxation factor The transition time steps gradually increase from 0 to 1 over several time steps before and after the switch. Determined by the following formula:

[0027] in This refers to the transition time length.

[0028] Furthermore, a machine learning-based cross-validation algorithm was introduced to perform residual correction on the calculation results of BEM and GDW: For each time step, construct a feature vector. :

[0029] in The average induction factor, For the gradient of inducing factors, The tip speed ratio; Training a Gaussian process regression (GPR) model to predict the residuals of BEM and GDW:

[0030]

[0031] The predicted value of the GPR model is:

[0032] in For the feature matrix of the training set, For kernel function vectors, For the kernel matrix, For noise variance; using the radial basis function (RBF) kernel:

[0033] in For signal variance, For length scale parameters; K-fold cross-validation was used to train and validate the residual correction model. The training dataset is randomly divided into K subsets of similar size. Each subset is used as the validation set, and the remaining K-1 subsets are used as the training set. Calculate the mean absolute error (MAE) and coefficient of determination (R²) for K validations. If the average R² ≥ 0.9, the residual correction model is accepted; otherwise, the eigenvectors or kernel function parameters are adjusted. Corrected hybrid solution results:

[0034]

[0035] In actual calculations, when new high-precision reference data becomes available, it is added to the training set and the GPR model is retrained to achieve online updates of the residual correction model.

[0036] Secondly, the present invention provides a hybrid aerodynamic solution system based on wake interference and dynamic inflow time, comprising: The data acquisition module is used to acquire the operating condition parameters of the wind turbine, and calculate the wake interference intensity index and dynamic inflow time scale based on the operating condition parameters. The switching parameter calculation module is used to calculate the comprehensive switching parameters based on the wake interference intensity index and the dynamic inflow time scale, and compare the comprehensive switching parameters with a preset first threshold and a second threshold, wherein the first threshold is less than the second threshold. The selection module is used to select the blade element momentum theory method for aerodynamic solution when the integrated switching parameters are less than the first threshold, and to select the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are greater than the second threshold. The hybrid module is used to select a weighted mixture of the blade element momentum theory method and the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are between the first threshold and the second threshold.

[0037] Thirdly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the hybrid aerodynamic solution method based on wake interference and dynamic inflow time.

[0038] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the hybrid aerodynamic solution method based on wake interference and dynamic inflow time.

[0039] Compared with the prior art, the present invention has the following technical effects: This invention acquires the operating parameters of wind turbine generators and calculates wake interference intensity and dynamic inflow time scale based on these parameters. It extracts two key quantitative indicators from actual operating data, characterizing the degree of wake influence and the rate of dynamic inflow change, providing an objective and scientific basis for subsequent adaptive switching decisions. By calculating comprehensive switching parameters based on the wake interference intensity and dynamic inflow time scale, and comparing these parameters with preset first and second thresholds, a fusion assessment of current operating conditions is achieved. The comprehensive switching parameters simultaneously reflect the strength of wake interference and the rate of dynamic inflow, overcoming the limitations of traditional fixed-threshold methods that only consider a single factor. This allows switching decisions to adapt to complex and ever-changing operating environments.

[0040] By selecting the blade element momentum theory method for aerodynamic solution when the integrated switching parameters are less than the first threshold, the high computational efficiency of the blade element momentum theory method can be fully utilized under conditions of low wake interference and slow dynamic changes. By selecting the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are greater than the second threshold, the ability of the generalized dynamic wake theory method to capture dynamic inflow effects and wake distortion can be utilized under conditions of strong wake interference and fast dynamic changes, ensuring computational accuracy. By selecting a weighted hybrid of the blade element momentum theory method and the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are between the first and second thresholds, a smooth transition between the two methods is achieved, avoiding load jumps during method switching and ensuring the temporal continuity and numerical stability of aerodynamic load calculations.

[0041] This invention achieves an optimal balance between the computational efficiency of the blade element momentum theory method and the computational accuracy of the generalized dynamic wake theory method. Under complex working conditions, it significantly improves the aerodynamic load calculation accuracy compared to the single blade element momentum theory method, while the overall computational efficiency is significantly better than that of the generalized dynamic wake theory method used throughout the entire process. Attached Figure Description

[0042] Figure 1 This is a logic block diagram of the present invention.

[0043] Figure 2 Phase diagram for BEM / GDW adaptive switching criteria.

[0044] Figure 3 This is a schematic diagram of the BEM-GDW state variable transfer mechanism.

[0045] Figure 4 A comparison chart of aerodynamic load accuracy using BEM / GDW / hybrid methods.

[0046] Figure 5 This is a flowchart of the present invention. Detailed Implementation

[0047] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0048] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0049] The present invention will now be described in further detail with reference to the accompanying drawings: See Figure 5 In one embodiment of the present invention, a hybrid aerodynamic solution method based on wake interference and dynamic inflow time is provided to solve the technical problem of accurate calculation of aerodynamic loads of wind turbines under complex wake interference and dynamic inflow conditions.

[0050] Specifically, the hybrid aerodynamic solution method based on wake interference and dynamic inflow time of this invention includes the following steps: S1: Obtain the operating parameters of the wind turbine, and calculate the wake interference intensity index and dynamic inflow time scale based on the operating parameters.

[0051] S2: Based on the wake interference intensity index and the dynamic inflow time scale, calculate the comprehensive switching parameters, and compare the comprehensive switching parameters with the preset first threshold and second threshold, wherein the first threshold is less than the second threshold.

[0052] S3: When the integrated switching parameter is less than the first threshold, the blade element momentum theory method is selected for aerodynamic solution; when the integrated switching parameter is greater than the second threshold, the generalized dynamic wake theory method is selected for aerodynamic solution.

[0053] S4: When the integrated switching parameters are between the first threshold and the second threshold, a weighted mixture of the blade element momentum theory method and the generalized dynamic wake theory method is selected for aerodynamic solution.

[0054] For S1, the operating parameters of the wind turbine are the basic input data for aerodynamic calculations. These parameters originate from the wind turbine's data acquisition and monitoring system, including but not limited to the inflow wind speed at infinity measured by the wind speed sensor, the wind direction data measured by the wind direction sensor, the yaw angle fed back by the yaw system, the rotor speed measured by the speed sensor, and the blade pitch angle. Real-time acquisition and accurate acquisition of these parameters are prerequisites for achieving adaptive hybrid aerodynamic solutions. The wake interference intensity index is a key parameter used to quantify the influence of the upstream turbine's wake on the velocity field within the swept surface of the downstream turbine's rotor. In actual wind farm operation, after the upstream turbine captures wind energy, it forms a low-speed wake region behind it. When the wake propagates to the downstream turbine's rotor position, it causes the inflow wind speed to decrease and exhibit a non-uniform distribution, directly affecting the aerodynamic performance of the downstream turbine. The wake interference intensity index is obtained by integrating and averaging the velocity deficit caused by the wake over the entire swept area of ​​the impeller. A larger value indicates a more severe interference of the wake with the inflow conditions, and a greater degree of violation of the fundamental assumption of blade element momentum theory (uniform and steady inflow). The dynamic inflow timescale is a key parameter characterizing the rate of change in inflow conditions. When wind speed fluctuates rapidly or the unit performs yaw maneuvers, the flow field around the impeller requires a delay to re-establish equilibrium; this delay effect is the dynamic inflow effect. The dynamic inflow timescale reflects the relative relationship between the flow field response time and the inflow change timescale; a smaller value indicates a more drastic inflow change, a more significant dynamic inflow effect, and the quasi-static assumption of blade element momentum theory no longer applies. This step extracts the wake interference intensity index and the dynamic inflow timescale from the operating condition parameters, abstracting the complex physical conditions into two quantifiable numerical indicators, laying the foundation for subsequent adaptive switching decisions.

[0055] For S2, in the actual operation of wind turbines, wake interference and dynamic inflow effects often coexist and are coupled, making it difficult to fully reflect the aerodynamic calculation method selection requirements of the current operating conditions using a single indicator. Therefore, by weightedly fusing the wake interference intensity index and the dynamic inflow time scale, a comprehensive switching parameter is constructed, which can comprehensively reflect the strength of wake interference and the speed of dynamic inflow effects. The calculation of the comprehensive switching parameter integrates information from two dimensions: its value is positively correlated with the wake interference intensity index (the stronger the wake interference, the larger the comprehensive switching parameter); and negatively correlated with the dynamic inflow time scale (the more significant the dynamic inflow effect, the larger the comprehensive switching parameter). By comparing the comprehensive switching parameter with preset first and second thresholds, the operating condition range can be divided into three regions: a region with low wake interference and slow dynamic changes, a region with strong wake interference and fast dynamic changes, and a transitional region between the two. The first and second thresholds are determined based on extensive simulation calculations and experimental calibrations. Their values ​​should reasonably distinguish the applicable boundaries of the blade element momentum theory method and the generalized dynamic wake theory method under different operating conditions, ensuring that the blade element momentum theory method is selected within its applicable operating condition range, and the generalized dynamic wake theory method is selected within its necessary operating condition range. This step, through a dual-threshold comparison mechanism that integrates switching parameters, achieves automated and quantitative discrimination of complex operating conditions, avoiding the limitations of traditional methods that rely on manual experience to set a single fixed threshold.

[0056] For S3, when the integrated switching parameter is less than the first threshold or greater than the second threshold, either the blade element momentum theory method or the generalized dynamic wake theory method is selected for aerodynamic solution. When the integrated switching parameter is less than the first threshold, it indicates that the current operating condition is a steady-state condition with low wake interference and slow dynamic changes. At this time, the inflow conditions are relatively uniform and steady, and the quasi-static assumption of the blade element momentum theory method is basically valid. The calculation accuracy of this method can meet engineering requirements. The blade element momentum theory method divides the blade along the spanwise direction into several blade elements, establishes momentum balance equations and blade element force balance equations for each blade element, iteratively solves the axial induction factor and tangential induction factor, and then calculates the aerodynamic forces and moments on each blade element. This method has a simple calculation model, high computational efficiency, and good engineering applicability under steady-state conditions. When the integrated switching parameter is greater than the second threshold, it indicates that the current operating condition is a complex condition with strong wake interference and rapid dynamic changes. At this time, the inflow conditions exhibit significant non-uniformity and unsteadiness, and the basic assumptions of the blade element momentum theory method are severely violated. The generalized dynamic wake theory method, based on unsteady potential flow theory, describes rotor wake dynamics through modal expansion of the pressure potential function. It naturally captures the dynamic evolution of the wake, wake distortion effects, and dynamic inflow effects. In this method, the pressure potential function is expressed as a series expansion of time-varying modal coefficients and associated Legendre functions. The modal coefficients satisfy dynamic equations including the mass matrix, damping matrix, and stiffness matrix. The aerodynamic excitation vector is obtained by projecting the blade aerodynamic load onto the modal space. This step enables independent application of the two methods within different advantageous ranges, allowing them to achieve optimal performance under their respective suitable operating conditions.

[0057] For S4, in actual operation, the operating parameters change continuously, with continuous transition intervals between low-wake interference and strong-wake interference, and between slow-dynamic and fast-dynamic changes. If a hard switching method is used in these transition intervals—that is, completely different calculation methods are applied to both sides of the critical point—the differences in mathematical models, discretization methods, and approximation assumptions between the two methods will lead to sudden jumps in the calculation results, severely affecting the continuity and numerical stability of aerodynamic load calculations. Therefore, a weighted hybrid solution scheme combining blade element momentum theory and generalized dynamic wake theory is adopted in the transition region. The core idea of ​​the weighted hybrid solution is to calculate using both methods simultaneously, and then weight the results according to certain weighting coefficients, which smoothly change with the overall switching parameters. At the left boundary of the transition zone (near the first threshold), the weight of the blade element momentum theory method approaches 1, while the weight of the generalized dynamic wake theory method approaches 0, and the calculation results are dominated by the blade element momentum theory method. At the right boundary of the transition zone (near the second threshold), the weight of the blade element momentum theory method approaches 0, while the weight of the generalized dynamic wake theory method approaches 1, and the calculation results are dominated by the generalized dynamic wake theory method. Inside the transition zone, the weights of the two methods change continuously, achieving a smooth transition in the calculation results. This step, through a weighted hybrid solution mechanism, eliminates the load jump phenomenon during method switching, ensuring the temporal continuity and numerical stability of aerodynamic load calculations.

[0058] This invention presents a hybrid aerodynamic solution method based on wake interference and dynamic inflow time. It organically links steps such as acquiring operating parameters and calculating indices, calculating integrated switching parameters and comparing thresholds, selecting methods and executing switching, and weighted mixing in the transition zone, forming a complete and adaptive hybrid aerodynamic solution. Specifically, the calculation of wake interference intensity indices and dynamic inflow timescales provides quantitative basis for operating condition judgment; the design of integrated switching parameters enables the fusion evaluation of the two physical effects of wake interference and dynamic inflow; the dual threshold comparison mechanism automatically divides the operating condition into three regions and matches corresponding solution strategies; and the weighted mixing solution mechanism ensures a smooth and continuous method switching process. The progressive and collaborative steps among these steps achieve an optimal balance between the computational efficiency of the blade element momentum theory method and the computational accuracy of the generalized dynamic wake theory method.

[0059] Example 2: This invention provides a hybrid aerodynamic solution method based on wake interference and dynamic inflow time, including: Mathematical Models of BEM and GDW Methods The BEM model used in this invention introduces the following improvements based on the traditional leaf element momentum theory: Dynamic inflow correction: The Pitt-Peters dynamic inflow model is introduced, and the quasi-steady-state induction factor is replaced with a dynamic induction factor. :

[0060]

[0061] in, For uniform inflow time constant, The linear inflow time constant is and These are the quasi-steady-state uniform and linear induction factor components, respectively.

[0062] The complete mathematical model of the GDW method: The potential function of GDW expands to:

[0063] The dynamic equations for the modal coefficients are:

[0064] The matrix elements are:

[0065]

[0066]

[0067] Aerodynamic excitation vector Obtained from the projection of aerodynamic loads on the blades:

[0068] The relationship between the axial induced velocity and the gradient of the potential function is as follows:

[0069] 4.2 Adaptive Switching Criterion Based on Wake Interference Intensity and Dynamic Inflow Time Scale One of the core innovations of this invention is the establishment of an adaptive handover criterion based on two parameters. The comprehensive handover parameters are defined. :

[0070] in, and Let be the weighting coefficient, satisfying , The indicator of wake interference intensity. For reference wake interference intensity, For dynamic inflow timescale, For reference, the dynamic inflow timescale.

[0071] Wake Interference Intensity Index Definition:

[0072] in, The velocity loss is caused by the wake. This represents the impeller swept area. For cases with partial wake obstruction, a weighted calculation based on the obstruction area ratio is required:

[0073] in The impeller area affected by the wake.

[0074] Dynamic inflow timescale Definition:

[0075] in This represents the rate of change of wind speed. For the transient yaw condition:

[0076] in The rate of change of yaw angle. This represents the impeller speed.

[0077] Adaptive switching rules: when When the BEM method is used to solve the problem (low wake interference, slow dynamic conditions), it is necessary to consider the following: when When the strong wake interference and fast dynamic conditions are present, the GDW method is used to solve the problem. when When the time is right, a weighted mixed method is used to solve the problem:

[0078] The weight function Use a Sigmoid-type smooth transition function:

[0079] in As a transition center, For transition bandwidth. This function ensures that... and Smooth transitions in the vicinity to avoid load jumps during method switching.

[0080] Weighting coefficient and Adaptively adjust based on current operating conditions:

[0081] When wake interference dominates Increase when the dynamic inflow effect dominates Increase.

[0082] 4.3 State Variable Transfer and Continuity Guarantee Mechanism of Hybrid Solvers BEM and GDW use different state variable representation methods. When switching methods, a mapping relationship needs to be established to ensure the continuity of aerodynamic load calculation.

[0083] BEM → GDW State Mapping (Forward Mapping) ): Axial inducing factor distribution of BEM Modal coefficients mapped to GDW via Legendre polynomial expansion:

[0084] Written in matrix form:

[0085] Where the mapping matrix The elements are:

[0086] This mapping is essentially a Galerkin projection, which projects the radially discrete induced distribution of the BEM onto the modal space of the GDW.

[0087] GDW → BEM State Mapping (Reverse Mapping) ): The potential function mode coefficients of GDW are reconstructed to obtain the induction factor distribution of BEM:

[0088] Written in matrix form:

[0089] The inverse mapping matrix is ​​the generalized inverse of the forward mapping:

[0090] Continuity guarantee mechanism: To ensure load continuity during method switching, a relaxation smoothing factor is introduced. :

[0091] in This is the state vector after the switch. The mapped state vector, This is the state vector from the previous time step. Relaxation factor. The transition time steps gradually increase from 0 to 1 over several time steps before and after the switch. Determined by the following formula:

[0092] in The transition time is typically 1-2 impeller rotation cycles.

[0093] Furthermore, to eliminate high-frequency oscillations that may be introduced during the mapping process, the induced factor after mapping is low-pass filtered:

[0094] in For filtering half-width, it is usually taken as 1-2 leaf element spacing.

[0095] 4.4 Machine Learning-Based Cross-Validation Algorithm To further improve the accuracy of the hybrid solution, this invention introduces a machine learning-based cross-validation algorithm to perform residual correction on the calculation results of BEM and GDW.

[0096] Feature vector construction: For each time step, construct a feature vector. :

[0097] in The average induction factor, For the gradient of inducing factors, The tip speed ratio.

[0098] Residual Correction Model: Training a Gaussian Process Regression (GPR) model to predict the residuals of BEM and GDW:

[0099]

[0100] The predicted value of the GPR model is:

[0101] in For the feature matrix of the training set, For kernel function vectors, For the kernel matrix, For noise variance. A radial basis function (RBF) kernel is used:

[0102] in For signal variance, This is the length scale parameter.

[0103] Cross-validation strategy: K-fold cross-validation (K=5) is used to train and validate the residual correction model. The training dataset is randomly divided into K subsets of similar size. Each subset is used as the validation set, and the remaining K-1 subsets are used as the training set. Calculate the mean absolute error (MAE) and coefficient of determination (R²) for K validations. If the average R² ≥ 0.9, the residual correction model is accepted; otherwise, the eigenvectors or kernel function parameters are adjusted. Corrected hybrid solution results:

[0104]

[0105] Online update mechanism: During actual calculations, when new high-precision reference data (such as CFD calculation results or experimental measurement data) becomes available, it is added to the training set and the GPR model is retrained, achieving online updates to the residual correction model. Online updates employ an incremental learning strategy to avoid retraining the entire model.

[0106] in Added for Schur.

[0107] The purpose of this invention is to overcome the above-mentioned shortcomings of the prior art and propose an adaptive hybrid aerodynamic solution method based on blade element momentum theory and generalized dynamic wake. By establishing an adaptive switching criterion based on wake interference intensity and dynamic inflow time scale, designing a state variable transfer and continuity guarantee mechanism for the hybrid solver, and introducing a cross-validation algorithm based on machine learning, the optimal balance between the computational efficiency of the BEM method and the computational accuracy of the GDW method is achieved. Under complex working conditions, the aerodynamic load calculation accuracy is significantly improved compared with the single BEM method.

[0108] Figure 1 BEM / GDW Adaptive Hybrid Solution Flowchart. This diagram illustrates the complete solution process of this method: after inputting wind conditions and blade geometry parameters, aerodynamic state initialization is first performed, followed by calculation of switching criterion parameters based on wake interference intensity and dynamic inflow timescale. ;according to The solution is obtained by selecting the BEM branch, GDW branch, or weighted hybrid solution of the transition region. After the solution is passed through the state variable transfer and continuity guarantee module, it enters the machine learning cross-validation algorithm for residual correction. Finally, the aerodynamic loads (thrust, torque, power) are output, and the aerodynamic state is updated through the feedback loop for iterative calculation until convergence.

[0109] Figure 2 BEM / GDW adaptive switching criterion phase diagram. This diagram uses wake interference intensity as the basis for the criterion. The horizontal axis represents the dynamic inflow time scale. The vertical axis represents the three regions: the BEM region, the GDW region, and the transition region. When integrating switching parameters... The time is the BEM region (low wake interference + slow dynamics). It is in the GDW region (strong wake interference + fast dynamics). The transition zone uses a weighted mixing method. The locations of typical operating points are marked on the figure.

[0110] Figure 3 A schematic diagram of the BEM-GDW state variable transfer mechanism. This diagram illustrates the bidirectional mapping relationship between the BEM state vector and the GDW state vector. BEM→GDW direction: Axial induction factor. BEM state variables are mapped through the Galerkin projection matrix potential function modal coefficients projected onto GDW GDW→BEM direction: Modal coefficients through inverse mapping matrix Reconstructed as the radial induction factor distribution of BEM.

[0111] Figure 4 Comparison chart of aerodynamic load accuracy using BEM / GDW / hybrid methods. This chart contains two sub-charts: (a) Power coefficient (a) Curve of variation with azimuth angle; (b) Thrust coefficient Curves showing variation with azimuth angle. The calculation results of the single BEM method, the single GDW method, the BEM-GDW hybrid method, and the reference solution (CFD / experiment) were compared, demonstrating the accuracy advantage of the hybrid method in achieving a high degree of agreement with the reference solution under dynamic inflow conditions.

[0112] Example 3: BEM-GDW hybrid solution for NREL 5MW benchmark wind turbine under yaw / dynamic inflow conditions Using the NREL 5MW benchmark wind turbine as an example, the effectiveness of the proposed BEM / GDW adaptive hybrid aerodynamic solution method is verified.

[0113] Wind turbine parameters: impeller diameter m, rated wind speed m / s, rated speed rpm, number of blades Wheel hub height m.

[0114] Operating conditions settings: Operating Condition 1 – Steady-state uniform flow: m / s, yaw angle There is no wake obstruction. The wake interference intensity at this time... Dynamic inflow timescale Switch parameters The method chosen was BEM.

[0115] Operating Condition 2 – Yaw Dynamic Inflow: m / s, yaw angle from within 2 seconds Change to .at this time , The method chosen is GDW.

[0116] Operating Condition 3 – Partial Wake Obstruction: At m / s, there is a similar unit at 7D upstream, with a wake obstruction area of ​​approximately 40%. , In the transition zone, the method selection is weighted mixing ( ).

[0117] Calculation results:

[0118] The calculation results show that: (1) Under steady-state conditions, the hybrid method automatically selects BEM, and its computational efficiency is comparable to that of pure BEM; (2) Under yaw dynamic inflow conditions, the hybrid method... The error is only 3.1%, which is about 83% higher than the 18.2% of the BEM method, and the calculation time is only 44% of the pure GDW method; (3) Under the condition of partial wake obstruction, the accuracy of the hybrid method is better than BEM and GDW, and the calculation efficiency is significantly better than the pure GDW method.

[0119] Residual correction model training: 300 sets of CFD calculation results were used as training data. The feature vector contains 10 parameters. After 5-fold cross-validation, the average R² of the GPR model was 0.94. The average accuracy of the hybrid method after correction was further improved by about 2 percentage points.

[0120] In another embodiment of the present invention, a hybrid aerodynamic solution system based on wake interference and dynamic inflow time is provided, which can be used to implement the above-mentioned hybrid aerodynamic solution method based on wake interference and dynamic inflow time. Specifically, the hybrid aerodynamic solution system based on wake interference and dynamic inflow time includes: a data acquisition module, a switching parameter calculation module, a selection module, and a mixing module.

[0121] The data acquisition module is used to acquire the operating condition parameters of the wind turbine, and based on the operating condition parameters, calculate the wake interference intensity index and the dynamic inflow time scale.

[0122] The switching parameter calculation module is used to calculate the comprehensive switching parameters based on the wake interference intensity index and the dynamic inflow time scale, and compare the comprehensive switching parameters with the preset first threshold and second threshold, wherein the first threshold is less than the second threshold.

[0123] The selection module is used to select the blade element momentum theory method for aerodynamic solution when the integrated switching parameters are less than the first threshold, and to select the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are greater than the second threshold.

[0124] The hybrid module is used to select a weighted mixture of the blade element momentum theory method and the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are between the first threshold and the second threshold.

[0125] All relevant content of each step involved in the aforementioned embodiments of the hybrid aerodynamic solution method based on wake interference and dynamic inflow time can be referenced to the functional description of the corresponding functional module of the hybrid aerodynamic solution system based on wake interference and dynamic inflow time in the embodiments of the present invention, and will not be repeated here.

[0126] The module division in this embodiment of the invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in the various embodiments of the invention can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0127] In another embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions from the computer storage medium to achieve a corresponding method flow or function. The processor described in this embodiment can be used for operating a hybrid aerodynamic solution method based on wake interference and dynamic inflow time.

[0128] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the hybrid aerodynamic solution method based on wake interference and dynamic inflow time in the above embodiments.

[0129] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0130] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0131] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0132] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A hybrid aerodynamic solution method based on wake interference and dynamic inflow time, characterized in that, Includes the following steps: Obtain the operating parameters of the wind turbine, and calculate the wake interference intensity index and dynamic inflow time scale based on the operating parameters; Based on the wake interference intensity index and the dynamic inflow time scale, the comprehensive switching parameters are calculated, and the comprehensive switching parameters are compared with a preset first threshold and a second threshold, wherein the first threshold is less than the second threshold. When the overall switching parameters are less than the first threshold, the blade element momentum theory method is selected for aerodynamic solution; when the overall switching parameters are greater than the second threshold, the generalized dynamic wake theory method is selected for aerodynamic solution. When the integrated switching parameters are between the first and second thresholds, a weighted hybrid of the leaf element momentum theory method and the generalized dynamic wake theory method is selected for aerodynamic solution.

2. The hybrid aerodynamic solution method based on wake interference and dynamic inflow time according to claim 1, characterized in that, The process of acquiring the operating parameters of the wind turbine and calculating the wake interference intensity index based on these parameters includes: The wake interference intensity index is calculated as follows: in, The velocity loss is caused by the wake. The impeller swept area is used; for cases with partial wake obstruction, a weighted calculation is performed based on the obstruction area ratio. in The impeller area affected by the wake.

3. The hybrid aerodynamic solution method based on wake interference and dynamic inflow time according to claim 1, characterized in that, The acquisition of wind turbine operating condition parameters and the calculation of dynamic inflow time scale based on these parameters include: Dynamic inflow timescale Definition: in For wind speed change rate; for yaw transient conditions: in The rate of change of yaw angle. This represents the impeller speed.

4. The hybrid aerodynamic solution method based on wake interference and dynamic inflow time according to claim 1, characterized in that, The calculation of integrated switching parameters based on wake interference intensity index and dynamic inflow time scale includes: Comprehensive switching parameters : in, and Let be the weighting coefficient, satisfying , The indicator of wake interference intensity. For reference wake interference intensity, For dynamic inflow timescale, For reference, the dynamic inflow timescale.

5. The hybrid aerodynamic solution method based on wake interference and dynamic inflow time according to claim 1, characterized in that, When the integrated switching parameters are between the first and second thresholds, a weighted hybrid approach combining the blade element momentum theory method and the generalized dynamic wake theory method is selected for aerodynamic solution, including: Solving using a weighted mixture method: The weight function Use a Sigmoid-type smooth transition function: in As a transition center, For transition bandwidth; this function ensures that during... and Smooth transition nearby; Weighting coefficient and Adaptively adjust based on current operating conditions: When wake interference dominates Increase when the dynamic inflow effect dominates Increase.

6. The hybrid aerodynamic solution method based on wake interference and dynamic inflow time according to claim 1, characterized in that, When switching from the leaf element momentum theory method to the generalized dynamic wake theory method, a positive mapping of state variables is performed; Axial inducible factor distribution of leaf element momentum theory BEM The modal coefficients are mapped to the generalized dynamic wake theory (GDW) via Legendre polynomial expansion: Written in matrix form: Where the mapping matrix The elements are: This mapping is essentially a Galerkin projection, which projects the radially discrete induced distribution of BEM onto the modal space of GDW; When switching from the generalized dynamic wake theory method to the leaf element momentum theory method, the inverse mapping of state variables is performed; The potential function modal coefficients of the generalized dynamic wake theory (GDW) are reconstructed to obtain the induction factor distribution of the leaf element momentum theory (BEM): Written in matrix form: The inverse mapping matrix is ​​the generalized inverse of the forward mapping: Introducing a relaxation smoothing factor : in This is the state vector after the switch. The mapped state vector, The state vector of the previous time step; relaxation factor The transition time steps gradually increase from 0 to 1 over several time steps before and after the switch. Determined by the following formula: in This refers to the transition time length.

7. The hybrid aerodynamic solution method based on wake interference and dynamic inflow time according to claim 6, characterized in that, A machine learning-based cross-validation algorithm was introduced to perform residual correction on the calculation results of BEM and GDW: For each time step, construct a feature vector. : in The average induction factor, For the gradient of inducing factors, The tip speed ratio; Training a Gaussian process regression (GPR) model to predict the residuals of BEM and GDW: The predicted value of the GPR model is: in For the feature matrix of the training set, For kernel function vectors, For the kernel matrix, For noise variance; using the radial basis function (RBF) kernel: in For signal variance, For length scale parameters; K-fold cross-validation was used to train and validate the residual correction model. The training dataset is randomly divided into K subsets of similar size. Each subset is used as the validation set, and the remaining K-1 subsets are used as the training set. Calculate the mean absolute error (MAE) and coefficient of determination (R²) for K validations. If the average R² ≥ 0.9, the residual correction model is accepted; otherwise, the eigenvectors or kernel function parameters are adjusted. Corrected hybrid solution results: In actual calculations, when new high-precision reference data becomes available, it is added to the training set and the GPR model is retrained to achieve online updates of the residual correction model.

8. A hybrid aerodynamic solution system based on wake interference and dynamic inflow time, characterized in that, include: The data acquisition module is used to acquire the operating condition parameters of the wind turbine, and calculate the wake interference intensity index and dynamic inflow time scale based on the operating condition parameters. The switching parameter calculation module is used to calculate the comprehensive switching parameters based on the wake interference intensity index and the dynamic inflow time scale, and compare the comprehensive switching parameters with a preset first threshold and a second threshold, wherein the first threshold is less than the second threshold. The selection module is used to select the blade element momentum theory method for aerodynamic solution when the integrated switching parameters are less than the first threshold, and to select the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are greater than the second threshold. The hybrid module is used to select a weighted hybrid of the blade element momentum theory method and the generalized dynamic wake theory method for aerodynamic solution when the integrated switching parameters are between the first threshold and the second threshold.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the hybrid aerodynamic solution method based on wake interference and dynamic inflow time as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the hybrid aerodynamic solution method based on wake interference and dynamic inflow time as described in any one of claims 1 to 7.

Images