Wind power blade model parameter correction method, device, equipment and medium

Abstract

The invention provides a wind power blade model parameter correction method, device and equipment and a medium, and belongs to the technical field of wind power generation. The method comprises the steps that wind power blade simulation model parameters are obtained, and fixed parameters and to-be-corrected parameters are distinguished; simulation is carried out according to the fixed parameters and the to-be-corrected parameters to obtain evaluation index results, and errors are calculated by comparing the evaluation index results with preset test results of the evaluation indexes. And if the error exceeds the limit, adjusting the to-be-corrected parameters of the wind power blade simulation model and performing simulation again until the error reaches the standard, and finally outputting the adjusted to-be-corrected parameters of the wind power blade simulation model. According to the method, through automatic parameter adjustment, iterative optimization based on an error threshold value, an advanced optimization algorithm is adopted, and the generalization ability of the model is improved, so that the data processing efficiency and the simulation prediction precision are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a method for correcting wind turbine blade model parameters, a device for correcting wind turbine blade model parameters, an electronic device and a computer-readable storage medium. Background Art

[0002] With the transformation of the global energy structure, wind energy has become an important part of the global energy structure due to its clean and renewable characteristics. As a key device for converting wind energy into electrical energy, the blade length of wind turbines has exceeded 100 meters to capture more wind energy and improve power generation efficiency. However, the increase in blade length and the flexibility of the structure have led to increasingly significant nonlinear effects, which has brought great challenges to the simulation calculation of blades. Traditional simulation models cannot improve the accuracy of wind turbine blade simulation calculations when dealing with these long and flexible blades. Therefore, improving the accuracy of wind turbine blade simulation calculations has become a problem that must be solved at present.

[0003] At present, by determining a set of correction parameters as optimization variables, and defining the difference between the theoretical model and the actual model as our optimization target. In the iterative calculation process, we use the optimization algorithm to adjust these parameters to minimize the model error. In this process, the artificial neural network algorithm plays a vital role. It imitates the behavior of animal neural networks and can effectively deal with nonlinear mapping problems, which is crucial for identifying and screening key parameters that affect the output of the model. The predictive ability of artificial neural networks also provides guidance for the optimization algorithm, making parameter adjustment more accurate. In addition, the adaptive correction characteristics of artificial neural networks help to continuously improve the generalization ability and prediction accuracy of the model during the iteration process. In summary, through this method of combining optimization algorithms and artificial neural networks, we can not only effectively correct the model parameters, but also ensure that the model has better adaptability and prediction performance on new data sets.

[0004] However, the existing wind turbine blade model parameter correction methods still have some problems: (1) The parameter screening and optimization process may be limited by the quality and quantity of data, resulting in uncertainty in the correction results; (2) The training process of artificial neural networks may require a lot of computing resources and is prone to falling into local optimal solutions; (3) The generalization ability of the model is limited by the coverage of the training data set. For unseen working conditions, the model's predictive ability may decrease.

[0005] In general, the existing wind turbine blade model parameter correction methods are subject to data limitations and computing resource challenges, which lead to problems of parameter optimization uncertainty and limited generalization ability. Summary of the invention

[0006] The purpose of the embodiments of the present invention is to provide a method, device, equipment and medium for correcting wind turbine blade model parameters, so as to solve the problem that the existing model parameter correction methods are subject to data limitations and computing resource challenges, thereby leading to parameter optimization uncertainty and limited generalization capabilities.

[0007] In order to achieve the above object, an embodiment of the present invention provides a method for correcting wind turbine blade model parameters, comprising: Obtaining parameters of a wind turbine blade simulation model; wherein the parameters include parameters to be corrected and fixed parameters; Determine the simulation results of the evaluation index of the wind turbine blade simulation model according to the parameters of the wind turbine blade simulation model; Determining the error of the wind turbine blade simulation model according to the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index; If the error of the wind turbine blade simulation model is greater than a preset threshold, the parameters to be corrected of the wind turbine blade simulation model are adjusted using a model parameter adjustment method, and a simulation result of the evaluation index of the wind turbine blade simulation model is determined by performing simulation analysis based on the parameters of the wind turbine blade simulation model; If the error of the wind turbine blade simulation model is less than a preset threshold, the adjusted parameters of the wind turbine blade simulation model are output.

[0008] Optionally, the parameters to be corrected of the wind turbine blade model include: blade distributed mass, flapping direction stiffness, swing direction stiffness and torsional stiffness; The evaluation indicators of the wind turbine blade simulation model include: blade mass, swing frequency, flapping frequency and torsion frequency.

[0009] Optionally, determining the error of the wind turbine blade simulation model according to the simulation result of the evaluation index of the wind turbine blade simulation model and the preset test result of the evaluation index includes: The following formula is used to calculate the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index to obtain the error of the wind turbine blade simulation model; ;in, represents the error of the wind turbine blade simulation model, The simulation results representing the evaluation index of the wind turbine blade simulation model, Indicates the preset test results of the evaluation indicators.

[0010] Optionally, the parameters to be corrected of the wind turbine blade simulation model are adjusted using a model parameter adjustment method, including: Determine a plurality of wind turbine blade simulation model design schemes; wherein each wind turbine blade simulation model design scheme includes different parameters to be corrected of the wind turbine blade simulation model; Determining simulation results of evaluation indicators of multiple wind turbine blade simulation model design schemes based on multiple wind turbine blade simulation model design schemes and fixed parameters of the wind turbine blade simulation models; The pre-built convolutional neural network-long short-term memory neural network parallel recurrent neural network is trained using the simulation results of multiple wind turbine blade simulation model design schemes and corresponding evaluation indicators, and a wind turbine blade parameter correction model is obtained after the training is completed; Input the test results of the preset evaluation indexes into the wind turbine blade parameter correction model to obtain the target design scheme of the wind turbine blade simulation model; The parameters to be corrected of the wind turbine blade simulation model are updated to a target design scheme of the wind turbine blade simulation model.

[0011] Optionally, multiple wind turbine blade simulation model design schemes are determined, including: Determining value ranges of multiple parameters to be corrected in the parameters of the wind turbine blade model; Selecting multiple values ​​of the parameters to be corrected from the value range of each parameter to be corrected; By utilizing a uniform design scheme, multiple values ​​of each parameter to be corrected are combined to obtain multiple wind turbine blade simulation model design schemes.

[0012] Optionally, the convolutional neural network includes: using the Conv1D layer in the layers module of the Keras library to construct a one-dimensional convolutional neural network; the long short-term memory neural network includes: using the BidirectionalLSTM layer in the layers module of the Keras library to construct a bidirectional long short-term memory neural network model; the recurrent neural network includes: using the SimpleRNN layer in the layers module of the Keras library.

[0013] In a second aspect of the implementation manner of the present invention, the embodiment of the present invention further provides a wind turbine blade model parameter correction device, comprising: A data acquisition module is used to acquire parameters of a wind turbine blade simulation model; wherein the parameters include parameters to be corrected and fixed parameters; A simulation analysis module, used to determine the simulation results of the evaluation index of the wind turbine blade simulation model according to the parameters of the wind turbine blade simulation model; An error analysis module, used to determine the error of the wind turbine blade simulation model based on the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index; A simulation cycle module is used to adjust the parameters to be corrected of the wind turbine blade simulation model by using a model parameter adjustment method when the error of the wind turbine blade simulation model is greater than a preset threshold, and return to call the simulation analysis module; The parameter output module is used to output the adjusted parameters of the wind turbine blade simulation model when the error of the wind turbine blade simulation model is less than a preset threshold.

[0014] Optionally, the error analysis module is specifically used for: The following formula is used to calculate the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index to obtain the error of the wind turbine blade simulation model; ;in, represents the error of the wind turbine blade simulation model, The simulation results representing the evaluation index of the wind turbine blade simulation model, Indicates the preset test results of the evaluation indicators.

[0015] In a third aspect of the embodiments of the present invention, an electronic device is provided, including: a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and when the machine-readable instructions are executed by the processor, the above-mentioned wind turbine blade model parameter correction method is executed.

[0016] In a fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided, storing computer instructions. When the computer instructions are executed on a computer, the computer executes the above-mentioned wind turbine blade model parameter correction method.

[0017] Beneficial effects of the present invention: (1) Automated parameter adjustment: The existing technology adjusts parameters manually, which is inefficient and the results are easily affected by the operator's subjective influence. The automated parameter adjustment process of this application reduces manual intervention and improves adjustment efficiency and consistency of results.

[0018] (2) Iterative optimization based on error threshold: The existing technology may lack systematicity and clear optimization goals through iterative optimization process. However, this application adopts a preset error threshold to guide the iteration, ensuring that the optimization goal is clear and the optimization process is systematic.

[0019] (3) More advanced optimization algorithms: The optimization algorithms used in the prior art may be relatively simple and have limited ability to handle complex nonlinear problems. However, this application uses advanced algorithms such as artificial neural networks to more effectively solve nonlinear problems.

[0020] (4) Improving the generalization ability of the model: The existing technology may not be able to adapt the model to new data sets, which may affect the prediction accuracy. However, this application enhances the generalization ability of the model through iterative optimization, so that it can better adapt to new data sets.

[0021] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings: Figure 1 It is a flow chart of a method for correcting wind turbine blade model parameters provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the training process of the CNN-LSTM parallel RNN neural network model provided by an embodiment of the present invention; Figure 3 It is a flow chart of a method for correcting wind turbine blade model parameters provided by another embodiment of the present invention; Figure 4 It is a structural schematic diagram of a wind turbine blade model parameter correction device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0025] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0026] Embodiment 1 Please refer to Figure 1 , Figure 1 : is a flow chart of a method for correcting wind turbine blade model parameters provided by an embodiment of the present invention, the method comprising the following steps: S100, obtaining parameters of a wind turbine blade simulation model; wherein the parameters include parameters to be corrected and fixed parameters; In one embodiment, the parameters to be corrected of the wind turbine blade model include: blade distributed mass, flapping direction stiffness, swing direction stiffness and torsional stiffness.

[0027] Blade distributed mass refers to the distribution of mass on the blades of a wind turbine. It involves the weight distribution of the blade from root to tip along the span. The mass distribution of the blade has a significant impact on the dynamic characteristics of the wind turbine. An ideal blade mass distribution can reduce the stress caused by centrifugal force, flapping and shimmying motion, while improving power generation efficiency and stability. The unit is kg / m.

[0028] Flapping stiffness refers to the bending stiffness of the blade in the flapping direction (i.e. perpendicular to the plane of rotation). Flapping motion is the up and down swinging of the blade around its elastic axis. The size of the flapping stiffness determines the ability of the blade to resist flapping vibrations caused by factors such as wind speed changes, tower shadow effect and gravity. Higher flapping stiffness can reduce the flapping amplitude of the blade, thereby reducing structural fatigue. The unit is Nm^2.

[0029] The swing stiffness refers to the bending stiffness of the blade in the swing direction (i.e., the direction in the rotation plane). Swing is the back and forth swing of the blade around its long axis. The size of the swing stiffness affects the ability of the blade to resist swing vibrations caused by wind turbulence, yaw error, etc. Increasing the swing stiffness helps control the swing response of the blade and reduce the dynamic load caused by it. The unit is Nm^2.

[0030] Torsional stiffness refers to the ability of a blade to resist torsional deformation. Torsional stiffness affects the aerodynamic characteristics of the blade, including changes in the angle of attack, which in turn affects the lift and drag of the blade. Low torsional stiffness may cause excessive twisting of the blade under wind load, affecting the efficiency and safe operation of the wind turbine; while high torsional stiffness may reduce the adaptability of the blade to wind speed changes. The unit is Nm^2.

[0031] In one embodiment, the parameters to be corrected of the wind turbine blade model may be determined through literature research or expert experience.

[0032] In one embodiment, the number of parameters to be corrected of the wind turbine blade model should not be greater than five.

[0033] In one embodiment, the fixed parameters of the wind turbine blade simulation model include: blade length, elastic modulus, variable pitch control parameters, and the like.

[0034] In one embodiment, the evaluation indexes of the wind turbine blade simulation model include: blade mass, swing frequency, flapping frequency and torsion frequency.

[0035] Blade mass is the physical mass of a wind turbine or similar rotating machine blade. This mass has a significant impact on the blade's dynamic properties, including its vibration characteristics, load response, and interaction with the wind. Blade mass usually includes the mass of the blade structure itself and any attached mass such as cladding, coatings, or possible weights.

[0036] The flap frequency is the natural frequency at which the blade vibrates perpendicular to the plane of rotation. This is a natural vibration mode of the blade, where each section of the blade oscillates back and forth along its length. The flap frequency is critical to the stability and life of a wind turbine because it affects the blade's response to aerodynamic and centrifugal forces.

[0037] The flapping frequency is the natural frequency at which the blades vibrate in the plane of rotation. This vibration mode is when the blades flap back and forth around their roots, similar to the flapping of a helicopter rotor. The flapping frequency is also critical to the operation of a wind turbine because it determines how the blades respond to changes in the wind and how they maintain rotational balance.

[0038] The torsional frequency is the natural frequency of the blade's torsional vibration about its longitudinal axis. This vibration mode is the twisting motion of the blade along its length. The torsional frequency affects the aerodynamic performance of the blade because it is related to how the blade's angle of attack changes over time, thus affecting the lift and drag characteristics of the blade.

[0039] S200, determining simulation results of evaluation indicators of the wind turbine blade simulation model according to parameters of the wind turbine blade simulation model; Specifically, the simulation process of the wind turbine blade simulation model is as follows: Step 1 (Building a simulation model): Geometric modeling: Use CAD software (such as SolidWorks, CATIA, etc.) to build a three-dimensional model of the blade.

[0040] Meshing: Import the blade model into pre-processing software (such as ICEM CFD, HyperMesh, etc.) for meshing to ensure that the mesh quality meets the simulation requirements.

[0041] Model setup: Set boundary conditions and physical models in simulation software (such as ANSYS Fluent, Abaqus, Blade Element Theory, etc.).

[0042] Step 2 (simulation analysis): Computational Fluid Dynamics (CFD): Use CFD software to simulate aerodynamic performance, solve the Navier-Stokes equations, and obtain flow field distribution, pressure distribution, aerodynamic coefficients, etc.

[0043] Performance indicators: calculation of lift coefficient, drag coefficient, power coefficient, etc.

[0044] Finite Element Analysis (FEA): Use FEA software to perform structural strength and stiffness analysis, and solve the stress, strain, displacement, etc. of the blade under load.

[0045] Dynamic characteristics analysis: Perform modal analysis to determine the natural frequency and vibration mode of the blade to avoid resonance.

[0046] Stability analysis: Analyze the operating stability of blades in complex wind fields.

[0047] Noise prediction: Use acoustic simulation software to predict the noise level when the blade is running.

[0048] Step 3 (Results analysis and evaluation index determination): Data processing: Collect simulation output data and perform necessary post-processing, such as curve fitting, error analysis, etc.

[0049] Evaluation indicators: blade mass, swing frequency, flapping frequency and torsional frequency.

[0050] S300, determining an error of the wind turbine blade simulation model according to a simulation result of an evaluation index of the wind turbine blade simulation model and a preset test result of the evaluation index; In one embodiment, step S300 specifically includes: The following formula is used to calculate the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index to obtain the error of the wind turbine blade simulation model; ;in, represents the error of the wind turbine blade simulation model, The simulation results representing the evaluation index of the wind turbine blade simulation model, Indicates the preset test results of the evaluation indicators.

[0051] S400, if the error of the wind turbine blade simulation model is greater than a preset threshold, the parameters to be corrected of the wind turbine blade simulation model are adjusted using a model parameter adjustment method, and the process returns to step S200; In one embodiment, step S400 specifically includes: S410, determining a plurality of wind turbine blade simulation model design schemes; wherein each wind turbine blade simulation model design scheme includes different parameters to be corrected of the wind turbine blade simulation model; In one embodiment, step S410 specifically includes: S411, determining value ranges of multiple parameters to be corrected in the parameters of the wind turbine blade model; Specifically, according to the properties of the parameters to be corrected of the wind turbine blade model, the upper and lower limits of the parameter variation are determined, thereby determining the value range of the correction parameter.

[0052] S412, selecting multiple values ​​of the parameters to be corrected from the value range of each parameter to be corrected; Specifically, the number of levels of the parameter to be corrected is determined, but only within the value range of the parameter to be corrected.

[0053] S413, using a uniform design scheme, combining the values ​​of multiple parameters to be corrected for each parameter to be corrected, to obtain multiple wind turbine blade simulation model design schemes.

[0054] Specifically, according to the number of parameters to be corrected and the number of levels, a suitable uniform design table is selected. The uniform design table can be found in relevant literature or software. Among them, the factors in the uniform design table are the parameters to be corrected, and the levels are the values ​​of the parameters to be corrected.

[0055] Taking a 6-factor-10-level design table as an example, the uniform design scheme is shown in Table 1 below.

[0056] Table 1 6 Wind turbine blade simulation model design scheme

[0057] Among them, the factors are X1, X2, ..., X6; the levels are 1, 2, ..., 10, the numbers represent the specific parameter values, and the working conditions are named N1, N2, ..., N10.

[0058] S420, determining simulation results of evaluation indicators of multiple wind turbine blade simulation model design schemes according to multiple wind turbine blade simulation model design schemes and fixed parameters of the wind turbine blade simulation model; Regarding step S420, the simulation analysis process of the simulation model can refer to the above-mentioned step S200, which will not be repeated here.

[0059] S430, using the simulation results of multiple wind turbine blade simulation model design schemes and corresponding evaluation indicators to train a pre-built convolutional neural network-long short-term memory neural network parallel recurrent neural network, and obtaining a wind turbine blade parameter correction model after the training is completed; like Figure 2 As shown, Figure 2 This is the training process of the wind turbine blade parameter correction model. For better understanding, the training process is described in detail below: Step 1 (data preparation): Wind turbine blade simulation model design scheme: Collect or design multiple wind turbine blade simulation model schemes, which should cover different blade design parameters, such as blade length, chord length, twist angle, etc.

[0060] Conduct simulation experiments on each wind turbine blade design and record the simulation results.

[0061] Evaluation indicators: Determine the key indicators for evaluating the performance of wind turbine blades, such as efficiency, load, vibration characteristics, etc.

[0062] The simulation results are matched with the evaluation indicators to form a training data set.

[0063] Step 2 (data preprocessing): Clean the simulation data to remove outliers and noise.

[0064] Standardize or normalize the data to make it easier for the neural network to process.

[0065] Divide the dataset, usually into training set, validation set and test set.

[0066] Step 4 (Building a Convolutional Neural Network-Long Short-Term Memory Neural Network (CNN-LSTM) Parallel Recurrent Neural Network): Designing a Convolutional Neural Network (CNN) Purpose: CNN is used to extract spatial features from input data, such as leaf shape, structure, etc.

[0067] Structure: Design multiple convolutional layers and pooling layers to extract features and reduce data dimensions.

[0068] Designing Long Short-Term Memory Neural Networks (LSTM) Purpose: LSTM is used to process time series data to capture long-term dependencies in the data, such as the dynamic response of leaves in time series.

[0069] Architecture: Design one or more LSTM layers to process sequence data.

[0070] Designing a Recurrent Neural Network (RNN) Purpose: RNN is used to process sequence data and can capture short-term temporal dependencies.

[0071] Architecture: Design one or more RNN layers that can be run in parallel with LSTM layers to provide different perspectives on time series processing.

[0072] Parallel structure design CNN, LSTM and RNN are designed as parallel structures, each network processes different data features and then merges the outputs.

[0073] Merging strategies: There are several strategies that can be used to merge the outputs of different networks, such as concatenation, sum, or average.

[0074] Fully connected layer and output layer After feature extraction, one or more fully connected layers are added to further process and integrate the features.

[0075] The output layer design should match the task, for example, if the task is parameter modification, the output layer should output the modified parameters.

[0076] Loss Function and Optimizer Choose an appropriate loss function, such as mean squared error (MSE) or a custom loss function, to measure the difference between the predicted parameters and the actual parameters.

[0077] Choose an optimizer, such as Adam or SGD, to adjust the network weights during training.

[0078] Step 5 (Training the Neural Network): Set a preset number of training times, such as 100 epochs.

[0079] During the training process, the validation set is used to tune the model to prevent overfitting.

[0080] Input sample data, and the data enters the two paths of training CNN-LSTM and RNN respectively. In the path of training CNN-LSTM, the sample data is first convolved and pooled, and then the meaningless sample data is cleaned and eliminated. The cleaned data enters the LSTM layer, and the network layer of the LSTM layer is trained so that the network layer predicts the weights and offsets in the mapping relationship. In the path of training RNN, recurrent neurons are used to calculate and analyze the data to obtain appropriate weights and offsets, and then the training results are screened in the fully connected layer to establish the optimal neural network model.

[0081] Step 6 (Model Evaluation): After each epoch, the validation set is used to evaluate the model performance.

[0082] Record evaluation indicators, such as loss function value, accuracy, etc.

[0083] Step 7 (model training completed): When the training times reach the preset times, the training stops.

[0084] The model is finally evaluated using the test set to ensure that the model has good generalization ability.

[0085] In one embodiment, the convolutional neural network includes: constructing a one-dimensional convolutional neural network using a Conv1D layer in a layers module of a Keras library.

[0086] In one embodiment, the long short-term memory neural network includes: using the Bidirectional LSTM layer in the layers module of the Keras library to build a long short-term memory neural network model with bidirectionality.

[0087] In one embodiment, the recurrent neural network includes: a SimpleRNN layer using the layers module of the Keras library.

[0088] S440, inputting the test results of the preset evaluation index into the wind turbine blade parameter correction model to obtain a target design scheme of the wind turbine blade simulation model.

[0089] Specifically, a specific implementation is given below: Step 1 (data preparation): It is necessary to prepare a data set containing wind turbine blade design parameters and corresponding evaluation index test results. This data set should include various blade design parameters (such as length, width, angle, etc.) and the performance of these designs in simulation or actual testing (such as efficiency, load, noise, etc.).

[0090] Step 2 (Building the Autoencoder): Build an autoencoder model that is able to learn low-dimensional representations of wind turbine blade design parameters in a dataset.

[0091] Encoder: Maps the high-dimensional design parameters into a low-dimensional space. This low-dimensional representation captures the main features of the design parameters and may better reflect the key factors of blade performance than the original parameters.

[0092] Decoder: Remaps the low-dimensional representation back to the original design parameter space. This ensures that the autoencoder is able to reconstruct the original data through the compressed representation.

[0093] Step 3 (Training the Autoencoder): The autoencoder is trained using the dataset and the weights of the encoder and decoder are optimized by minimizing the reconstruction error. This process allows the autoencoder to learn how to effectively represent the designed parameters.

[0094] Step 4 (feature extraction): Once the autoencoder is trained, use the encoder part to extract low-dimensional features for each sample in the dataset.

[0095] Step 5 (Evaluation Index Input): The test results of the preset evaluation indicators are used as input and combined with the low-dimensional features extracted by the autoencoder. This can be achieved in the following ways: Supervised learning: Low-dimensional features are used as input and evaluation metrics are used as labels to train a supervised learning model (such as a regression model or classifier) ​​to predict or classify the performance of the blade design.

[0096] Parameter optimization: Using low-dimensional features and evaluation metrics as guidance, we search for the optimal design parameters in low-dimensional space through optimization algorithms such as gradient descent or genetic algorithms.

[0097] Step 6 (Generate target design solution): Based on the optimized parameters obtained in the above steps, the decoder part of the autoencoder is used to remap these low-dimensional features back to the original design parameter space to obtain an optimized wind turbine blade design solution.

[0098] S450: updating the parameters to be corrected of the wind turbine blade simulation model to a target design solution of the wind turbine blade simulation model.

[0099] S500, if the error of the wind turbine blade simulation model is less than a preset threshold, outputting the adjusted parameters of the wind turbine blade simulation model; Specifically, in the iterative cycle from step S200 to step S400, when the error of the wind turbine blade simulation model is less than a preset threshold, the adjusted parameters of the wind turbine blade simulation model are output.

[0100] In order to better understand the present invention, the following reference Figure 3 A specific implementation method is given: Step 1: Simulate and analyze the wind turbine blade model when the model parameters use theoretical values, obtain the simulation results of the evaluation indicators, and compare them with the test results. If the error meets the requirements, it proves that the model does not need to be corrected. If the error does not meet the requirements, the model needs to be corrected to improve its accuracy.

[0101] Step 2: Analyze the main parameters that affect the simulation results. Determine the wind turbine blade model parameters and evaluation indicators to be corrected through literature research or experience. (The more parameters corrected, the better the effect, but the higher the calculation cost.) The main parameters include blade distributed mass, flapping direction stiffness, swing direction stiffness, torsional stiffness, etc.

[0102] Step 3: According to the nature of the parameter, determine the upper and lower limits of the parameter change and the number of change levels, and formulate a factor (parameter)-level design table. (The more levels, the more precise the results, but the higher the calculation cost) Step 4: Based on the factor-level design table, a uniform design scheme is formulated. Then, according to the uniform design scheme, the settings of the wind turbine blade model parameters under various working conditions are adjusted, and simulation analysis is performed to obtain the evaluation index results of various working conditions.

[0103] Step 6: Establish a neural network learning sample library based on the parameters and results of each working condition.

[0104] Step 7: Conduct neural network learning based on the CNN-LSTM parallel RNN algorithm, and use this algorithm to establish a nonlinear mapping relationship model between evaluation indicators and model parameters.

[0105] Step 8: Input the test results as the target values ​​of the evaluation indicators into the nonlinear mapping relationship model, and calculate the wind turbine blade model parameter correction values ​​through the nonlinear mapping relationship model.

[0106] Step 9: Substitute the corrected parameters into the wind turbine blade model and simulate again. If the error between the simulation and test results meets the requirements, the model correction is terminated. If not, adjust the upper and lower limits of the parameters and the number of change levels, and then perform simulation analysis, build a larger neural network sample library, and perform parameter correction.

[0107] Beneficial effects of the present invention: (1) Automated parameter adjustment: The existing technology adjusts parameters manually, which is inefficient and the results are easily affected by the operator's subjective influence. The automated parameter adjustment process of this application reduces manual intervention and improves adjustment efficiency and consistency of results.

[0108] (2) Iterative optimization based on error threshold: The existing technology may lack systematicity and clear optimization goals through iterative optimization process. However, this application adopts a preset error threshold to guide the iteration, ensuring that the optimization goal is clear and the optimization process is systematic.

[0109] (3) More advanced optimization algorithms: The optimization algorithms used in the prior art may be relatively simple and have limited ability to handle complex nonlinear problems. However, this application uses advanced algorithms such as artificial neural networks to more effectively solve nonlinear problems.

[0110] (4) Improving the generalization ability of the model: The existing technology may not be able to adapt the model to new data sets, which may affect the prediction accuracy. However, this application enhances the generalization ability of the model through iterative optimization, so that it can better adapt to new data sets.

[0111] Embodiment 2 Based on the same inventive concept, Figure 4 As shown, the embodiment of the present invention further provides a wind turbine blade model parameter correction device 200, comprising: The data acquisition module 210 is used to acquire the parameters of the wind turbine blade simulation model; wherein the parameters include parameters to be corrected and fixed parameters; The simulation analysis module 220 is used to determine the simulation results of the evaluation index of the wind turbine blade simulation model according to the parameters of the wind turbine blade simulation model; The error analysis module 230 is used to determine the error of the wind turbine blade simulation model according to the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index; The simulation cycle module 240 is used to adjust the parameters to be corrected of the wind turbine blade simulation model by using the model parameter adjustment method when the error of the wind turbine blade simulation model is greater than a preset threshold, and return to call the simulation analysis module; The parameter output module 250 is used to output the adjusted parameters of the wind turbine blade simulation model when the error of the wind turbine blade simulation model is less than a preset threshold.

[0112] Optionally, the error analysis module 230 is specifically used for: The following formula is used to calculate the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index to obtain the error of the wind turbine blade simulation model; ;in, represents the error of the wind turbine blade simulation model, The simulation results representing the evaluation index of the wind turbine blade simulation model, Indicates the preset test results of the evaluation indicators.

[0113] It should be understood that the device corresponds to the above-mentioned wind turbine blade model parameter correction method embodiment, and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be found in the above description. To avoid repetition, the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or fixed in the operating system (OS) of the device.

[0114] Embodiment 3 Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, including: a processor and a memory, the memory storing machine-readable instructions executable by the processor, and the machine-readable instructions executing the above-mentioned wind turbine blade model parameter correction method when executed by the processor.

[0115] In a typical configuration, an electronic device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0116] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0117] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0118] Embodiment 4 Based on the same inventive concept, an embodiment of the present invention further provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer executes the above-mentioned wind turbine blade model parameter correction method.

[0119] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0120] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0121] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0122] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0123] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the embodiments of the present invention will not further describe various possible combinations.

[0124] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0125] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0126] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for correcting wind turbine blade model parameters, characterized in that: include: Obtaining parameters of a wind turbine blade simulation model; wherein the parameters include parameters to be corrected and fixed parameters; Determine the simulation results of the evaluation index of the wind turbine blade simulation model according to the parameters of the wind turbine blade simulation model; Determining the error of the wind turbine blade simulation model according to the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index; If the error of the wind turbine blade simulation model is greater than a preset threshold, the parameters to be corrected of the wind turbine blade simulation model are adjusted using a model parameter adjustment method, and a simulation result of the evaluation index of the wind turbine blade simulation model is determined by performing simulation analysis based on the parameters of the wind turbine blade simulation model; If the error of the wind turbine blade simulation model is less than a preset threshold, the adjusted parameters of the wind turbine blade simulation model are output.

2. The wind turbine blade model parameter correction method according to claim 1, characterized in that: The parameters to be corrected for the wind turbine blade model include: blade distributed mass, flapping direction stiffness, swing direction stiffness and torsional stiffness; The evaluation indicators of the wind turbine blade simulation model include: blade mass, swing frequency, flapping frequency and torsion frequency.

3. The method for correcting wind turbine blade model parameters according to claim 1, characterized in that: According to the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index, the error of the wind turbine blade simulation model is determined, including: The following formula is used to calculate the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index to obtain the error of the wind turbine blade simulation model; ;in, represents the error of the wind turbine blade simulation model, The simulation results representing the evaluation index of the wind turbine blade simulation model, Indicates the preset test results of the evaluation indicators.

4. The method for correcting wind turbine blade model parameters according to claim 1, characterized in that: The parameters to be corrected of the wind turbine blade simulation model are adjusted using the model parameter adjustment method, including: Determine a plurality of wind turbine blade simulation model design schemes; wherein each wind turbine blade simulation model design scheme includes different parameters to be corrected of the wind turbine blade simulation model; Determining simulation results of evaluation indicators of multiple wind turbine blade simulation model design schemes based on multiple wind turbine blade simulation model design schemes and fixed parameters of the wind turbine blade simulation models; The pre-built convolutional neural network-long short-term memory neural network parallel recurrent neural network is trained using the simulation results of multiple wind turbine blade simulation model design schemes and corresponding evaluation indicators, and a wind turbine blade parameter correction model is obtained after the training is completed; Input the test results of the preset evaluation indexes into the wind turbine blade parameter correction model to obtain the target design scheme of the wind turbine blade simulation model; The parameters to be corrected of the wind turbine blade simulation model are updated to a target design scheme of the wind turbine blade simulation model.

5. The method for correcting wind turbine blade model parameters according to claim 4, characterized in that: Determine multiple wind turbine blade simulation model design options, including: Determining value ranges of multiple parameters to be corrected in the parameters of the wind turbine blade model; Selecting multiple values ​​of the parameters to be corrected from the value range of each parameter to be corrected; By utilizing a uniform design scheme, multiple values ​​of each parameter to be corrected are combined to obtain multiple wind turbine blade simulation model design schemes.

6. The method for correcting wind turbine blade model parameters according to claim 1, characterized in that: The convolutional neural network includes: using the Conv1D layer in the layers module of the Keras library to build a one-dimensional convolutional neural network; the long short-term memory neural network includes: using the Bidirectional LSTM layer in the layers module of the Keras library to build a bidirectional long short-term memory neural network model; the recurrent neural network includes: using the SimpleRNN layer in the layers module of the Keras library.

7. A wind turbine blade model parameter correction device, characterized in that: include: A data acquisition module is used to acquire parameters of a wind turbine blade simulation model; wherein the parameters include parameters to be corrected and fixed parameters; A simulation analysis module, used to determine the simulation results of the evaluation index of the wind turbine blade simulation model according to the parameters of the wind turbine blade simulation model; An error analysis module, used to determine the error of the wind turbine blade simulation model based on the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index; A simulation cycle module is used to adjust the parameters to be corrected of the wind turbine blade simulation model by using a model parameter adjustment method when the error of the wind turbine blade simulation model is greater than a preset threshold, and return to call the simulation analysis module; The parameter output module is used to output the adjusted parameters of the wind turbine blade simulation model when the error of the wind turbine blade simulation model is less than a preset threshold.

8. The wind turbine blade model parameter correction device according to claim 7, characterized in that: The error analysis module is specifically used for: The following formula is used to calculate the simulation results of the evaluation index of the wind turbine blade simulation model and the preset test results of the evaluation index to obtain the error of the wind turbine blade simulation model; ;in, represents the error of the wind turbine blade simulation model, The simulation results representing the evaluation index of the wind turbine blade simulation model, Indicates the preset test results of the evaluation indicators.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and when the machine-readable instructions are executed by the processor, the method for correcting wind turbine blade model parameters according to any one of claims 1 to 7 is executed.

10. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed on a computer, the computer is enabled to execute the wind turbine blade model parameter correction method described in any one of claims 1 to 7.

Images