Wind power blade corrosion resistance evaluation method and system based on multiple rainwater conditions

Through comprehensive experiments and simulation data, combined with neural network model, the problem of high time-consuming and cost-effective evaluation of traditional wind power blade corrosion resistance performance is solved, and the comprehensive evaluation of wind power blades under various rainwater conditions is achieved to adapt to changing environments. Reduced assessment costs.

CN119066801APending Publication Date: 2024-12-03CHANGZHOU HEXIN DAXU NEW ENERGY TECH DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411042812.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The traditional wind power blade corrosion resistance evaluation method is time-consuming and costly, making it difficult to conduct large-scale evaluation under different environmental conditions. The existing simulation methods lack comprehensive matching with the actual environment, and ignore the corrosion resistance of variable rainwater conditions on the blade. Comprehensive impact.

Method used

A method of corrosion resistance evaluation of wind power blades based on a variety of rainwater conditions is adopted, and a neural network model is established for corrosion resistance evaluation through the comprehensive use of experimental and simulated data. The specific steps include setting experimental conditions for rain erosion testing, establishing geometric models for simulated rain erosion testing, combining experimental and simulation data, and selecting appropriate neural network structures for model training and optimization.

Benefits of technology

It realizes a comprehensive and accurate corrosion resistance evaluation of wind power blades under different rainwater conditions, provides more reliable evaluation results, adapts to variable environmental conditions, reduces experimental costs, and improves the cost of the evaluation process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066801A_ABST
    Figure CN119066801A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of corrosion resistance evaluation, in particular to a wind power blade corrosion resistance evaluation method and system based on multiple rainwater conditions, and the method comprises the steps: carrying out rain erosion tests under multiple types of rainwater conditions, and collecting the experimental data of the rain erosion tests; performing a rain erosion simulation test, and collecting simulation data of the rain erosion simulation test; merging the experimental data and the simulation data to obtain a comprehensive data set; selecting a neural network structure as a wind power blade corrosion resistance evaluation model, and training, evaluating and optimizing the wind power blade corrosion resistance evaluation model; and performing wind power blade corrosion resistance evaluation by using the optimized wind power blade corrosion resistance evaluation model. According to the method, the corrosion resistance of the wind power blade under different rainwater conditions can be comprehensively and accurately evaluated, a more reliable evaluation result is provided, and the method can adapt to variable environmental conditions, so that the corrosion resistance of the wind power blade in an actual environment is more comprehensively reflected, and the performance optimization of the wind power blade is facilitated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of corrosion resistance evaluation, and in particular to a method and system for evaluating the corrosion resistance of wind turbine blades based on various rain conditions. Background Art

[0002] Wind turbine blades are key components in modern wind power generation systems, and their design and performance directly affect the efficiency and economy of wind turbines. Wind turbine blades are generally made of high-strength composite materials to provide sufficient strength and durability. These blades are subjected to huge aerodynamic pressure during wind power generation and are also exposed to various meteorological conditions, such as wind, rain, snow, and ultraviolet radiation. Since wind turbine blades are exposed to the natural environment for a long time, their surfaces may suffer various forms of corrosion and damage, especially in rainy or humid environments, where the blade surface may be eroded by raindrop impact. Rain erosion refers to raindrops hitting the surface of the blade, causing the peeling and wear of the surface material. This phenomenon will reduce the structural strength and performance of the blade, and affect the power generation efficiency and life of the wind turbine. Therefore, it is particularly important to evaluate the corrosion resistance of wind turbine blades.

[0003] Traditional wind turbine blade corrosion performance evaluation methods are usually time-consuming and costly, requiring long-term actual testing in the laboratory or on-site, and difficult to conduct large-scale evaluations under different environmental conditions. Although existing simulation methods can provide some data support, they usually lack comprehensive matching with the actual environment, resulting in a certain gap between the simulation results and the actual situation. In addition, most existing methods focus on the simulation of a single type of rainwater, ignoring the comprehensive impact of the variable rainwater conditions in the actual environment on the corrosion resistance of the blades. Therefore, the wind turbine blade corrosion performance evaluation method based on multiple rainwater conditions has become one of the current research directions. Summary of the invention

[0004] The present invention provides a method and system for evaluating the corrosion resistance of wind turbine blades based on various rain conditions, which can effectively solve the problems in the background technology.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: A method for evaluating the corrosion resistance of wind turbine blades based on various rain conditions, the method comprising: Setting experimental conditions for a wind turbine blade rain erosion test, conducting the rain erosion test under several types of rain conditions, and collecting experimental data of the rain erosion test; Establishing a geometric model of the wind turbine blade, setting several types of rain conditions, performing a simulated rain erosion test, and collecting simulation data of the simulated rain erosion test; Combining the experimental data with the simulation data to obtain a comprehensive data set; Selecting a suitable neural network structure as a wind turbine blade corrosion resistance performance evaluation model, and training, evaluating and optimizing the wind turbine blade corrosion resistance performance evaluation model through the comprehensive data set; The wind turbine blade corrosion resistance performance evaluation is performed using the optimized wind turbine blade corrosion resistance performance evaluation model.

[0006] Furthermore, experimental conditions for the wind turbine blade rain erosion test are set, the rain erosion test is performed under several types of rain conditions, and experimental data of the rain erosion test is collected, including: Determine the experimental objectives, prepare several wind turbine blade samples, and set the experimental conditions; Before the rain erosion test, a preliminary measurement is performed on the wear conditions of the wind turbine blade samples; Setting several types of rain conditions and test times to respectively carry out the rain erosion test, and measuring the wear conditions of the several wind turbine blade samples again; Comparing the wear conditions before and after the rain erosion test to obtain wear data of the wind turbine blade; The wear data and the environmental parameters of the rain erosion test are collected to obtain experimental data of the rain erosion test.

[0007] Furthermore, during the rain erosion test, a high-speed camera is used to record the raindrop impact process.

[0008] Furthermore, a geometric model of the wind turbine blade is established, several types of rain conditions are set, a simulated rain erosion test is performed, and simulation data of the simulated rain erosion test is collected, including: Creating a geometric model of the wind turbine blade and defining material properties of the geometric model; Add the physical field of raindrop flow, set several types of rain conditions, and define the boundary conditions and initial conditions of the simulation; Meshing the geometric model, starting simulation calculation, and obtaining simulated wear data of the wind turbine blade; The simulated wear data and environmental parameters of the simulation process are collected to obtain simulation data of the simulated rain erosion test.

[0009] Furthermore, meshing the geometric model includes: Select the mesh type, set the global mesh parameters, and generate the global mesh using the mesh generation algorithm; Determine the refinement area, set local mesh refinement parameters, and run the refinement algorithm to generate the local mesh; Performing a quality check on the generated global grid and the local grid, and optimizing the global grid and the local grid according to the quality check result.

[0010] Furthermore, the experimental data and the simulation data are combined to obtain a comprehensive data set, including: Preprocessing the experimental data and the simulation data; Mapping the preprocessed experimental data and the simulation data to the same standard, performing data conversion, and achieving alignment between the experimental data and the simulation data; The aligned experimental data and simulation data are converted into a unified format, and data splicing is performed to obtain a comprehensive data set.

[0011] Furthermore, the experimental data and the simulation data are combined to perform consistency check and outlier detection.

[0012] A wind turbine blade corrosion resistance evaluation system based on various rain conditions, the system comprising: An experimental data collection module is used to set experimental conditions for the wind turbine blade rain erosion test, conduct the rain erosion test under several types of rain conditions, and collect experimental data of the rain erosion test; A simulation data collection module is used to establish a geometric model of the wind turbine blade, set several types of rain conditions, perform a simulated rain erosion test, and collect simulation data of the simulated rain erosion test; A data merging module, merging the experimental data and the simulation data to obtain a comprehensive data set; An evaluation model building module selects a suitable neural network structure as a wind turbine blade corrosion resistance performance evaluation model, and trains, evaluates and optimizes the wind turbine blade corrosion resistance performance evaluation model through the comprehensive data set; The evaluation model application module uses the optimized wind turbine blade corrosion resistance performance evaluation model to evaluate the wind turbine blade corrosion resistance performance.

[0013] Furthermore, the experimental data collection module includes: Experiment preparation unit, which determines the experimental objectives, prepares several wind turbine blade samples, and sets the experimental conditions; A preliminary measurement unit, which performs preliminary measurement on the wear conditions of the wind turbine blade samples before the rain erosion test; The measuring unit is used to set several types of rain conditions and test times to respectively perform the rain erosion test, and to measure the wear conditions of the several wind turbine blade samples again; A data comparison unit, for comparing the wear conditions before and after the rain erosion test to obtain wear data of the wind turbine blade; The experimental data collection unit collects the wear data and the environmental parameters of the rain erosion test to obtain the experimental data of the rain erosion test.

[0014] Furthermore, the simulation data collection module includes: A model creation unit, which creates a geometric model of the wind turbine blade and defines material properties of the geometric model; Simulation preparation unit, add the physical field of raindrop flow, set several types of rain conditions, and define the boundary conditions and initial conditions of the simulation; Running a simulation unit, meshing the geometric model, starting simulation calculation, and obtaining simulated wear data of the wind turbine blade; The simulation data collection unit collects the simulated wear data and environmental parameters of the simulation process to obtain simulation data of the simulated rain erosion test.

[0015] The technical solution of the present invention can achieve the following technical effects: The present invention uses a combination of experimental data and simulation data, and combines with advanced neural network models to comprehensively and accurately evaluate the corrosion resistance of wind turbine blades under different rain conditions, providing more reliable evaluation results. Moreover, the evaluation is based on a variety of rain conditions and can adapt to changing environmental conditions, thereby more comprehensively reflecting the corrosion resistance of wind turbine blades in actual environments, which is helpful for optimizing the performance of wind turbine blades, while greatly reducing experimental costs and making the evaluation process more economical and efficient.

[0016] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flow chart of the wind turbine blade corrosion resistance evaluation method based on various rain conditions; Figure 2 A schematic diagram of the process of meshing a geometric model; Figure 3 This is a schematic diagram of the structure of the wind turbine blade corrosion resistance evaluation system based on various rain conditions. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0021] Embodiment 1 like Figure 1 As shown, a method for evaluating the corrosion resistance of wind turbine blades based on various rain conditions includes: A1: Set experimental conditions for wind turbine blade rain erosion test, conduct rain erosion tests under several types of rain conditions, and collect experimental data of rain erosion test; This step evaluates the corrosion resistance of blade materials under different rain conditions through rain erosion test experiments. Detailed experimental conditions can be set to ensure the reliability and repeatability of experimental data, and the setting of multiple rain conditions can fully reflect the impact of different environments on blades. In the laboratory environment, different types of rain are simulated, including acid rain, ocean rain, ordinary rain, etc. The environmental data, experimental conditions and experimental result data during the test are collected as one of the inputs of the subsequent model.

[0022] A2: Establish a geometric model of wind turbine blades, set several types of rain conditions, conduct simulated rain erosion tests, and collect simulated data of the simulated rain erosion tests; Specifically, computer simulation can be used to evaluate the corrosion resistance of wind turbine blades under different rain conditions, saving experimental costs and time. Detailed simulation data can also be provided. Similarly, environmental data, simulation conditions, and simulation result data collected during the simulation process can be used as one of the inputs for subsequent models.

[0023] A3: Combine experimental data and simulation data to obtain a comprehensive data set; A4: Select a suitable neural network structure as the wind turbine blade corrosion performance evaluation model, and train, evaluate and optimize the wind turbine blade corrosion performance evaluation model through a comprehensive data set; In this step, a suitable neural network structure can be selected according to the characteristics of the data, such as a multi-layer perceptron, a convolutional neural network, a recursive neural network, etc. The neural network architecture is trained through a comprehensive data set. During the training process, the comprehensive data set is divided into a training set, a validation set, and a test set. The model is trained using the training set. The neural network can be trained using a back propagation algorithm and an optimizer. The optimizer can update the network weights and biases using a gradient descent method to minimize the loss function. The model performance is evaluated and hyperparameters are adjusted using a validation set. The validation set can be used to evaluate the performance indicators of the model, such as accuracy, F1-score, mean square error, etc. The hyperparameters are adjusted according to the verification results, including the learning rate, batch size, number of neural network layers, number of neurons in each layer, etc. Grid search, random search, or Bayesian optimization can be used to find the best hyperparameter combination. If the amount of data is sufficient, a cross-validation method can be used to further verify the stability and reliability of the model, and finally a test set can be used to evaluate the generalization ability of the model.

[0024] A5: Use the optimized wind turbine blade corrosion resistance performance evaluation model to evaluate the corrosion resistance of wind turbine blades.

[0025] Specifically, in order to obtain more comprehensive data, experimental data and simulation data can be collected, and the input features of the model can be provided by combining real experimental data and a large amount of simulation data, which can cover more variables and environmental conditions, enrich the diversity of training data, and improve the generalization ability of the model. By training and optimizing the evaluation model with experimental data and simulation data, the neural network model can learn more comprehensive features, thereby improving the accuracy and reliability of the prediction.

[0026] The present invention uses a combination of experimental data and simulation data, and combines with advanced neural network models to comprehensively and accurately evaluate the corrosion resistance of wind turbine blades under different rain conditions, providing more reliable evaluation results. Moreover, the evaluation is based on a variety of rain conditions and can adapt to changing environmental conditions, thereby more comprehensively reflecting the corrosion resistance of wind turbine blades in actual environments, which is helpful for optimizing the performance of wind turbine blades, while greatly reducing experimental costs and making the evaluation process more economical and efficient.

[0027] On the basis of the above embodiment, experimental conditions for the wind turbine blade rain erosion test are set, rain erosion tests are carried out under several types of rain conditions, and experimental data of the rain erosion test are collected, including: B1: Determine the experimental objectives, prepare several wind turbine blade samples, and set the experimental conditions; In this embodiment, the experimental goal is first determined to evaluate the corrosion resistance of wind turbine blades under different types of rain conditions, such as acid rain, salt spray, industrial pollution rain, etc. Several groups of wind turbine blade samples of the same material and size can be prepared for the experiment to ensure the comparability of the test, and the selected wind turbine blade samples should be representative to obtain reliable data. The experimental conditions are clarified to include the type of rain, such as acid rain, salt spray, polluted rain, etc., the physical parameters of raindrops, such as diameter, speed, impact angle, and environmental parameters, such as temperature, humidity, etc. The rain erosion test experiment can be equipped with equipment that can simulate different types of rain, such as acid rain generators, salt spray test chambers, etc., with the function of adjusting the spray intensity and flow rate, and a mechanism that can control the rotation speed is installed to simulate the actual working state of the wind turbine blade. It should be noted that in order to ensure the standardization and repeatability of the test process and to obtain reliable experimental data, the samples can be measured multiple times at different time points or different locations to obtain more comprehensive data.

[0028] B2: Conduct preliminary measurements of the wear of several wind turbine blade samples before the rain erosion test; B3: Set several types of rain conditions and test times to conduct rain erosion tests, and measure the wear of several wind turbine blade samples again; B4: Compare the wear conditions before and after the rain erosion test to obtain the wear data of the wind turbine blades; B5: Collect wear data and environmental parameters of rain erosion test to obtain experimental data of rain erosion test.

[0029] Specifically, the rain simulation system and the rotating mechanism can be adjusted during the rain erosion test to ensure accurate parameters. The blades are measured before the rain erosion test to measure the surface wear, weight and microstructure of the sample. After the test, the sample is measured again to record the surface wear, weight change and microstructure change. At the end of the experiment, the experimental data can be recorded to create a standardized data record table, including the sample number, experimental conditions, preliminary measurement data, test time, subsequent measurement data, etc.

[0030] Furthermore, during the rain erosion test, a high-speed camera was used to record the raindrop impact process.

[0031] On the basis of the above embodiments, in order to fully capture and analyze the dynamic behavior of raindrop impact, evaluate its impact on the surface of wind turbine blades, optimize test and simulation conditions, record extreme situations, and provide strong data support, a high-speed camera can be used to record the raindrop impact process. The high-speed camera can record the impact process of raindrops at an extremely high frame rate, thereby capturing the dynamic characteristics of the raindrops such as the motion trajectory, impact angle and impact force in detail. The actual impact process recorded by the high-speed camera is compared with the computer simulation results to verify the accuracy and reliability of the simulation. And the test parameters, such as raindrop speed, size, impact angle, etc., can be adjusted according to the actual impact behavior to ensure that the test conditions are consistent with the actual working conditions, optimize the test conditions and simulation parameters, and improve the accuracy and effectiveness of the corrosion resistance evaluation of wind turbine blades.

[0032] As a preferred embodiment of this invention, a geometric model of a wind turbine blade is established, several types of rain conditions are set, a simulated rain erosion test is performed, and simulation data of the simulated rain erosion test is collected, including: C1: Create the geometric model of the wind turbine blade and define the material properties of the geometric model; Specifically, the first step of the simulation experiment is to determine the specific goals that need to be simulated, such as blade wear, mass change, chemical composition change, etc. under specific rain types. After that, this step can obtain the design drawings or CAD files of the wind turbine blades, use computer-aided design software or finite element analysis software, such as SolidWorks, ANSYS, etc., to create a three-dimensional geometric model to ensure that the model accurately reflects the actual size and shape of the blade; and define material properties for the geometric model, including density, elastic modulus, Poisson's ratio, hardness, etc.

[0033] C2: Add the physical field of raindrop flow, set several types of rain conditions, and define the boundary conditions and initial conditions of the simulation; The simulated rain erosion test simulates the corrosion resistance of wind turbine blades under different rain conditions. Therefore, the physical field of raindrops hitting the surface of wind turbine blades can be set, including the speed, diameter and impact angle of raindrops. Since it is necessary to simulate the impact of different rain conditions on blades, the chemical properties of rainwater can be defined, such as acidity and salinity. Boundary conditions are constraints or conditions imposed on the boundaries of the calculation area. Boundary conditions in the simulation are set, such as fixed points of blades, applied external forces, etc. Initial conditions refer to the initial values ​​assigned to physical quantities at each point in the calculation domain at the beginning of time integration or iterative calculations. The initial conditions include initial raindrop distribution, initial fluid velocity field, etc.

[0034] C3: Mesh the geometric model, start simulation calculation, and obtain simulated wear data of wind turbine blades; C4: Collect simulated wear data and environmental parameters of the simulation process to obtain simulated data of the simulated rain erosion test.

[0035] In this embodiment, the geometric model can be divided into small calculation units to generate a calculation grid. The quality and density of the grid will affect the accuracy of the simulation results. Therefore, when the model is meshed, it is necessary to ensure that the quality and detail of the grid meet the simulation accuracy requirements. When performing a simulation, the running status of the simulation can be monitored to ensure stability and accuracy during the simulation process.

[0036] Furthermore, if Figure 2 As shown, meshing of the geometric model is performed, including: D1: Select the mesh type, set the global mesh parameters, and generate the global mesh through the mesh generation algorithm; In this step, you can select a suitable mesh type based on the complexity of the geometric model and the requirements of the simulation. The mesh types are: (1) Structured grid Structured grids are suitable for regular geometric shapes and are composed of regular grid cells, such as cubes or cuboids. Structured grid cells are arranged neatly, easy to generate, and have high computational efficiency, but they are not suitable for complex geometric shapes.

[0037] (2) Unstructured grid Unstructured grids are suitable for complex geometric shapes, where the grid cells are arranged irregularly, such as triangles, tetrahedrons, etc. Unstructured grids are highly adaptable and can handle complex geometries, but they have low computational efficiency and are more complex to generate and optimize.

[0038] (3) Hybrid grid Hybrid grids combine the advantages of structured grids and unstructured grids and are used to combine the advantages of different regions. Structured grids can be used in simple areas and unstructured grids can be used in complex areas. The disadvantage is that grid generation and optimization are complex.

[0039] After the mesh type is determined, you can define the overall density of the mesh and the size of the initial mesh cells, select the mesh generation algorithm, such as Delaunay triangulation, octree decomposition, etc., and use the mesh generation tool to create the initial mesh of the entire geometric model. It should be noted that setting the mesh density and cell size will directly affect the accuracy and calculation time of the simulation. A higher mesh density provides more accurate results, but the calculation burden is heavier.

[0040] D2: Determine the refinement area, set local mesh refinement parameters, and run the refinement algorithm to generate the local mesh; This step can identify key areas, determine key areas that need to be refined, and mark these areas for local mesh refinement; setting local mesh refinement parameters can define the refinement level and set the size and density of the mesh in the refined area. For example, the mesh in the boundary layer area may require a smaller unit size; you can also set the parameters of the refined mesh, such as the local mesh size, gradient ratio, etc.

[0041] D3: Perform quality check on the generated global mesh and local mesh, and optimize the global mesh and local mesh based on the quality check results.

[0042] After the mesh is built, a comprehensive check can be performed to check the quality of the overall mesh, including the global mesh and the local refined mesh, including the shape, size and uniformity of the mesh cells, to ensure that the transition between the refined mesh and the global mesh is smooth and there are no irregular or distorted mesh cells. According to the results of the quality check, the mesh parameters can be adjusted, such as increasing the mesh refinement, adjusting the local mesh density or improving the mesh generation algorithm, to improve the quality of the mesh and the accuracy of the simulation results.

[0043] Based on the above embodiment, the experimental data and the simulation data are combined to obtain a comprehensive data set, including: E1: Preprocessing of experimental and simulation data; In order to ensure the accuracy and consistency of the data, the experimental data and simulation data can be preprocessed, including data cleaning, standardization and normalization. Data cleaning can remove noise and outliers in the data, fill or process missing values ​​in the data, delete duplicate data, and ensure that each data is independent; data standardization can convert the data into a distribution with zero mean and unit variance, and normalization can scale the data to the range of [0, 1] or [-1, 1]. Standardization and normalization help to compare different features on the same scale, preventing some features from having an unbalanced impact on the model due to too large or too small numerical ranges.

[0044] E2: Map the preprocessed experimental data and simulation data to the same standard, perform data conversion, and achieve alignment between experimental data and simulation data; In this step, data alignment can select common standards, such as time and conditions, as the basis for data alignment. For data collected at different time points, time interpolation or resampling can be performed to ensure that the data are aligned at the same time point. Experimental and simulation conditions can also be mapped to ensure that data are compared under the same conditions, for example, alignment is performed based on experimental and simulated parameters such as temperature and humidity. Data alignment can ensure that experimental data and simulation data are compared at the same time point and under the same conditions, avoiding errors caused by differences in time and conditions.

[0045] E3: Convert the aligned experimental data and simulation data into a unified format and perform data splicing to obtain a comprehensive data set.

[0046] Specifically, data splicing can be achieved through vertical splicing and horizontal splicing. Before data splicing, a unified data format can be selected for the experimental data and the simulation data to maintain field consistency and ensure that the experimental data and the simulation data contain the same fields or columns for easy splicing.

[0047] In order to ensure the high quality and consistency of the merged dataset, the experimental data and simulation data were merged and then subjected to consistency check and outlier detection.

[0048] In this embodiment, the consistency check verifies the merged data set to ensure the consistency of the data in terms of structure, format, unit and range, and ensure that the various parts of the data set are consistent in structure and format, which can avoid processing errors caused by different data formats, and the consistent data structure and format will help the subsequent data analysis and model training, and reduce errors caused by data problems. Outlier detection is to identify and process those values ​​in the data set that deviate significantly from the normal range. Outliers can be detected through statistical methods, visualization methods, rule methods, etc., to avoid the negative impact of outliers on model training and analysis results, and improve the accuracy and reliability of the model.

[0049] like Figure 3 As shown, a wind turbine blade corrosion resistance evaluation system based on various rain conditions includes: The experimental data collection module sets the experimental conditions for the wind turbine blade rain erosion test, conducts rain erosion tests under several types of rain conditions, and collects experimental data for the rain erosion test; The simulation data collection module establishes the geometric model of the wind turbine blade, sets several types of rain conditions, performs simulated rain erosion tests, and collects simulation data of the simulated rain erosion tests; The data merging module merges the experimental data and the simulation data to obtain a comprehensive data set; The evaluation model building module selects a suitable neural network structure as the wind turbine blade corrosion performance evaluation model, and trains, evaluates and optimizes the wind turbine blade corrosion performance evaluation model through a comprehensive data set; The evaluation model application module uses the optimized wind turbine blade corrosion resistance performance evaluation model to evaluate the corrosion resistance performance of wind turbine blades.

[0050] The above-mentioned adjustment system in the present invention can effectively implement the wind turbine blade corrosion resistance performance evaluation method based on various rain conditions, and the technical effects that can be achieved are as described in the above-mentioned embodiments and will not be repeated here.

[0051] Furthermore, the experimental data collection module includes: Experiment preparation unit, which determines the experimental objectives, prepares several wind turbine blade samples, and sets the experimental conditions; The preliminary measurement unit conducts preliminary measurement of the wear of several wind turbine blade samples before the rain erosion test; The measurement unit was measured again, and several types of rain conditions and test times were set to conduct rain erosion tests respectively, and the wear of several wind turbine blade samples was measured again; A data comparison unit compares the wear conditions before and after the rain erosion test to obtain the wear data of the wind turbine blades; The experimental data collection unit collects the wear data and environmental parameters of the rain erosion test to obtain the experimental data of the rain erosion test.

[0052] Furthermore, the simulation data collection module includes: A model creation unit, which creates a geometric model of a wind turbine blade and defines material properties of the geometric model; Simulation preparation unit, add the physical field of raindrop flow, set several types of rain conditions, and define the boundary conditions and initial conditions of the simulation; Run the simulation unit, mesh the geometric model, start the simulation calculation, and obtain the simulated wear data of the wind turbine blades; The simulation data collection unit collects the simulated wear data and the environmental parameters of the simulation process to obtain the simulation data of the simulated rain erosion test.

[0053] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0054] Although the present application has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined herein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for evaluating the corrosion resistance of wind turbine blades based on various rain conditions, characterized in that: The method comprises: Setting experimental conditions for a wind turbine blade rain erosion test, conducting the rain erosion test under several types of rain conditions, and collecting experimental data of the rain erosion test; Establishing a geometric model of the wind turbine blade, setting several types of rain conditions, performing a simulated rain erosion test, and collecting simulation data of the simulated rain erosion test; Combining the experimental data with the simulation data to obtain a comprehensive data set; Selecting a suitable neural network structure as a wind turbine blade corrosion resistance performance evaluation model, and training, evaluating and optimizing the wind turbine blade corrosion resistance performance evaluation model through the comprehensive data set; The wind turbine blade corrosion resistance performance evaluation is performed using the optimized wind turbine blade corrosion resistance performance evaluation model.

2. The method for evaluating the corrosion resistance of wind turbine blades based on multiple rain conditions according to claim 1 is characterized in that: Setting experimental conditions for the wind turbine blade rain erosion test, conducting the rain erosion test under several types of rain conditions, and collecting experimental data of the rain erosion test, including: Determine the experimental objectives, prepare several wind turbine blade samples, and set the experimental conditions; Before the rain erosion test, a preliminary measurement is performed on the wear conditions of the wind turbine blade samples; Setting several types of rain conditions and test times to respectively carry out the rain erosion test, and measuring the wear conditions of the several wind turbine blade samples again; Comparing the wear conditions before and after the rain erosion test to obtain wear data of the wind turbine blade; The wear data and the environmental parameters of the rain erosion test are collected to obtain experimental data of the rain erosion test.

3. The method for evaluating the corrosion resistance of wind turbine blades based on multiple rain conditions according to claim 2 is characterized in that: During the rain erosion test, a high-speed camera was used to record the raindrop impact process.

4. The method for evaluating the corrosion resistance of wind turbine blades based on multiple rain conditions according to claim 1 is characterized in that: Establishing a geometric model of the wind turbine blade, setting several types of rain conditions, conducting a simulated rain erosion test, and collecting simulation data of the simulated rain erosion test, including: Creating a geometric model of the wind turbine blade and defining material properties of the geometric model; Add the physical field of raindrop flow, set several types of rain conditions, and define the boundary conditions and initial conditions of the simulation; Meshing the geometric model, starting simulation calculation, and obtaining simulated wear data of the wind turbine blade; The simulated wear data and environmental parameters of the simulation process are collected to obtain simulation data of the simulated rain erosion test.

5. The method for evaluating the corrosion resistance of wind turbine blades based on multiple rain conditions according to claim 4 is characterized in that: Meshing the geometric model includes: Select the mesh type, set the global mesh parameters, and generate the global mesh using the mesh generation algorithm; Determine the refinement area, set local mesh refinement parameters, and run the refinement algorithm to generate the local mesh; Performing a quality check on the generated global grid and the local grid, and optimizing the global grid and the local grid according to the quality check result.

6. The method for evaluating the corrosion resistance of wind turbine blades based on multiple rain conditions according to claim 1, characterized in that: The experimental data and the simulation data are combined to obtain a comprehensive data set, including: Preprocessing the experimental data and the simulation data; Mapping the preprocessed experimental data and the simulation data to the same standard, performing data conversion, and achieving alignment between the experimental data and the simulation data; The aligned experimental data and simulation data are converted into a unified format, and data splicing is performed to obtain a comprehensive data set.

7. The method for evaluating the corrosion resistance of wind turbine blades based on multiple rain conditions according to claim 6 is characterized in that: The experimental data and the simulation data are combined and then consistency check and outlier detection are performed.

8. A wind turbine blade corrosion resistance evaluation system based on various rain conditions, characterized in that: The system comprises: An experimental data collection module is used to set experimental conditions for the wind turbine blade rain erosion test, conduct the rain erosion test under several types of rain conditions, and collect experimental data of the rain erosion test; A simulation data collection module is used to establish a geometric model of the wind turbine blade, set several types of rain conditions, perform a simulated rain erosion test, and collect simulation data of the simulated rain erosion test; A data merging module, merging the experimental data and the simulation data to obtain a comprehensive data set; An evaluation model building module selects a suitable neural network structure as a wind turbine blade corrosion resistance performance evaluation model, and trains, evaluates and optimizes the wind turbine blade corrosion resistance performance evaluation model through the comprehensive data set; The evaluation model application module uses the optimized wind turbine blade corrosion resistance performance evaluation model to evaluate the wind turbine blade corrosion resistance performance.

9. The wind turbine blade corrosion resistance evaluation system based on multiple rain conditions according to claim 8, characterized in that: The experimental data collection module comprises: Experiment preparation unit, which determines the experimental objectives, prepares several wind turbine blade samples, and sets the experimental conditions; A preliminary measurement unit, which performs preliminary measurement on the wear conditions of the wind turbine blade samples before the rain erosion test; The measuring unit is used to set several types of rain conditions and test times to respectively perform the rain erosion test, and to measure the wear conditions of the several wind turbine blade samples again; A data comparison unit, for comparing the wear conditions before and after the rain erosion test to obtain wear data of the wind turbine blade; The experimental data collection unit collects the wear data and the environmental parameters of the rain erosion test to obtain the experimental data of the rain erosion test.

10. The wind turbine blade corrosion resistance evaluation system based on multiple rain conditions according to claim 8, characterized in that: The simulation data collection module comprises: A model creation unit, which creates a geometric model of the wind turbine blade and defines material properties of the geometric model; Simulation preparation unit, add the physical field of raindrop flow, set several types of rain conditions, and define the boundary conditions and initial conditions of the simulation; Running a simulation unit, meshing the geometric model, starting simulation calculation, and obtaining simulated wear data of the wind turbine blade; The simulation data collection unit collects the simulated wear data and environmental parameters of the simulation process to obtain simulation data of the simulated rain erosion test.