Wind vibration performance evaluation method and evaluation system

Abstract

The invention provides a wind vibration performance evaluation method and evaluation system, and relates to the photovoltaic technical field, the method comprises the following steps: establishing a finite element model of a photovoltaic flexible support, and obtaining modal parameters of the finite element model; obtaining meteorological data of a to-be-operated environment of the photovoltaic flexible support, and determining a wind speed range and a wind direction range of the to-be-operated environment based on the meteorological data and the modal parameters; in the wind vibration simulation experiment based on the wind speed range and the wind direction range, recording wind vibration parameters of the photovoltaic flexible support under different wind directions and different wind speeds; and the wind vibration performance of the photovoltaic flexible support is obtained based on wind vibration parameter evaluation. The wind vibration simulation experiment is carried out based on the wind speed range and the wind direction range of the to-be-operated environment, the wind vibration parameters are obtained, and the wind vibration performance of the photovoltaic flexible support is accurately evaluated based on the wind vibration parameters.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic technology, and in particular to a wind vibration performance evaluation method and evaluation system. Background Art

[0002] With the growing global demand for renewable energy, solar photovoltaic technology has been widely used. As a new type of support system, photovoltaic flexible bracket has broad application prospects in the field of solar photovoltaic due to its advantages such as light weight, flexible installation and strong adaptability. However, in the actual use of photovoltaic flexible bracket, especially in severe weather conditions such as strong winds, its wind vibration performance often becomes a key factor affecting its safety and stability.

[0003] Existing wind vibration performance evaluation methods for photovoltaic flexible brackets mainly rely on actual field tests and theoretical calculations. When evaluating wind vibration performance through actual field tests, it will be limited by factors such as environment, time and cost. At the same time, actual field tests are also easily affected by test conditions and the technical level of testers. In addition, when evaluating wind vibration performance through theoretical calculations, it is necessary to simplify the bracket structure and external environmental factors. The simplified situation is often inconsistent with the actual situation. Therefore, the results of the two existing methods when evaluating wind vibration performance are usually greatly deviated from the actual wind vibration performance, and the accuracy of the wind vibration performance evaluation cannot be guaranteed. Summary of the invention

[0004] The present invention provides a wind vibration performance evaluation method and an evaluation system to solve the technical problem of low accuracy in evaluating the wind vibration performance of a photovoltaic flexible bracket in the existing method.

[0005] An embodiment of the present invention provides a method for evaluating wind vibration performance, which includes the following steps: establishing a finite element model of a photovoltaic flexible bracket, and obtaining modal parameters of the finite element model; wherein the modal parameters include natural frequency and vibration mode; obtaining meteorological data of the working environment of the photovoltaic flexible bracket, and determining the wind speed range and wind direction range of the working environment based on the meteorological data and the modal parameters; in a wind vibration simulation experiment based on the wind speed range and the wind direction range, recording the wind vibration parameters of the photovoltaic flexible bracket under different wind directions and different wind speeds; based on the wind vibration parameters, evaluating the wind vibration performance of the photovoltaic flexible bracket; wherein the wind vibration parameters include stress, strain and displacement of the photovoltaic flexible bracket.

[0006] Optionally, the step of establishing a finite element model of the photovoltaic flexible bracket includes: establishing the finite element model based on computer-aided design software, importing the finite element model into computer-aided engineering software, and setting model parameters of the finite element model; determining the load type of the working environment based on the meteorological data and applying different types of loads to the finite element model.

[0007] Optionally, the step of establishing the finite element model based on computer-aided design software includes: establishing the finite element model based on the initial file of the photovoltaic flexible bracket through computer-aided design software, taking into account the geometric shape, size, connection method and relative positions between components of the photovoltaic flexible bracket; wherein the initial file includes: design drawings, technical specifications and material lists.

[0008] Optionally, the step of setting the model parameters of the finite element model includes: setting the unit type, material parameters and boundary conditions of the finite element model; wherein the unit types include shell units, beam units and solid units, the material parameters include elastic modulus, Poisson's ratio, density and yield strength, and the boundary conditions include fixed constraints, sliding constraints and free boundaries.

[0009] Optionally, the step of obtaining meteorological data of the working environment of the photovoltaic flexible bracket includes: collecting meteorological data of the working environment, and preprocessing and statistically analyzing the meteorological data to obtain processed meteorological data; wherein the preprocessing includes cleaning, sorting and formatting.

[0010] Optionally, the step of determining the wind speed range and wind direction range of the working environment based on the meteorological data and the modal parameters includes: determining the wind speed range of the working environment based on the processed meteorological data in combination with a wind speed profile model, and removing the wind speed values ​​corresponding to the natural frequency within the wind speed range; determining the wind direction range of the working environment based on the processed meteorological data in combination with a wind direction stability analysis, and taking the wind direction in the wind direction range that causes the largest change in the vibration mode as the main wind direction.

[0011] Optionally, the step of obtaining the wind vibration performance of the photovoltaic flexible bracket based on the wind vibration parameters includes: training the machine learning algorithm model based on the collected sample data on the basis of the wind vibration performance evaluation rules; obtaining the characteristic parameters of the photovoltaic flexible bracket, preprocessing the collected wind vibration parameters, and inputting the preprocessed wind vibration parameters and the characteristic parameters into the trained machine learning algorithm model to obtain the wind vibration performance of the photovoltaic flexible bracket; wherein the characteristic parameters include wind speed, wind direction, and bracket material.

[0012] Optionally, the step of obtaining the wind vibration performance of the photovoltaic flexible bracket based on the wind vibration parameters also includes: inputting verification data into the trained machine learning algorithm model for verification, and establishing evaluation indicators based on the verification results to evaluate the performance of the machine learning algorithm model to obtain model performance evaluation results; wherein the evaluation indicators include accuracy, recall rate, and F1 score.

[0013] An embodiment of the present invention provides a method for evaluating wind vibration performance. By establishing a finite element model of a photovoltaic flexible bracket, the simplified processing of the photovoltaic flexible bracket is avoided, and the accuracy of the modal parameters of the photovoltaic flexible bracket is ensured. The wind direction range and wind speed range are obtained based on the modal parameters and the meteorological data of the working environment. A wind vibration simulation experiment is performed based on the wind speed range and wind direction range, avoiding the simplified processing of the working environment of the photovoltaic flexible bracket. The combination of wind speed and wind direction to which the photovoltaic flexible bracket is subjected in the working environment can be simulated to ensure the authenticity of the wind vibration simulation experiment. The wind vibration parameters of the photovoltaic flexible bracket under different wind speed and wind direction combinations are recorded, and the wind vibration performance of the photovoltaic flexible bracket is evaluated based on the wind vibration parameters, thereby ensuring the accuracy of the evaluation of its wind vibration performance.

[0014] An embodiment of the present invention further provides a wind-induced vibration performance evaluation system, which is used to execute the above-mentioned wind-induced vibration performance evaluation method.

[0015] Other features and advantages of the embodiments of the present invention will be described in the following description, or some features and advantages can be inferred or determined without doubt from the description, or can be learned by implementing the above-mentioned techniques of the embodiments of the present invention.

[0016] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. 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 embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0018] Figure 1 A flow chart of a method for evaluating wind vibration performance provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] The embodiment of the present invention provides a method for evaluating wind vibration performance, which can be applied to, for example, Figure 1 A flow chart of a method for evaluating wind vibration performance is shown, and the method for evaluating wind vibration performance comprises the following steps:

[0021] S102, establishing a finite element model of the photovoltaic flexible support, and obtaining modal parameters of the finite element model; wherein the modal parameters include natural frequency and vibration mode;

[0022] The photovoltaic flexible finite element model established in the above steps facilitates the acquisition of the modal parameters of the photovoltaic flexible bracket through the method based on finite element calculation modal analysis. The modal parameters are important parameters of the photovoltaic flexible bracket under dynamic conditions, which serve as the basis for the subsequent determination of wind speed and wind direction.

[0023] S104, obtaining meteorological data of the working environment of the photovoltaic flexible support, and determining the wind speed range and wind direction range of the working environment based on the meteorological data and modal parameters;

[0024] In the above steps, the meteorological data of the working environment is obtained to ensure the authenticity of the subsequent simulation. The corresponding wind speed within the natural frequency range is likely to cause resonance damage to the photovoltaic flexible bracket. Therefore, the subsequent experimental process should avoid the wind speed corresponding to the natural frequency range in the meteorological data. The vibration of the photovoltaic flexible bracket in different wind directions can be known through the vibration shape. The wind direction that causes the photovoltaic flexible bracket to have a larger vibration shape in the meteorological data can be used as the main research wind direction.

[0025] S106, in a wind vibration simulation experiment based on a wind speed range and a wind direction range, recording wind vibration parameters of the photovoltaic flexible support under different wind directions and different wind speeds;

[0026] In the above steps, a wind vibration simulation experiment based on the wind speed range and wind direction range is carried out, wind loads with different wind directions and wind speeds are applied to the photovoltaic flexible bracket, and the wind vibration parameters of the photovoltaic flexible bracket under different wind directions and wind speeds are recorded. The wind vibration parameters are used as the main indicators for subsequent evaluation of wind vibration performance.

[0027] S108, obtaining the wind vibration performance of the photovoltaic flexible support based on the wind vibration parameter evaluation; wherein the wind vibration parameter includes the stress, strain and displacement of the photovoltaic flexible support;

[0028] In the above steps, the stress of the photovoltaic flexible bracket reflects the force it bears per unit area. In the wind vibration assessment, the magnitude and distribution of the stress are directly related to the safety and stability of the structure; the strain reflects the shape change of the photovoltaic flexible bracket after being subjected to force. In the wind vibration assessment, the magnitude and distribution of the strain can help analyze the stiffness and deformation of the photovoltaic flexible bracket, and then evaluate the adaptability of the photovoltaic flexible bracket under wind load; the displacement reflects the moving distance of the photovoltaic flexible bracket under the action of wind load. In the wind vibration assessment, the magnitude and direction of the displacement are directly related to the stability of the photovoltaic flexible bracket.

[0029] The above-mentioned wind vibration performance evaluation method provided by the embodiment of the present invention avoids simplifying the photovoltaic flexible bracket by establishing a finite element model of the photovoltaic flexible bracket, ensures the accuracy of the modal parameters of the photovoltaic flexible bracket, obtains the wind direction range and wind speed range based on the modal parameters and the meteorological data of the working environment, and conducts a wind vibration simulation experiment based on the wind speed range and wind direction range, thereby avoiding simplifying the working environment of the photovoltaic flexible bracket. The combination of wind speed and wind direction to which the photovoltaic flexible bracket is subjected in the working environment can be simulated to ensure the authenticity of the wind vibration simulation experiment, and the wind vibration parameters of the photovoltaic flexible bracket under different wind speed and wind direction combinations are recorded. The wind vibration performance of the photovoltaic flexible bracket is evaluated based on the wind vibration parameters, thereby ensuring the accuracy of the evaluation of its wind vibration performance.

[0030] In one embodiment, this embodiment provides a specific implementation method for establishing a finite element model of a photovoltaic flexible bracket:

[0031] Establish a finite element model based on computer-aided design software, import the finite element model into computer-aided engineering software, and set the model parameters of the finite element model;

[0032] Exemplarily, in the above steps, a finite element model of a photovoltaic flexible bracket is established using computer-aided design (CAD) software (such as AutoCAD, SolidWorks, etc.), and the finite element model is imported into computer-aided engineering (CAE) software (such as ANSYS, Abaqus, etc.), and the integrity and accuracy of the finite element model are ensured during the import process to avoid loss or deformation of the finite element model data. The model parameters required for the finite element model are set based on the CAE software, and the finite element model is established and the model parameters of the finite element are set based on the joint use of CAD software and CAE software to ensure that a precise and accurate finite element model is obtained.

[0033] Determine the load type of the working environment based on the meteorological data and apply different types of loads to the finite element model;

[0034] Exemplarily, in the above steps, the load type and load value to be simulated are determined based on the meteorological data of the working environment of the photovoltaic flexible bracket, wherein the load types include wind load, temperature load, gravity load, etc. Specifically, the wind pressure distribution is calculated according to the wind speed and wind direction, and applied to the surface of the finite element model as a wind load; the corresponding temperature field is set as a temperature load in the finite element model according to the ambient temperature change; the gravity load is set based on gravity and applied to the finite element model; based on the determined load type and load value, different loads are combined and applied to the finite element model, so as to realize the simulation of the dynamic response of the photovoltaic flexible bracket in actual work while considering the influence of factors such as temperature, wind force, sunshine and so on in the working environment on the photovoltaic flexible bracket, and more accurately evaluate the performance of the finite element model of the photovoltaic flexible bracket in a complex environment.

[0035] In one embodiment, this embodiment provides a specific implementation method for establishing a finite element model based on computer-aided design software:

[0036] Under the premise of considering the geometric shape, size, connection method and relative position between components of the photovoltaic flexible bracket, a finite element model is established through computer-aided design software based on the initial files of the photovoltaic flexible bracket; wherein the initial files include: design drawings, technical specifications and material lists;

[0037] Exemplarily, in the above steps, necessary design data such as design drawings, technical specifications, and material lists of the photovoltaic flexible bracket are first collected as initial files, and three-dimensional modeling is performed based on the initial files through CAD software under the premise of accurately considering the geometric shape, size, connection method and relative positions of the components of the photovoltaic flexible bracket to establish a finite element model. In the process of establishing the finite element model, the key components of the photovoltaic flexible bracket (such as connectors, supporting structures, etc.) and possible detailed features (such as chamfers, fillets, etc.) need to be processed in detail to ensure the accuracy and completeness of the finite element model.

[0038] In one embodiment, this embodiment provides a specific implementation method for setting model parameters of a finite element model:

[0039] Set the unit type, material parameters and boundary conditions of the finite element model; the unit types include shell unit, beam unit and solid unit, the material parameters include elastic modulus, Poisson's ratio, density and yield strength, and the boundary conditions include fixed constraint, sliding constraint and free boundary;

[0040] Exemplarily, in the above steps, the unit type, material parameters and boundary conditions of the finite element model are set based on CAE software. Specifically, the unit type is set: according to the structural characteristics and material properties of the photovoltaic flexible bracket, a suitable unit type is selected. For flexible components, shell units or beam units can be used for simulation. For key parts such as connectors, solid units can be used for refined simulation; material parameters are set: according to the material type and properties used in the bracket, corresponding material parameters (including elastic modulus, Poisson's ratio, density, yield strength, etc.) are set. For multi-material composite photovoltaic flexible brackets, the material parameters of different materials need to be set separately; boundary conditions are set: according to the actual working environment and installation method of the bracket, appropriate boundary conditions are set. For fixedly installed photovoltaic flexible brackets, fixed constraints and other constraints can be set. For movable or adjustable photovoltaic flexible brackets, sliding constraints, free boundaries or corresponding constraints can be set to simulate their actual movement.

[0041] In one embodiment, this embodiment provides a specific implementation method for obtaining modal parameters of a finite element model:

[0042] Divide the finite element model into grids, select the solver, set the solution parameters, and perform nonlinear processing and multi-load coupling analysis based on the solver to obtain the modal parameters of the finite element model;

[0043] Exemplarily, in the above steps, first ensure that the geometry, size, material parameters, etc. of the finite element model imported into the CAE software are consistent with the design drawings, and ensure that the finite element model has no missing parts, wrong connections, or inconsistent material properties. Then, select the appropriate meshing method based on the characteristics and requirements of the finite element model. For the complex structure of the photovoltaic flexible bracket, it may be necessary to use adaptive meshing or local refinement meshing methods to ensure the analysis accuracy of key parts. After meshing the finite element model, it is necessary to check the mesh quality, such as the mesh distortion, aspect ratio, etc., to ensure that the mesh quality meets the analysis requirements;

[0044] After ensuring that the quality of the mesh based on the finite element model meets the analysis requirements, modal solution is required. Specifically, first select a suitable solver according to the characteristics of the finite element model and the analysis requirements. For large-scale or complex finite element models, parallel computing or high-performance computing technology can be used to improve the solution efficiency and set appropriate solution parameters, including solution accuracy, number of iterations, etc. If the photovoltaic flexible bracket has nonlinear characteristics, it is necessary to introduce nonlinear processing capabilities in the modal analysis process so that the finite element model can more accurately simulate the actual working state of the photovoltaic flexible bracket. In the process of solving the problem through the solver, the solution progress and convergence can be monitored to ensure the smooth progress of the solution, and the nonlinear stress-strain relationship of the material and the geometric nonlinearity need to be considered. Linear factors, appropriate iterative algorithms or incremental methods are used to deal with the nonlinear problems of photovoltaic flexible brackets to ensure the accuracy of the solution results. At the same time, considering that different loads are applied to the finite element model, it is necessary to consider the multi-load coupling effect and establish a multi-load coupling model, which includes determining the interaction relationship between different loads, defining coupling parameters, etc. By adopting multi-physical field coupling analysis technology or multi-load superposition method, the influence of multiple loads on the finite element model is considered to obtain more comprehensive modal analysis results. After performing modal analysis on the finite element model, the modal parameters of the finite element model are obtained from the solver and displayed through visualization technology, so as to more intuitively understand the modal characteristics of the photovoltaic flexible bracket, such as vibration mode and displacement distribution at different frequencies.

[0045] In one embodiment, this embodiment provides a specific implementation method for obtaining meteorological data of the working environment of the photovoltaic flexible support:

[0046] Collect meteorological data of the working environment, and perform pre-processing and statistical analysis on the meteorological data to obtain processed meteorological data; wherein the pre-processing includes cleaning, sorting and formatting;

[0047] Exemplarily, in the above steps, based on the working environment of the photovoltaic flexible bracket, appropriate monitoring equipment is selected, such as a weather station, anemometer, satellite remote sensing system, etc., and the frequency and duration of meteorological data collection are set based on the monitoring equipment to obtain sufficient meteorological data samples. At the same time, for different regions and seasons, the meteorological data collection strategy can be adjusted to adapt to different climatic conditions, thereby ensuring that these monitoring equipment can accurately measure and record key meteorological parameters such as wind speed, wind direction, temperature, and humidity. The meteorological data collected by the monitoring equipment is transmitted to the data center or analysis platform in real time using a reliable data transmission method (such as wireless transmission or wired transmission) to ensure the integrity and security of the meteorological data and avoid data loss or tampering;

[0048] The meteorological data transmitted to the data center or analysis platform will be cleaned, sorted and formatted to remove outliers and noise data to ensure the accuracy of subsequent analysis results based on the meteorological data. The pre-processed meteorological data will be statistically analyzed using statistical software or analysis tools to obtain statistical analysis results based on the meteorological data, including wind speed distribution, wind direction frequency, joint distribution of wind speed and wind direction, etc. The statistical analysis results will be used as important data for subsequently determining the wind speed range and wind direction range.

[0049] In one embodiment, this embodiment provides a specific implementation method for determining the wind speed range and wind direction range of the working environment based on meteorological data and modal parameters:

[0050] Based on the processed meteorological data combined with the wind speed profile model, the wind speed range of the working environment is determined, and the wind speed value corresponding to the natural frequency within the wind speed range is removed; based on the processed meteorological data combined with the wind direction stability analysis, the wind direction range of the working environment is determined, and the wind direction that causes the largest vibration mode change in the wind direction range is taken as the main wind direction;

[0051] Exemplarily, in the above steps, further analysis is performed based on the processed meteorological data, i.e., the statistical analysis results of the meteorological data, combined with wind engineering theory (such as wind speed profile model, wind direction stability analysis, and wind speed and wind direction joint distribution, etc.), so as to more accurately describe the wind condition characteristics of the working environment. Specifically, the maximum wind speed and minimum wind speed that the photovoltaic flexible bracket may encounter during normal use are determined based on the processed meteorological data (mainly based on wind speed distribution) combined with the wind speed profile model, and a certain safety margin is considered based on the maximum wind speed and the minimum wind speed to ensure the structural stability of the photovoltaic flexible bracket under extreme weather conditions. At the same time, since the wind speed of the working environment is within the resonant wind speed range of the photovoltaic flexible bracket, it is easy to cause the photovoltaic flexible bracket to respond violently, resulting in damage to the photovoltaic flexible bracket. Therefore, it is necessary to exclude the wind speed value corresponding to the natural frequency of the photovoltaic flexible bracket or the nearby range of these values ​​from the wind speed range of the working environment;

[0052] Based on the processed meteorological data (mainly based on wind direction frequency) combined with wind direction stability analysis, the wind direction range of the working environment is determined, and the wind direction within the wind direction range that makes the vibration mode of the finite element model show a large displacement or stress concentration is analyzed as the main wind direction. At the same time, for areas with multiple wind directions, a combination of multiple main wind directions can be considered.

[0053] In one embodiment, this embodiment provides a specific implementation method for obtaining the wind vibration performance of a photovoltaic flexible bracket based on wind vibration parameters:

[0054] Training the machine learning algorithm model based on the collected sample data on the basis of the wind vibration performance evaluation rules;

[0055] Exemplarily, two wind vibration performance evaluation rules are established in the above steps, including classification rules or regression rules. Under the classification rules, the wind vibration performance of the photovoltaic flexible bracket can be divided into different levels (such as excellent, good, medium, and poor); under the regression rules, a certain performance indicator of the photovoltaic flexible bracket (such as maximum displacement, maximum stress, etc.) can be directly predicted, and under the regression rules, according to the design requirements and industry standards of the photovoltaic flexible bracket, the maximum displacement does not exceed the preset displacement threshold, the maximum stress does not exceed the material strength, and other rules can be set; under the two rules, the wind vibration parameters of the photovoltaic flexible bracket are compared with the preset wind vibration parameters, and any abnormal wind vibration parameters or those that exceed the design standards are identified.

[0056] In order to evaluate the wind vibration performance of the photovoltaic flexible support, a machine learning algorithm (such as support vector machine, random forest, etc.) or a traditional statistical method (such as regression analysis, variance analysis, etc.) can be selected. In the embodiment of the present invention, a support vector machine (SVM) algorithm is selected as the machine learning algorithm. The SVM algorithm finds the optimal hyperplane by maximizing the interval. Compared with the traditional machine learning method (such as linear regression or decision tree), it can provide better generalization ability and facilitate the processing of the collected high-dimensional sample data and the nonlinear problems in the photovoltaic flexible support. The basic form of the optimization problem of the SVM algorithm can be expressed as:

[0057]

[0058] Among them, ω is the weight vector, b is the bias term, C is the regularization parameter, ξ i is the slack variable, φ(x i ) is the mapping function, y i is the label of the i-th sample; in the SVM algorithm, by adjusting the regularization parameter C, the complexity and fitting error of the SVM algorithm model can be balanced to avoid overfitting.

[0059] The SVM algorithm is trained based on the set wind vibration performance evaluation rules and the collected sample data, and the parameters of the SVM algorithm (such as kernel function, penalty coefficient, etc.) are adjusted during the training process to enable the model to achieve better performance on the training set. Specifically, in the example of the present invention, by adjusting the radial basis function kernel of the SVM algorithm, the SVM algorithm is allowed to find the optimal hyperplane in the high-dimensional space, which is conducive to processing the complex nonlinear wind vibration response problems encountered by the photovoltaic flexible bracket, provides a more flexible decision boundary, and improves the prediction accuracy of the SVM algorithm for wind vibration performance evaluation. The radial basis function kernel can be expressed as:

[0060]

[0061] Among them, γ is the kernel function parameter, which controls the influence range of a single training sample, ||x i-x j || is x i With x j The Euclidean distance between .

[0062] Acquire characteristic parameters of the photovoltaic flexible bracket, preprocess the collected wind vibration parameters, and input the preprocessed wind vibration parameters and characteristic parameters into the trained machine learning algorithm model to obtain the wind vibration performance of the photovoltaic flexible bracket; wherein the characteristic parameters include wind speed, wind direction, and bracket material;

[0063] Exemplarily, in the above steps, the collected wind vibration parameters are preprocessed, including noise removal, outlier processing, data smoothing and other operations. Subsequently, data processing and analysis techniques (such as FFT analysis, time domain and frequency domain analysis, etc.) may be used to extract key information of the wind vibration parameters to improve the quality of the wind vibration parameters and the accuracy of analysis. In the process of preprocessing the wind vibration parameters, the embodiment of the present invention uses a Gaussian filter for processing, and its formula is expressed as follows:

[0064]

[0065] Among them, x(t) is the original signal, y(t) is the filtered signal, σ is the standard deviation of the Gaussian distribution, and τ is the noise signal; Gaussian filter is used for preprocessing. Compared with simple moving average filtering, Gaussian filtering has better local characteristics in the time domain, can more effectively retain the high-frequency components in the wind vibration parameters, and effectively remove noise and outliers, improve the quality of wind vibration parameters, and make the wind vibration parameters smooth, which helps to more accurately capture the wind vibration response characteristics of photovoltaic flexible brackets;

[0066] The features that have a greater impact on the wind-vibration performance of the photovoltaic flexible bracket are selected as feature parameters (such as wind speed, wind direction, bracket material, etc.), and the wind-vibration performance of the photovoltaic flexible bracket is obtained based on the feature parameters and the preprocessed wind-vibration parameters input into the trained SVM algorithm model.

[0067] In one embodiment, this embodiment also provides a specific implementation method for obtaining the wind vibration performance of the photovoltaic flexible bracket based on the wind vibration parameters:

[0068] Input the verification data into the trained machine learning algorithm model for verification, and establish evaluation indicators based on the verification results to evaluate the performance of the machine learning algorithm model to obtain the model performance evaluation results; the evaluation indicators include accuracy, recall rate, and F1 score;

[0069] Exemplarily, in the above steps, the validation data is input into the trained machine learning algorithm model, i.e., the SVM algorithm model, for validation to evaluate the performance of the SVM algorithm model. Commonly used evaluation indicators include accuracy, recall, F1 score, etc. In the embodiment of the present invention, mean square error (MSE) and K-fold cross validation in the cross validation method are used as evaluation indicators, and the formula for the mean square error is:

[0070]

[0071] Where N is the total number of samples, y i is the true value, is the predicted value; the mean square error is used as the evaluation index to provide a quantitative measure of the wind vibration performance prediction error of the SVM algorithm model, which provides a more comprehensive wind vibration performance evaluation result than a simple correct rate or error rate;

[0072] The accuracy of K-fold cross validation can be expressed as:

[0073]

[0074] Among them, K is the fold number, n k is the number of samples in the kth fold, I is the indicator function, y i is the true value, is the predicted value, CV k It is the mean square error of K-fold cross validation. Using K-fold cross validation instead of simple data set segmentation method enables the SVM algorithm model to more comprehensively evaluate the generalization ability of the SVM algorithm model even when the amount of input data is limited.

[0075] According to the established wind vibration performance evaluation rules, the maximum displacement, maximum stress, wind vibration performance level, etc. of the photovoltaic flexible bracket under different wind speeds and wind directions are compared with the design requirements and industry standards, and the wind vibration performance based on the SVM algorithm model is output in the form of a visual chart or report to intuitively understand the wind vibration performance of the photovoltaic flexible bracket.

[0076] This embodiment also provides a wind vibration performance evaluation system for executing the above-mentioned wind vibration performance evaluation method.

[0077] In the embodiment of the present invention, before conducting a wind vibration simulation experiment, a fan or wind tunnel of appropriate power and model can be selected according to the wind speed range of the working environment, and a wind direction adjustment device with high precision and rapid response capability can be provided so as to quickly adjust the wind direction during the experiment. In addition, a wind speed control device is provided to ensure the stability and accuracy of the wind speed. The key components such as the fan or wind tunnel, the wind direction adjustment device and the wind speed control device are installed according to the design drawings to form a wind load simulation system. After the installation is completed, the wind load simulation system is debugged to ensure that the connection between the components is stable and can operate normally. According to the wind speed range and the wind direction range, different wind conditions are set by the wind speed control device and the wind direction adjustment device, including constant wind speed, variable wind speed, different wind directions, etc., to simulate various wind conditions that the photovoltaic flexible bracket may encounter in the working environment;

[0078] At the same time, the photovoltaic flexible bracket to be tested needs to be inspected and tested to ensure that its structure and performance meet the test requirements, and then the photovoltaic flexible bracket is installed on the experimental platform. During the installation process, the connection must be firm and reliable to avoid experimental failure or inaccurate data due to connection problems. After the installation is completed, it is also necessary to check to ensure that it will not accidentally fall off or be damaged during the experiment.

[0079] When conducting wind vibration simulation experiments, multiple combinations of wind speeds and wind directions are set through the wind load simulation system to simulate the wind vibration response of the photovoltaic flexible bracket under different wind conditions, and the sampling frequency and recording time of the high-speed response recording system are set. By setting a sufficiently high sampling frequency, the rapid response of the photovoltaic flexible bracket under wind load can be captured, and a sufficiently long recording time is set to obtain sufficient wind vibration parameters for analysis.

[0080] In an embodiment of the present invention, based on the verification results of the machine learning algorithm model, namely the SVM algorithm model, the machine learning algorithm model is optimized (such as adjusting parameters, adding features, changing the model structure, etc.) to improve the generalization ability and performance of the machine learning algorithm model.

[0081] In the embodiment of the present invention, the trained machine learning algorithm model, namely the SVM algorithm model, is applied to the wind vibration performance evaluation of the actual photovoltaic flexible bracket to ensure the real-time performance of the wind vibration performance evaluation and provide instant wind vibration performance evaluation feedback for engineering practice. The machine learning algorithm model, namely the SVM algorithm model, can be iteratively optimized to improve its performance and accuracy. Specifically, an online learning algorithm can be used for iterative optimization, and its update rule can be expressed as:

[0082]

[0083] Among them, ω t is the weight at time t, η(t) is the learning rate, is the loss function, The gradient of x with respect to the weight ω, t+1 and t+1 are the features and labels of new samples.

[0084] In an embodiment of the present invention, based on the data of the photovoltaic flexible bracket collected in advance, a finite element model of the photovoltaic flexible bracket is established through CAD software, ensuring the accuracy of the model and being able to reflect the actual structure of the photovoltaic flexible bracket, the finite element model is imported into the CAE software and the unit type, material parameters and boundary conditions are set, and an accurate finite element model is established in view of the diversity of materials of the photovoltaic flexible bracket, and in order to consider the influence of environmental factors such as temperature, wind force, and sunshine on the photovoltaic flexible bracket, a corresponding load model is established and applied to the finite element model, so as to truly simulate the wind vibration response of the photovoltaic flexible bracket, based on the introduction of nonlinear processing and multi-load coupling model in the process of modal analysis of the finite element model, so as to obtain more accurate and comprehensive modal parameters, significantly improve the calculation efficiency while ensuring the calculation accuracy, obtain the meteorological data of the working environment of the photovoltaic flexible bracket, and determine the wind speed range and wind direction range based on the meteorological data and modal parameters. , ensure the authenticity and accuracy of subsequent wind vibration simulation tests, obtain the wind vibration parameters of wind vibration simulation experiments, and first train the machine learning algorithm model based on the collected sample data on the basis of wind vibration performance evaluation rules, and preprocess the collected wind vibration parameters to ensure the smoothness of wind vibration parameter data and reduce the influence of noise on wind vibration performance evaluation results. Input the characteristic parameters and preprocessed wind vibration parameters into the trained machine learning algorithm model to obtain the wind vibration performance evaluation results. Based on the accuracy and authenticity when establishing the finite element model, as well as the accuracy and authenticity of the wind speed and wind direction in the working environment when conducting wind vibration simulation experiments, the wind vibration performance is evaluated by training the machine learning algorithm model first and then evaluating it, which ensures the accuracy of wind vibration performance evaluation of photovoltaic flexible brackets. In the process of using the machine learning algorithm model for wind vibration performance evaluation, the model can be verified and optimized and iteratively optimized, thereby further improving the accuracy of wind vibration performance evaluation.

[0085] In the description of the present invention, it should be noted that the terms "upper", "lower", "front", "horizontal", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on the present invention.

[0086] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the term "installation" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating wind vibration performance, characterized in that: The wind vibration performance evaluation method comprises the following steps: Establishing a finite element model of a photovoltaic flexible bracket and obtaining modal parameters of the finite element model; wherein the modal parameters include natural frequency and vibration mode; Acquire meteorological data of the working environment of the photovoltaic flexible support, and determine the wind speed range and wind direction range of the working environment based on the meteorological data and the modal parameters; In a wind vibration simulation experiment based on the wind speed range and the wind direction range, the wind vibration parameters of the photovoltaic flexible bracket under different wind directions and different wind speeds are recorded; The wind vibration performance of the photovoltaic flexible bracket is evaluated based on the wind vibration parameters; wherein the wind vibration parameters include stress, strain and displacement of the photovoltaic flexible bracket.

2. The wind vibration performance evaluation method according to claim 1, characterized in that: The step of establishing a finite element model of the photovoltaic flexible support comprises: Establishing the finite element model based on computer-aided design software, importing the finite element model into computer-aided engineering software, and setting model parameters of the finite element model; The load type of the working environment is determined based on the meteorological data and different types of loads are applied to the finite element model.

3. The wind vibration performance evaluation method according to claim 2, characterized in that: The step of establishing the finite element model based on computer-aided design software comprises: Under the premise of considering the geometric shape, size, connection method and relative positions between components of the photovoltaic flexible bracket, the finite element model is established through computer-aided design software based on the initial files of the photovoltaic flexible bracket; wherein the initial files include: design drawings, technical specifications and material lists.

4. The wind vibration performance evaluation method according to claim 2, characterized in that: The step of setting the model parameters of the finite element model comprises: The unit type, material parameters and boundary conditions of the finite element model are set; wherein the unit types include shell units, beam units and solid units, the material parameters include elastic modulus, Poisson's ratio, density and yield strength, and the boundary conditions include fixed constraints, sliding constraints and free boundaries.

5. The wind vibration performance evaluation method according to claim 1, characterized in that: The step of obtaining the modal parameters of the finite element model comprises: The finite element model is meshed, a solver is selected, solution parameters are set, and nonlinear processing and multi-load coupling analysis are performed based on the solver to obtain the modal parameters of the finite element model.

6. The wind vibration performance evaluation method according to claim 1, characterized in that: The step of obtaining meteorological data of the working environment of the photovoltaic flexible support includes: The meteorological data of the working environment is collected, and the meteorological data is pre-processed and statistically analyzed to obtain processed meteorological data; wherein the pre-processing includes cleaning, sorting and formatting.

7. The wind vibration performance evaluation method according to claim 6, characterized in that: The step of determining the wind speed range and wind direction range of the working environment based on the meteorological data and the modal parameters includes: Determine the wind speed range of the working environment based on the processed meteorological data in combination with a wind speed profile model, and remove the wind speed value corresponding to the natural frequency within the wind speed range; The wind direction range of the working environment is determined based on the processed meteorological data combined with wind direction stability analysis, and the wind direction in the wind direction range that causes the largest change in the vibration mode is taken as the main wind direction.

8. The wind vibration performance evaluation method according to claim 1, characterized in that: The step of obtaining the wind vibration performance of the photovoltaic flexible bracket based on the wind vibration parameters includes: Training the machine learning algorithm model based on the collected sample data on the basis of the wind vibration performance evaluation rules; Acquire characteristic parameters of the photovoltaic flexible bracket, preprocess the collected wind vibration parameters, and input the preprocessed wind vibration parameters and the characteristic parameters into the trained machine learning algorithm model to obtain the wind vibration performance of the photovoltaic flexible bracket; wherein the characteristic parameters include wind speed, wind direction, and bracket material.

9. The wind vibration performance evaluation method according to claim 8, characterized in that: Also includes: The verification data is input into the trained machine learning algorithm model for verification, and evaluation indicators are established based on the verification results to evaluate the performance of the machine learning algorithm model to obtain model performance evaluation results; wherein the evaluation indicators include accuracy, recall rate, and F1 score.

10. A wind vibration performance evaluation system, characterized in that: Used to execute the wind vibration performance evaluation method described in any one of claims 1-9.

Images