An optimized system and method for tracking photovoltaic mounting

Abstract

The present invention relates to the field of data processing technology, and more specifically, to an optimization system and method for a tracking photovoltaic support, the method comprising: calculating the noise performance of each light intensity data in the acquired light intensity data sequence, performing variational modal decomposition on the light intensity data sequence based on multiple decomposition levels, and constructing a scatter plot with the noise performance of each light intensity data as the horizontal coordinate and the correction amplitude of each light intensity data at any decomposition level as the vertical coordinate, and measuring the denoising effect of the decomposition level by using the average value of the distance between the data points in the scatter plot and the regression curve, and controlling the orientation of the photovoltaic panel based on the denoised data of the decomposition level with the best denoising effect. The present invention can accurately control the orientation of the photovoltaic panel, thereby improving the efficiency of the photovoltaic panel in collecting light energy.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology. More specifically, the present invention relates to an optimization system and method for a tracking photovoltaic support. Background Art

[0002] A tracking photovoltaic bracket is a photovoltaic system that can automatically track sunlight. The principle of this is that the photovoltaic panel is always kept perpendicular to the sun through the motor and control system installed on the bracket to maximize the absorption of solar energy. In order to improve the energy collection efficiency of the tracking photovoltaic bracket, it is necessary to collect the light intensity in real time and feed it back to the control system to adjust the direction of the photovoltaic panel in real time. Therefore, it is very important to control the direction of the photovoltaic bracket.

[0003] In the related technology, as disclosed in the patent application document with publication number CN116305358A, an adjustable bracket control model training method, device, equipment and storage medium are disclosed. The method includes: according to the light coverage area of ​​the photovoltaic panel, selecting from various preset control angles a control angle that makes the preset power generation parameters of the photovoltaic panel meet the preset conditions, as the selected control angle corresponding to the target environmental simulation parameter value; iteratively optimizing the adjustable bracket control model based on multiple training samples generated based on various environmental simulation parameter values ​​to obtain the target adjustable bracket control model.

[0004] However, the above scheme did not take into account the influence of noise data in the actual application of the photovoltaic bracket when training the adjustable bracket control model, so that the control angle determined by the target adjustable bracket control model trained according to the above scheme may have a certain deviation from the optimal angle, which may cause the sunlight to fail to fully cover the photovoltaic panel, thereby reducing the efficiency of the photovoltaic panel in collecting light energy to a certain extent. Summary of the invention

[0005] In order to solve the problem of low efficiency of photovoltaic panels in collecting light energy, the present invention proposes an optimization system and method for a tracking photovoltaic support.

[0006] According to a first aspect of an embodiment of the present invention, there is provided a method for optimizing a tracking photovoltaic support, comprising:

[0007] Obtain the reading of the light intensity sensor installed on the surface of the photovoltaic panel to obtain a light intensity data sequence, and calculate the noise performance of each light intensity data in the light intensity data sequence;

[0008] Based on the pre-acquired multiple decomposition levels, variational mode decomposition is performed on the light intensity data sequence to obtain denoised data at each decomposition level, and the difference before and after denoising any light intensity data at any decomposition level is calculated to obtain the correction amplitude of any light intensity data at any decomposition level;

[0009] The noise performance of each light intensity data is used as the horizontal axis, and the correction amplitude of each light intensity data at any decomposition level is used as the vertical axis. A scatter plot is constructed to determine the regression curve of the scatter plot. The average value of the distance between the data points in the scatter plot and the regression curve is used to measure the denoising effect of any decomposition level. The denoising effect is negatively correlated with the average value.

[0010] The denoising effect corresponding to each decomposition layer number is obtained, and the orientation of the photovoltaic panel is controlled based on the denoised data obtained based on the decomposition layer number with the best denoising effect, so as to optimize the control of the photovoltaic panel.

[0011] The present invention utilizes the correlation between the noise performance of each light intensity data and the correction amplitude under the decomposition number to measure the denoising effect of any decomposition number, which can ensure the accuracy of the denoising effect corresponding to each determined decomposition number, and based on the denoised data obtained based on the decomposition number with the best denoising effect, controls the orientation of the photovoltaic bracket so that the orientation of the photovoltaic panel can remain perpendicular to the light, thereby improving the efficiency of the photovoltaic panel in collecting light energy.

[0012] In an exemplary embodiment of the present invention, the average value of the distance between the data points in the scatter plot and the regression curve is used to measure the denoising effect of any decomposition layer, satisfying the following relationship:

[0013] ;

[0014] In the formula, For the Denoising effect of the number of decomposition layers; is the number of illumination intensity data in the illumination intensity data sequence; Based on The denoised data obtained by the decomposition layer number, the first The distance between a data point and the corresponding regression curve; for the summation symbol; The natural constant An exponential function with base .

[0015] In an exemplary embodiment of the present invention, a method for obtaining noise performance includes:

[0016] Obtaining the number of extreme value points within a preset range of any light intensity data, when the number of extreme value points is less than or equal to the preset value, the noise performance of any light intensity data is zero;

[0017] When the number of extreme value points is greater than a preset value, the noise performance of any light intensity data is positively correlated with the number and the average value of the difference between adjacent extreme value points within the preset range.

[0018] The present invention sets the values ​​of two noise representations according to the number of extreme value points within a preset range, and can accurately evaluate the possibility that each light intensity data is noise data.

[0019] In an exemplary embodiment of the present invention, when the number of extreme value points is greater than a preset value, the noise performance of any light intensity data satisfies the following relationship:

[0020] ;

[0021] In the formula, For the Noise performance of light intensity data; , Respectively The number of extreme value points within a preset range of the light intensity data, and the number of data points within the preset range; , Respectively The first light intensity data within the preset range The extreme points and extreme points; for the summation symbol; is the absolute value symbol; is the normalization function.

[0022] The present invention utilizes a normalization function to unify the dimensions of different data, so that the noise performance of the light intensity data can be determined by integrating a variety of data, thereby ensuring the accuracy of the determined noise performance.

[0023] In an exemplary embodiment of the present invention, a method for obtaining multiple decomposition levels includes:

[0024] Obtain a preset fixed number of superposition layers, and use the maximum kurtosis principle to determine the initial number of decomposition layers for variational modal decomposition of the light intensity data sequence;

[0025] The difference between the initial decomposition number of layers and the fixed number of superimposed layers, as well as the cumulative sum between the initial decomposition number of layers and the fixed number of superimposed layers are calculated, and the positive integer values ​​in the range of layers composed of the difference and the cumulative sum are taken as the values ​​of each decomposition number to obtain multiple decomposition numbers.

[0026] The present invention can ensure the accuracy of the determined initial decomposition number of layers, and when performing variational modal decomposition on the light intensity data sequence, the optimal decomposition number of layers is within the local range of the determined initial decomposition number of layers. Therefore, based on the initial decomposition number of layers and the preset fixed superposition number of layers, the range of the selected multiple decomposition number of layers is determined, which can ensure that the optimal decomposition number is included in each decomposition number of layers.

[0027] In an exemplary embodiment of the present invention, variational modal decomposition is performed on the illumination intensity data sequence based on the pre-acquired multiple decomposition levels to obtain denoised data at each decomposition level, including:

[0028] Based on any number of decomposition layers, variational modal decomposition is performed on the light intensity data sequence to obtain multiple intrinsic modal components, and the number of intrinsic modal components corresponds to any number of decomposition layers;

[0029] The wavelet threshold is used to denoise each intrinsic modal component, and all the intrinsic modal components after denoising are reconstructed to obtain the denoised data at any decomposition level, and then the denoised data at each decomposition level is determined.

[0030] In an exemplary embodiment of the present invention, a method for obtaining a regression curve of a scatter plot includes:

[0031] The least squares fitting is performed on each data point in the scatter plot to obtain a fitting curve composed of the predicted values ​​of each data point. The fitting curve is used to minimize the error between each data point and the corresponding predicted value to obtain the regression curve of the scatter plot.

[0032] The present invention utilizes the least square method to determine the regression curve of the scatter plot, the method is simple, and the accuracy of the determined regression curve is high.

[0033] According to a second aspect of an embodiment of the present invention, an optimization system for a tracking photovoltaic bracket is provided. The optimization system for a tracking photovoltaic bracket includes a memory and a processor. A computer program is stored in the memory. The processor executes the computer program to implement the steps of the first aspect of the embodiment of the present invention.

[0034] The present invention has the following effects:

[0035] The present invention can accurately evaluate the correlation between two variables corresponding to each data point in the scatter plot, that is, the correlation between the noise performance of each light intensity data and the correction amplitude of each light intensity data at the decomposition layer number corresponding to the scatter plot, thereby ensuring the accuracy of the denoising effect corresponding to each determined decomposition layer number; and based on the denoised data obtained based on the decomposition layer number with the best denoising effect, the orientation of the photovoltaic bracket is controlled so that the orientation of the photovoltaic panel can be kept perpendicular to the light, thereby improving the efficiency of the photovoltaic panel in collecting light energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0037] Figure 1 It is a schematic flow chart of the steps of an optimization method of a tracking photovoltaic support according to an embodiment of the present invention;

[0038] Figure 2 4 is a flowchart of the steps of determining the noise performance of each light intensity data in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0040] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0041] Reference Figure 1 , an optimization method for a tracking photovoltaic support, comprising steps S1 to S4, specifically as follows:

[0042] S1: Obtain the reading of the light intensity sensor installed on the surface of the photovoltaic panel, obtain a light intensity data sequence, and calculate the noise performance of each light intensity data in the light intensity data sequence.

[0043] Specifically, the sampling interval may be set to 1 second, and then the readings of the light intensity sensor installed on the surface of the photovoltaic panel are continuously collected to obtain a light intensity data sequence. This embodiment does not specifically limit the length of the light intensity data sequence.

[0044] Among them, noise performance refers to data used to measure the possibility that light intensity data is noise data. For example, the greater the noise performance of any light intensity data sequence, the greater the possibility that the light intensity data sequence is noise data. On the contrary, if the noise performance of any light intensity data sequence is smaller, the possibility that the light intensity data sequence is noise data is smaller.

[0045] In an exemplary embodiment of the present invention, reference Figure 2 As shown, the noise performance of each light intensity data can be determined by the following steps:

[0046] S101: Obtain the number of extreme value points within a preset range of any light intensity data. When the number is less than or equal to a preset value, the noise of any light intensity data is zero.

[0047] It should be noted that noise data will cause changes in the local characteristics of the data sequence, and in the process of collecting light intensity data, the light intensity sensor installed on the surface of the photovoltaic panel may be blocked by clouds, and the clouds move relatively quickly, which may cause some of the collected light intensity data sequences to contain drastically fluctuating noise data. Therefore, based on this feature, the data fluctuations in the local range of any light intensity data can be used to evaluate the possibility that the light intensity data is noise data.

[0048] The preset range of any light intensity data refers to the area around the light intensity data. data segment, and when the amount of data on either side of the light intensity data is less than , the data on that side is discarded and the light intensity data is compared with the data on the other side. data segment as the preset range of the light intensity data, where: is a rounding down function. In this embodiment, .

[0049] Optionally, the differential method can be used to identify extreme points within a preset range of the light intensity data, or the data points within the preset range can be fitted to determine the extreme points within the preset range. This embodiment does not specifically limit the method for determining the extreme points.

[0050] Optionally, when the number of extreme points within the preset range of any light intensity data is less than or equal to a preset value, such as 3, it means that the data change within the preset range of the light intensity data is relatively stable, and the possibility that the light intensity data is noise data is extremely small. Therefore, the noise performance of the light intensity data can be set to zero. Of course, a preset value of appropriate size can also be set according to the specific situation. This embodiment does not specifically limit the size of the preset value.

[0051] S102: When the number is greater than a preset value, the noise performance of any light intensity data is positively correlated with the number and the average value of the difference between adjacent extreme value points within a preset range.

[0052] Specifically, when the number of extreme value points within a preset range of any light intensity data is greater than a preset value, such as 3, the noise performance of the light intensity data satisfies the following relationship:

[0053] ;

[0054] In the formula, For the Noise performance of light intensity data; , Respectively The number of extreme value points within a preset range of the light intensity data, and the number of data points within the preset range; , Respectively The first light intensity data within the preset range The extreme points and extreme points; for the summation symbol; is the absolute value symbol; is the normalization function.

[0055] in, For the The difference between adjacent extreme points within the preset range of light intensity data, is the overall difference between adjacent extreme points in the range, reflecting the overall fluctuation intensity of the data in the range. The larger the value, the greater the fluctuation of the data in the range. The greater the possibility that the light intensity data is noise data.

[0056] In another embodiment, the noise performance of any light intensity data may also be calculated by other formulas, such as by a product formula instead of a summation formula.

[0057] S2: Based on the pre-acquired multiple decomposition levels, variational mode decomposition is performed on the light intensity data sequence to obtain denoised data at each decomposition level, and the difference before and after denoising of any light intensity data at any decomposition level is calculated to obtain the correction amplitude of any light intensity data at any decomposition level.

[0058] It should be noted that variational mode decomposition (VMD) is an algorithm suitable for nonlinear and non-stationary signal processing. It can decompose complex signals into multiple intrinsic modal components representing different frequencies, and then process different intrinsic modal components to achieve the purpose of denoising.

[0059] The number of decomposition levels refers to the number of intrinsic modal components obtained when performing variational modal decomposition on the light intensity data sequence. When the light intensity data sequence is subjected to variational mode decomposition, we can obtain An intrinsic modal component.

[0060] In an exemplary embodiment of the present invention, the determination of multiple decomposition levels can be achieved by the following steps:

[0061] Obtain a preset fixed number of stacking layers, and use the maximum kurtosis principle to determine the initial number of decomposition layers for variational modal decomposition of the light intensity data sequence; calculate the difference between the initial number of decomposition layers and the fixed number of stacking layers, as well as the cumulative sum between the initial number of decomposition layers and the fixed number of stacking layers, and take the positive integers in the range of layers composed of the difference and the cumulative sum as the values ​​of each decomposition layer, to obtain multiple decomposition layer numbers.

[0062] Optionally, the fixed number of stacking layers can be set to 2 based on empirical values, and when the initial number of decomposition layers determined by the maximum kurtosis principle is When , the difference between the initial decomposition number and the fixed stacking number is , the cumulative sum between the initial decomposition number of layers and the fixed superposition number of layers is , the corresponding composition layer number range is , , , , , thereby obtaining multiple decomposition levels. It should be noted that the process of determining the corresponding initial decomposition level using the maximum kurtosis principle is a prior art, and this embodiment will not be described in detail.

[0063] In an exemplary embodiment of the present invention, after determining multiple decomposition levels, the following steps may be performed to determine the denoised data at each decomposition level:

[0064] Based on any number of decomposition levels, variational modal decomposition is performed on the light intensity data sequence to obtain multiple intrinsic modal components, and the number of intrinsic modal components corresponds to any number of decomposition levels; each intrinsic modal component is denoised using a wavelet threshold, and all the intrinsic modal components after denoising are reconstructed to obtain denoised data at any number of decomposition levels, and then the denoised data at each decomposition level is determined.

[0065] For example, when based on the number of decomposition layers Perform variational mode decomposition on the light intensity data sequence and obtain After the intrinsic mode functions (IMF) with different frequencies are obtained, The connotation modal components with different frequencies are denoised by wavelet threshold respectively, so as to obtain the denoised The denoised modal components are then calculated The sum of the intrinsic modal components with different frequencies is used to reconstruct the intrinsic modal components after denoising, and then the denoised data of the light intensity data sequence at the decomposition level K can be obtained.

[0066] Optionally, in the process of reconstructing all the intrinsic modal components after denoising, it can be determined whether to use the decomposition layer number as the basis according to the specific situation. The residuals obtained by performing variational mode decomposition on the illumination intensity data sequence are summed. This embodiment does not specifically limit whether to sum the residuals when determining the denoised data.

[0067] It should be noted that the process of performing variational mode decomposition based on a determined number of decomposition levels and the process of denoising the obtained intrinsic mode components using a wavelet threshold are prior arts and will not be described in detail in this embodiment.

[0068] Furthermore, after performing variational mode decomposition on the illumination intensity data sequence based on the determined multiple decomposition levels and obtaining the denoised data at each decomposition level, the difference between any illumination intensity data before and after denoising at any decomposition level can be calculated to obtain the correction amplitude of any illumination intensity data at any decomposition level. For example, the illumination intensity data sequence can be recorded as { }, based on the number of decomposition levels The number of decomposition layers obtained when performing variational mode decomposition on the light intensity data sequence The denoised data under , then Light intensity data , in the decomposition level The difference before and after denoising , is the absolute value symbol, and the corresponding light intensity data is in the decomposition layer The correction range is , so that the difference before and after denoising of any light intensity data at any decomposition level can be determined, and the correction amplitude of any light intensity data at any decomposition level can be obtained.

[0069] S3: With the noise performance of each light intensity data as the horizontal axis and the correction amplitude of each light intensity data at any decomposition level as the vertical axis, a scatter plot is constructed, and the regression curve of the scatter plot is determined. The average value of the distance between the data points in the scatter plot and the regression curve is used to measure the denoising effect of any decomposition level. The denoising effect is negatively correlated with the average value.

[0070] In an exemplary embodiment of the present invention, a scatter plot refers to a graph used to reflect the relationship between each light intensity data and the correction amplitude of each light intensity data at any decomposition level. It should be noted that the number of constructed scatter plots is the same as the number of decomposition levels, that is, when the number of decomposition levels is , , , , When there are 5 in total, the number of scatter plots constructed is also 5, and the stronger the positive correlation between the data points in any scatter plot, the better the denoising effect under the corresponding number of decomposition layers.

[0071] The regression curve of the scatter plot refers to a curve that can reflect the relationship between two variables corresponding to each data point in the scatter plot. The regression curve of the scatter plot can be a linear regression curve or a nonlinear regression curve. This embodiment does not specifically limit the type of regression curve.

[0072] In an exemplary embodiment of the present invention, the determination of the regression curve of any scatter plot can be achieved by the following steps:

[0073] The least squares fitting is performed on each data point in the scatter plot to obtain a fitting curve composed of the predicted values ​​of each data point. The fitting curve is used to minimize the error between each data point and the corresponding predicted value to obtain the regression curve of the scatter plot.

[0074] Optionally, the regression curve of the scatter plot is determined by using the Ordinary Least Square (OLS) method, which is simple in design and has a low algorithm complexity, so that the regression curve corresponding to each scatter plot can be efficiently determined; of course, the regression curve of the scatter plot can also be determined by using the optimized least squares method, that is, the ridge regression algorithm. This embodiment does not specifically limit the method for determining the regression curve of the scatter plot. It should be noted that the process of determining the regression curve of the scatter plot using the least squares method is a prior art, and this embodiment will not be described in detail here.

[0075] Furthermore, the average value of the distance between the data points in the scatter plot and the regression curve is used to measure the denoising effect of any decomposition layer, satisfying the following relationship:

[0076] ;

[0077] In the formula, For the Denoising effect of the number of decomposition layers; is the number of illumination intensity data in the illumination intensity data sequence; Based on The denoised data obtained by the decomposition layer number, the first The distance between a data point and the corresponding regression curve; for the summation symbol; The natural constant An exponential function with base .

[0078] in, Reflects the The denoised data obtained by the decomposition layer number, the first The smaller the value is, the closer the data point is to the regression curve. The stronger the correlation between the noise performance of the data point and the correction amplitude of the data point at the decomposition level is, the better the denoising effect of the decomposition level is.

[0079] In another embodiment, the denoising effect of any decomposition level may also be determined by other calculation formulas. For example, the inverse of the average value of the distance between a data point in any scatter plot and the regression curve of the scatter plot may be used as the denoising effect of the decomposition level corresponding to the scatter plot.

[0080] S4: obtaining denoising effects corresponding to each decomposition layer number, and controlling the orientation of the photovoltaic panel based on the denoised data obtained from the decomposition layer number with the best denoising effect, so as to optimize the control of the photovoltaic panel.

[0081] It should be noted that if the number of decomposition layers is set to be large, then when performing variational mode decomposition on the light intensity data sequence, it may lead to overfitting of the light intensity data sequence. If the number of decomposition layers is set to be small, then when performing variational mode decomposition on the light intensity data sequence, it may not be able to effectively capture the complex structure and noise data of the light intensity data sequence, resulting in poor denoising effect. Therefore, it is necessary to set an appropriate number of decomposition layers.

[0082] Specifically, the denoising effect of any decomposition layer can be measured by using the average value of the distance between the data points in the scatter plot and the regression curve, and the denoising effect corresponding to each decomposition layer can be calculated respectively. Then, the denoised data obtained based on the decomposition layer with the best denoising effect can be used as the basic data for the control system of the tracking photovoltaic bracket to control the orientation of the photovoltaic panel, so that the control system can accurately select the direction with the highest energy output as the orientation of the photovoltaic panel, reducing the influence of the noise data on the control system when controlling the orientation of the photovoltaic panel, and can achieve accurate control of the photovoltaic panel, thereby optimizing the control of the photovoltaic panel.

[0083] The present invention also provides an optimization system for a tracking photovoltaic bracket. The system includes a memory and a processor, and a computer program is stored in the memory. The computer program integrates the function of an optimization method for a tracking photovoltaic bracket. When the computer program is executed, the orientation of the photovoltaic panel can be accurately controlled through an optimization method for a tracking photovoltaic bracket, thereby improving the efficiency of the photovoltaic panel in collecting light energy.

[0084] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0085] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for optimizing a tracking photovoltaic support, characterized in that: include: Obtaining a reading of a light intensity sensor installed on the surface of a photovoltaic panel to obtain a light intensity data sequence, and calculating the noise performance of each light intensity data in the light intensity data sequence, including: Obtaining the number of extreme value points within a preset range of any light intensity data, when the number of extreme value points is less than or equal to a preset value, the noise of any light intensity data is zero; When the number of extreme value points is greater than the preset value, the noise performance of any light intensity data satisfies the following relationship: ； In the formula, For the Noise performance of light intensity data; , Respectively The number of extreme value points within a preset range of the light intensity data, and the number of data points within the preset range; , Respectively The first light intensity data within the preset range The light intensity at the extreme point and the The light intensity at the extreme point; for the summation symbol; is the absolute value symbol; is the normalization function; Based on the pre-acquired multiple decomposition levels, the illumination intensity data sequence is subjected to variational mode decomposition to obtain denoised data at each decomposition level, including: Based on any number of decomposition levels, variational modal decomposition is performed on the illumination intensity data sequence to obtain a plurality of intrinsic modal components, wherein the number of the intrinsic modal components corresponds to the number of any decomposition levels; De-noising each of the intrinsic modal components using a wavelet threshold, and reconstructing all the intrinsic modal components after the de-noising to obtain the de-noised data at any of the decomposition levels, and then determining the de-noised data at each of the decomposition levels; Calculate the difference between any light intensity data before and after denoising at any decomposition level, and obtain the correction amplitude of any light intensity data at any decomposition level; The noise performance of each of the light intensity data is used as the horizontal coordinate, and the correction amplitude of each of the light intensity data at any decomposition level is used as the vertical coordinate, a scatter plot is constructed, a regression curve of the scatter plot is determined, and the denoising effect of any decomposition level is measured by using the average value of the distance between the data points in the scatter plot and the regression curve, and the denoising effect is negatively correlated with the average value; The denoising effect corresponding to each of the decomposition layers is obtained, and the orientation of the photovoltaic panel is controlled based on the denoised data obtained by the decomposition layer with the best denoising effect, so as to optimize the control of the photovoltaic panel.

2. The optimization method of a tracking photovoltaic support according to claim 1, characterized in that: The denoising effect of any decomposition layer is measured by using the average value of the distance between the data points in the scatter plot and the regression curve, satisfying the following relationship: ； In the formula, For the Denoising effect of the number of decomposition layers; is the number of illumination intensity data in the illumination intensity data sequence; Based on The denoised data obtained by the decomposition layer number, the first The distance between a data point and the corresponding regression curve; for the summation symbol; The natural constant An exponential function with base .

3. The optimization method of a tracking photovoltaic support according to claim 1, characterized in that: The method for obtaining the multiple decomposition levels includes: Obtaining a preset fixed number of stacking layers, and determining an initial number of decomposition layers for performing variational modal decomposition on the illumination intensity data sequence using the maximum kurtosis principle; Calculate the difference between the initial decomposition number of layers and the fixed number of superimposed layers, as well as the cumulative sum between the initial decomposition number of layers and the fixed number of superimposed layers, and take the positive integer values ​​in the layer number range composed of the difference and the cumulative sum as the values ​​of each decomposition number to obtain multiple decomposition numbers.

4. The optimization method of a tracking photovoltaic support according to claim 1, characterized in that: The method for obtaining the regression curve of the scatter plot comprises: A least squares fit is performed on each of the data points in the scatter plot to obtain a fitting curve consisting of the predicted values ​​of each of the data points, and the fitting curve is used to minimize the error between each of the data points and the corresponding predicted value to obtain a regression curve of the scatter plot.

5. An optimization system for a tracking photovoltaic support, characterized in that: The optimization system for a tracking photovoltaic support includes a memory and a processor, wherein a computer program is stored in the memory, and the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 4.

Images