Method for optimizing light reflectivity of graphene nano coating of photovoltaic panel

Abstract

The invention provides a method for optimizing the luminous reflectance of a graphene nano coating of a photovoltaic panel, which belongs to the technical field of photovoltaic panels and comprises the following steps: firstly, acquiring optical characteristic parameters of the surface of the photovoltaic panel by using various instruments such as a laser reflectometer and a scanning electron microscope; performing multivariate decomposition on the reflectivity data by adopting a Fourier transform method, and constructing an optical characteristic matrix by applying a least square method; 100 groups of coating structure parameters are generated through a K-means clustering algorithm and are subjected to clustering analysis, and 10 clustering centers are selected; a sample is prepared by adopting a chemical vapor deposition method, and experimental verification is carried out under the illumination intensity of 1000 watts per square meter. And finally, through 50-generation iterative optimization of a genetic algorithm, obtaining optimal structural parameters of the graphene nano-coating, and preparing the optimized photovoltaic panel surface coating by using a plasma enhanced chemical vapor deposition technology, so that the technical problem that the optical characteristics of the graphene nano-coating are difficult to accurately regulate and control to achieve a better reflection inhibition effect is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic panels, and more specifically, relates to a method for optimizing the light reflectivity of a graphene nano - coating on a photovoltaic panel. Background Art

[0002] Reflection loss on the surface of a photovoltaic cell is one of the important factors affecting its power generation efficiency. Usually, anti - reflection coating technology is adopted on the surface of crystalline silicon cells, such as inorganic film layers like silicon nitride or titanium oxide to reduce reflection loss. However, such inorganic film layers usually have a relatively large thickness, which will affect the light absorption of the cell, and at the same time, the preparation process is complex and the cost is relatively high.

[0003] In recent years, graphene, as a new two - dimensional carbon material, has received extensive attention due to its excellent optoelectronic properties. Existing research has shown that depositing a graphene thin film with a nanoscale thickness on the surface of a photovoltaic cell can not only effectively reduce reflection loss but also improve the photoelectric conversion efficiency. However, how to accurately regulate the optical properties of the graphene nano - coating to achieve a better reflection suppression effect is still an urgent technical problem to be solved. Summary of the Invention

[0004] In view of this, the present invention provides a method for optimizing the light reflectivity of a graphene nano - coating on a photovoltaic panel, which can solve the technical problem in the prior art that it is difficult to accurately regulate the optical properties of the graphene nano - coating to achieve a better reflection suppression effect.

[0005] The present invention is implemented as follows: In the first aspect of the present invention, a method for optimizing the light reflectivity of a graphene nano - coating on a photovoltaic panel is provided, which includes the following steps:

[0006] S10. Measure the surface characteristic parameters of the photovoltaic panel, including: collecting the light reflectivity measurement data of multiple test points on the surface of the photovoltaic panel at incident angles from 0 to 90 degrees by a laser reflectometer; measuring the graphene nano - coating thickness measurement data by a scanning electron microscope; measuring the graphene nano - coating refractive index measurement data by an ellipsometer; measuring the graphene nano - coating extinction coefficient measurement data by an ultraviolet spectrophotometer; measuring the incident light intensity measurement data by a laser power meter;

[0007] S20. Use the Fourier transform method to perform multivariate decomposition calculation on the light reflectivity measurement data to obtain the steady - state reflection component measurement data and the dynamic reflection component measurement data;

[0008] S30. Use the least - squares method to establish the mapping relationship between the graphene nano - coating thickness measurement data, the graphene nano - coating refractive index measurement data, the graphene nano - coating extinction coefficient measurement data and the steady - state reflection component measurement data, the dynamic reflection component measurement data, and construct an optical property matrix;

[0009] S40. Generate 100 sets of graphene nanocoating structure parameters according to the optical property equations, and calculate the light reflectivity calculation data corresponding to the 100 sets of graphene nanocoating structure parameters;

[0010] S50. Use the K-means clustering algorithm to perform clustering analysis on the 100 sets of graphene nanocoating structure parameters, and obtain the graphene nanocoating structure parameters corresponding to 10 cluster centers;

[0011] S60. Prepare the graphene nanocoating structures corresponding to the 10 cluster centers by chemical vapor deposition, conduct experimental verification under a light intensity of 1000 watts per square meter, and collect the light reflectivity verification data;

[0012] S70. Use the genetic algorithm to perform 50 generations of optimization iterations on the light reflectivity verification data to obtain the optimal graphene nanocoating structure parameters;

[0013] S80. Adopt plasma-enhanced chemical vapor deposition technology, apply the optimal graphene nanocoating structure parameters to the surface of the photovoltaic panel to prepare a graphene nanocoating, and obtain an optimized photovoltaic panel.

[0014] Further, the Fourier transform multi-decomposition calculation process in step S20 is specifically expressed as follows:

[0015]

[0016] In the formula, R(t) is the function of the light reflectivity changing with time; f k is the k-th frequency component, with the unit of Hz; a k , b k are the Fourier coefficients corresponding to the frequencies; N is the truncation term number of the Fourier series, and the value range is 10 - 100; R s is the steady-state reflection component; t is the sampling time, with the unit of s.

[0017] The Fourier coefficient calculation formula is: In the formula, T is the sampling period, with the unit of s.

[0018] The least squares mapping relationship in step S30 is specifically expressed as follows:

[0019]

[0020] In the formula, R si is the steady-state reflection component of the i-th test point; d i is the coating thickness of the i-th test point; n i is the refractive index of the i-th test point; k iis the extinction coefficient at the i-th test point; α, β, γ, δ are undetermined coefficients; ε i is the error term.

[0021] The Fresnel reflection equation is further expressed as follows:

[0022]

[0023] where r p , r s are the amplitude reflection coefficients of the p-wave and s-wave respectively; N 1 , n 2 are the refractive indices of the incident medium and the transmitted medium respectively; θ 1 is the angle of incidence, in degrees; σ is the interface roughness, in nm; λ is the wavelength of the incident light, in nm; α T is the temperature coefficient, ranging from 10 -6 to 10 -4 ; T is the actual temperature, in K; T 0 is the reference temperature, taken as 298 K; ε r is the reflection error term.

[0024] The interface phase delay equation is further expressed as follows:

[0025]

[0026] where δ is the phase delay; λ is the wavelength of the incident light, in nm; n is the refractive index of the coating; d is the thickness of the coating, in nm; θ is the angle of incidence, in degrees; β T is the thermo-optic coefficient, ranging from 10 -6 to 10 -4 ; T is the actual temperature, in K; T 0 is the reference temperature, taken as 298 K; γ P is the photoelastic coefficient, ranging from 10 -6 to 10 -4 ; P is the actual pressure, in Pa; P 0 is the standard atmospheric pressure, taken as 1.013×10 5 Pa; ω is the angular frequency of the light, in rad / s; c is the speed of light in vacuum; η j is the amplitude of the j-th harmonic component; L is the modulation period, in nm; x is the spatial position coordinate; M is the number of harmonics considered, with a value range of 3 to 10; ε δ is the phase error term.

[0027] The multi-layer film interference equation is further expressed as follows:

[0028]

[0029] Among them, each function is defined as:

[0030] F(θ) = 1 + f 1 sin 2 θ + f 2 sin 4 θ;

[0031] G(T) = 1 + g 1 (T - T 0 ) + g 2 (T - T 0 ) 2 ;

[0032]

[0033] In the formula, R is the total reflectivity; r 12 , r 23 , r 34 are the reflection coefficients of adjacent interfaces respectively; i is the imaginary unit; δ is the phase delay; f 1 , f 2 is the angle correction coefficient, with a range of -1 to 1; g 1 , g 2 is the temperature correction coefficient, with ranges of 10 -4 to 10 -2 and 10 -6 to 10 -4 ; h 1 , h 2 is the dispersion correction coefficient, with a range of -1 to 1; λ 0 is the central wavelength, in nm; ε R is the interference error term.

[0034] The absorption loss equation is further expressed as follows:

[0035]

[0036] In the formula, A is the absorption loss rate; I 0 is the incident light intensity, in W / m2; k is the extinction coefficient; d is the optical path length, in nm; λ is the incident light wavelength, in nm; μ T is the temperature absorption coefficient, with a range of 10 -4 to 10 -2 ; ΔT is the temperature change, in K; ν P is the pressure absorption coefficient, with a range of 10 -6 to 10 -4 Pa; ΔP is the pressure change, in Pa; ξ m is the influence coefficient of the m-th impurity; χ m is the concentration of the m-th impurity; N is the number of impurity types considered; ρl is the nonlinear absorption coefficient; I s is the saturation light intensity, with the unit of W / m2; L is the nonlinear order, and its value range is 2 to 4; ε A is the absorption error term.

[0037] The derivation process of each equation is explained in detail below:

[0038] 1. Derivation process of the Fourier transform equation:

[0039] (1) First, construct the time series R(t) based on the light reflectivity measurement data;

[0040] (2) According to the Fourier series expansion principle, decompose the periodic signal into the sum of trigonometric functions:

[0041]

[0042] (3) Considering the bandwidth limitation of the actual signal, intercept the first N terms:

[0043]

[0044] (4) Based on the orthogonality principle, derive the calculation formula for the Fourier coefficients:

[0045]

[0046] Explanation of parameter sources: f k The main frequency components are obtained by performing a fast Fourier transform on the measurement data; N is determined according to the requirements of the signal-to-noise ratio and calculation accuracy; T is determined according to the actual sampling time.

[0047] 2. Derivation process of the least squares mapping relationship:

[0048] (1) Establish a linear regression model: R si = α + βd i + γn i + δk i + ε i ;

[0049] (2) Write the n groups of measurement data in matrix form:

[0050]

[0051] Explanation of parameter sources: d i , n i , k i are obtained by measurement using a scanning electron microscope, an ellipsometer, and an ultraviolet spectrophotometer respectively; α, β, γ, δ are obtained by solving the normal equations using the least squares method.

[0052] 3. Optimization process of Fresnel reflection equation:

[0053] (1) Derive the basic Fresnel formula based on Snell's law:

[0054] (2) Substitute

[0055] (3) Considering the influence of interface roughness, introduce an exponential decay term

[0056] (4) Considering the temperature effect, add a quadratic term α T (T - T 0 ) 2 ;

[0057] Parameter source description: σ is obtained by measuring with an atomic force microscope; α T is calibrated through temperature - related reflectivity measurement experiments.

[0058] 4. Optimization process of interface phase delay equation:

[0059] (1) Basic phase delay formula:

[0060] (2) Considering the thermo - optic effect and the piezo - optic effect: [1 + β T (T - T 0 ) + γ P (P - P 0 )];

[0061] (3) Introduce the frequency dispersion effect:

[0062] (4) Considering the spatial modulation effect:

[0063] Parameter source description: β T , γ P are obtained through temperature - pressure control experiments; η j is obtained by Fourier - analyzing spatial modulation data.

[0064] 5. Optimization process of multi - layer film interference equation:

[0065] (1) Derive the basic interference formula based on the theory of electromagnetic wave propagation;

[0066] (2) Extend to three - layer interface reflection;

[0067] (3) Introduce correction functions F(θ), G(T), H(λ) to describe the angle, temperature, and wavelength dependencies;

[0068] Parameter source description: f 1 , f 2,g 1 ,g 2 ,h 1 ,h 2 Obtained by multi-variable non-linear fitting of experimental data.

[0069] 6. Optimization process of the absorption loss equation:

[0070] (1) Basic Lambert-Beer law: A = I 0 (1 - e -4πkd / λ );

[0071] (2) Considering temperature and pressure effects: (1 + μ T ΔT + v P ΔP);

[0072] (3) Introducing the influence of impurities:

[0073] (4) Considering non-linear absorption:

[0074] Description of parameter sources: μ T , v P Obtained through absorption experiments controlled by temperature and pressure; ξ m Obtained through impurity doping experiments; ρ l Obtained through intensity-dependent absorption measurements.

[0075] Compared with the prior art, a method for optimizing the light reflectivity of a graphene nano-coating on a photovoltaic panel provided by the present invention comprehensively measures the optical parameters on the surface of the photovoltaic panel, establishes an accurate optical property model, and adopts an advanced mathematical optimization algorithm to finally determine the optimal structural parameters of the graphene nano-coating, thereby effectively improving the light absorption performance of the photovoltaic panel. It has the following significant advantages compared with the prior art:

[0076] 1. Comprehensively measure optical parameters and establish an accurate optical property model. The method of the present invention uses advanced testing means such as a laser reflectometer, a scanning electron microscope, an ellipsometer, and an ultraviolet spectrophotometer to comprehensively measure and analyze key parameters such as the incident light intensity, light reflectivity, graphene coating thickness, refractive index, and extinction coefficient on the surface of the photovoltaic panel. On this basis, using mathematical methods such as Fourier analysis and least squares method, an accurate mapping relationship between the optical properties of the graphene nano-coating and the light reflectivity is constructed, laying a solid data foundation for the subsequent optimization of structural parameters.

[0077] 2. Implement precise parameter optimization using advanced mathematical optimization algorithms. The method of the present invention selects two mathematical optimization means, namely the K-means clustering algorithm and the genetic algorithm. First, through K-means clustering analysis, 10 representative optimal combinations are screened out from a large number of candidate structural parameters. Then, the genetic algorithm is used to iteratively optimize the light reflectivity data measured in experiments, and finally the structural parameters of the graphene nano-coating that can maximally suppress light reflection are determined. This method based on mathematical modeling and optimization greatly improves the accuracy and efficiency of parameter optimization.

[0078] 3. Achieve precise preparation of the graphene nano-coating on the surface of the photovoltaic panel. The method of the present invention finally applies the optimized structural parameters of the graphene nano-coating to the plasma-enhanced chemical vapor deposition process, and successfully prepares the optimized graphene nano-coating on the surface of the photovoltaic panel. Through strict control of process parameters and verification of structural parameters, it is ensured that the prepared photovoltaic panel can achieve the expected low-reflection characteristics.

[0079] In summary, the method for optimizing the light reflectivity of the graphene nano-coating on the photovoltaic panel proposed by the present invention makes full use of advanced optical measurement means and mathematical optimization algorithms, realizes precise regulation of the structural parameters of the graphene nano-coating, effectively reduces the reflection loss on the surface of the photovoltaic panel, and thus significantly improves the photovoltaic power generation efficiency; it solves the technical problem in the prior art that it is difficult to precisely regulate the optical properties of the graphene nano-coating to achieve a better reflection suppression effect. Brief Description of the Drawings

[0080] Figure 1 It is a flowchart of the method provided by the present invention. Detailed Embodiment

[0081] As Figure 1 shown, it is a flowchart of a method for optimizing the light reflectivity of the graphene nano-coating on a photovoltaic panel provided by the present invention. The specific implementation of each step will be described in detail below.

[0082] The specific implementation of step S10 is to comprehensively collect the optical characteristic parameters on the surface of the photovoltaic panel using various measurement means. First, a 10×10 test point grid array is set on the surface of the photovoltaic panel using a light power meter, and the incident light intensity is measured at each test point to obtain the incident light intensity measurement data. The purpose of this step is to provide the necessary input parameters for calculating the absorption loss of the photovoltaic panel in the subsequent process.

[0083] Secondly, a laser reflectometer is used to measure the light reflectivity at each test point in the 10×10 test point array at intervals of 5 degrees within the range of incident angles from 0 degrees to 90 degrees to obtain the light reflectivity measurement data. These data will be used to extract the steady-state reflection component and dynamic reflection component on the surface of the photovoltaic panel using Fourier analysis in the subsequent process.

[0084] Next, use a scanning electron microscope to measure the thickness of the graphene nanocoating for each test point in the 10 by 10 test point array, and obtain the graphene nanocoating thickness measurement data. This parameter is one of the key inputs for constructing the optical property equations.

[0085] At the same time, use an ellipsometer to measure the refractive index of the graphene nanocoating for each test point in the 10 by 10 test point array, and obtain the graphene nanocoating refractive index measurement data. This parameter is also a key input for constructing the optical property equations.

[0086] Finally, use an ultraviolet spectrophotometer to measure the extinction coefficient of the graphene nanocoating for each test point in the 10 by 10 test point array, and obtain the graphene nanocoating extinction coefficient measurement data. This parameter is also a key input for constructing the optical property equations.

[0087] Generally speaking, the purpose of step S10 is to comprehensively measure the optical parameters of the photovoltaic panel surface, including incident light intensity, light reflectivity, graphene coating thickness, refractive index, extinction coefficient, etc., providing a necessary data basis for subsequent optical modeling and optimization.

[0088] The specific implementation of step S20 is to perform multivariate decomposition calculation on the obtained light reflectivity measurement data using the Fourier transform method. First, construct the light reflectivity measurement data of each test point into time series data. Then, perform Fourier series expansion on these time series data, and set the number of truncation terms of the Fourier series to 50. Next, calculate the cosine coefficient and sine coefficient corresponding to each frequency component based on the orthogonality principle. Finally, substitute these cosine coefficients and sine coefficients into the Fourier series expansion formula to reconstruct the function of the light reflectivity changing with time. Two key parameters, namely the steady-state reflection component and the dynamic reflection component, can be extracted from this changing function.

[0089] The purpose of this step is to separate the steady-state and dynamic components of the light reflection on the photovoltaic panel surface through Fourier analysis. The steady-state component mainly comes from the reflection at the interface between the photovoltaic panel substrate and the coating, while the dynamic component mainly comes from the multiple interference reflections inside the coating. These two components play important roles in subsequent optical modeling.

[0090] The specific implementation of step S30 is to establish a mapping relationship between the thickness, refractive index, extinction coefficient of the graphene nanocoating and the steady-state reflection component using the least squares method. First, construct a linear regression model, with the steady-state reflection component as the dependent variable and the thickness, refractive index, and extinction coefficient of the graphene nanocoating as the independent variables. Then, use the least squares method to solve the normal equations and obtain the coefficients of the linear regression model. Finally, construct an optical property matrix based on these coefficients.

[0091] In this process, it is necessary to calculate the residuals at each test point and evaluate the goodness of fit of the model. Here, the criterion for judging the goodness of fit of the model is set as the sum of squared residuals being less than 0.01. The purpose of this step is to establish a quantitative mapping relationship between the optical parameters and the light reflectivity, providing a basis for the subsequent optimization of the structural parameters.

[0092] The specific implementation of step S40 is to generate 100 sets of different graphene nano-coating structural parameters according to the established optical property matrix and calculate the corresponding light reflectivities. First, using the Monte Carlo random sampling method, within the value range of the graphene nano-coating thickness from 10 to 100 nanometers, the refractive index from 1.2 to 3.5, and the extinction coefficient from 0.01 to 1.0, 100 sets of structural parameters are generated.

[0093] Then, these 100 sets of structural parameters are successively substituted into the Fresnel reflection equation to calculate the corresponding reflection coefficients for each set. Next, the reflection coefficients are substituted into the interface phase delay equation to calculate the phase delay amount. Then, the phase delay amount is substituted into the multi-layer film interference equation to calculate the total light reflectivity. Finally, the total reflectivity is substituted into the absorption loss equation to obtain the light reflectivity calculation data corresponding to each set of structural parameters.

[0094] The purpose of this step is to generate a large number of candidate graphene nano-coating structural parameters based on the established optical property relationship and calculate their corresponding light reflectivities, providing sufficient data support for the subsequent optimization of the structural parameters.

[0095] The specific implementation of step S50 is to perform clustering analysis on the generated 100 sets of graphene nano-coating structural parameters using the K-means clustering algorithm to obtain the optimal structural parameters corresponding to 10 cluster centers. First, calculate the Euclidean distances between every two of the 100 sets of structural parameters to obtain their similarities. Then, initialize 10 cluster centers and use the K-means algorithm for iterative optimization.

[0096] The conditions for terminating the iteration are set as: the change amount of the cluster center position is less than 0.001, or the number of iterations reaches 1000 times. Evaluate the clustering results using the silhouette coefficient to judge the goodness of the clustering effect. Finally, extract the graphene nano-coating structural parameters corresponding to the 10 cluster centers from the clustering results.

[0097] The purpose of this step is to screen out 10 representative and optimal parameter combinations from a large number of candidate structural parameters through clustering analysis, providing a targeted experimental plan for the subsequent experimental verification.

[0098] The specific implementation of step S60 is to use plasma-enhanced chemical vapor deposition technology to prepare graphene nano-coating samples corresponding to 10 clustering centers, and conduct a light reflectivity measurement experiment under standard lighting conditions. First, in the plasma-enhanced chemical vapor deposition system, set the flow ratio of reaction gases methane and hydrogen to 1:4, the reaction chamber temperature to 800 degrees Celsius, the pressure to 100 Pa, and the radio frequency power to 100 W. Then, according to the structural parameters corresponding to the 10 clustering centers, adjust the deposition process parameters to deposit graphene nano-coating samples on the surface of the photovoltaic panel.

[0099] Next, use a scanning electron microscope, ellipsometer, and ultraviolet spectrophotometer to measure the detailed structural parameters of these 10 samples to verify whether they meet the expectations. Finally, conduct a light reflectivity measurement experiment on these 10 samples under a standard light intensity of 1000 W / m² to obtain light reflectivity verification data.

[0100] The purpose of this step is to further confirm the feasibility of the 10 optimal structural parameters selected from a large number of candidate parameters in actual preparation through experimental verification, laying a foundation for subsequent parameter optimization. At the same time, the experimentally measured light reflectivity verification data will also be used for comparison with the calculated data to evaluate the optimization effect.

[0101] The specific implementation of step S70 is to use the genetic algorithm to perform 50 generations of optimization iterations on the obtained light reflectivity verification data, and finally determine the optimal structural parameters of the graphene nano-coating. First, construct the fitness function of the genetic algorithm, and use the root mean square error between the light reflectivity verification data and the calculated data as the fitness evaluation index.

[0102] Then, set the population size to 100, the chromosome coding length to 30 bits, the crossover probability to 0.8, and the mutation probability to 0.1. Perform 50 generations of optimization iterations. In each generation, select the 20 individuals with the highest fitness for crossover operations, and perform gene mutations on the crossed individuals according to the mutation probability. Finally, select the individual with the highest fitness from the last generation of the population as the finally determined optimal structural parameters of the graphene nano-coating.

[0103] The purpose of this step is to use the global optimization ability of the genetic algorithm to further optimize the structural parameters of the graphene nano-coating on the basis of experimental verification, so that its light reflectivity characteristics reach the best. By minimizing the error between the experimental data and the calculated data, it can be ensured that the obtained optimal parameters can truly reflect the optical characteristics of the surface of the photovoltaic panel.

[0104] The specific implementation of step S80 is to use plasma-enhanced chemical vapor deposition technology to apply the optimal graphene nanocoating structure parameters determined in step S70 to the surface of the photovoltaic panel, and prepare an optimized photovoltaic panel sample. First, in the plasma-enhanced chemical vapor deposition system, set the flow ratio of reaction gases methane and hydrogen to 1:4, the reaction chamber temperature to 800 degrees Celsius, the pressure to 100 Pa, and the radio frequency power to 100 W.

[0105] Then, according to the optimal graphene nanocoating structure parameters determined in step S70, adjust the deposition process parameters to deposit a graphene nanocoating on the surface of the photovoltaic panel. Finally, use a scanning electron microscope, ellipsometer, and ultraviolet spectrophotometer to measure the structure parameters of the deposited graphene nanocoating to ensure that they meet the requirements of the optimal parameters.

[0106] The purpose of this step is to apply the optimal graphene nanocoating structure parameters determined in all the previous steps to the actual surface of the photovoltaic panel and manufacture an optimized photovoltaic panel sample. Through strict control of process parameters and verification of structure parameters, ensure that the prepared photovoltaic panel can achieve the expected light reflectivity performance index.

[0107] In summary, this method for optimizing the light reflectivity of the graphene nanocoating on the photovoltaic panel comprehensively obtains the optical parameters of the photovoltaic panel surface through multiple measurement means, uses algorithms such as Fourier analysis, least squares method, and genetic algorithm for modeling and optimization, and finally determines the optimal graphene nanocoating structure parameters based on experimental verification and applies them to the manufacture of the photovoltaic panel. This systematic optimization method can effectively improve the light absorption performance of the photovoltaic panel and provide important technical support for improving the photovoltaic power generation efficiency.

[0108] Next, a specific embodiment 1 of the present invention is provided. The specific implementation of each step in this embodiment 1 is described in detail as follows: The specific implementation of step S10 is to comprehensively collect the optical characteristic parameters of the photovoltaic panel surface by using multiple measurement means. First, use a light power meter to set a 10×10 test point grid array on the surface of the photovoltaic panel, and measure the incident light intensity I 0 , and obtain the incident light intensity measurement data. The purpose of this step is to provide the necessary input parameters for subsequent calculation of the absorption loss of the photovoltaic panel.

[0109] Secondly, use a laser reflectometer to measure the light reflectivity E(θ) at each of the 10×10 test point arrays at intervals of 5 degrees in the range of incident angle θ from 0 degrees to 90 degrees, and obtain the light reflectivity measurement data. These data will be used to extract the steady-state reflection component and dynamic reflection component of the photovoltaic panel surface by using the Fourier analysis method in the future.

[0110] Next, use a scanning electron microscope to measure the thickness d of the graphene nano - coating for each test point in the 10 by 10 test - point array, and obtain the graphene nano - coating thickness measurement data. This parameter is one of the key inputs for constructing the optical property equations.

[0111] Meanwhile, use an ellipsometer to measure the refractive index n of the graphene nano - coating for each test point in the 10 by 10 test - point array, and obtain the graphene nano - coating refractive index measurement data. This parameter is also a key input for constructing the optical property equations.

[0112] Finally, use an ultraviolet spectrophotometer to measure the extinction coefficient k of the graphene nano - coating for each test point in the 10 by 10 test - point array, and obtain the graphene nano - coating extinction coefficient measurement data. This parameter is also a key input for constructing the optical property equations.

[0113] Generally speaking, the purpose of step S10 is to comprehensively measure the optical parameters on the surface of the photovoltaic panel, including the incident light intensity I 0 , the light reflectivity R(θ), the graphene coating thickness d, the refractive index n, and the extinction coefficient k, etc., providing a necessary data basis for subsequent optical modeling and optimization.

[0114] The specific implementation of step S20 is to perform a multivariate decomposition calculation on the obtained light reflectivity measurement data using the Fourier transform method. First, construct the light reflectivity measurement data R(t) of each test point into time - series data. Then, perform a Fourier series expansion on these time - series data:

[0115]

[0116] where f k is the k - th frequency component, with the unit of Hz; a k , b k are the Fourier coefficients corresponding to the frequencies; N is the truncation number of the Fourier series, with the value range of 10 - 100; R s is the steady - state reflection component; t is the sampling time, with the unit of s. The calculation formula for the Fourier coefficients is:

[0117]

[0118] where T is the sampling period, with the unit of s.

[0119] Next, calculate the cosine coefficients and sine coefficients corresponding to each frequency component based on the orthogonality principle. Finally, substitute these cosine coefficients and sine coefficients into the Fourier series expansion formula to reconstruct the function of the light reflectivity changing with time. The steady - state reflection component R s and the dynamic reflection component can be extracted from this changing function.

[0120] The purpose of this step is to separate the steady-state and dynamic components of the light reflection on the surface of the photovoltaic panel through Fourier analysis. The steady-state component mainly comes from the reflection at the interface between the photovoltaic panel substrate and the coating, while the dynamic component mainly comes from the multiple interference reflections inside the coating. These two components play important roles in subsequent optical modeling.

[0121] The specific implementation of step S30 is to establish the mapping relationship between the thickness d, refractive index n, extinction coefficient k of the graphene nano-coating and the steady-state reflection component R s using the least squares method. First, construct a linear regression model:

[0122]

[0123] where, R si is the steady-state reflection component at the i-th test point; d i , n i , k i are the coating thickness, refractive index and extinction coefficient at the i-th test point respectively; α, β, γ, δ are undetermined coefficients; ε i is the random error term.

[0124] Then, use the least squares method to solve the normal equations and obtain the coefficients of the linear regression model. Finally, construct an optical property matrix based on these coefficients. In this process, it is necessary to calculate the residual of each test point and evaluate the goodness of fit of the model. Here, the determination criterion for the goodness of fit of the model is that the sum of squared residuals is less than 0.01.

[0125] The purpose of this step is to establish a quantitative mapping relationship between the optical parameters and the light reflectivity, providing a basis for subsequent structural parameter optimization.

[0126] The specific implementation of step S40 is to generate 100 groups of different structural parameters of the graphene nano-coating according to the established optical property matrix and calculate the corresponding light reflectivity. First, use the Monte Carlo random sampling method to generate 100 groups of structural parameters within the value ranges of the graphene nano-coating thickness d from 10 to 100 nm, refractive index n from 1.2 to 3.5, and extinction coefficient k from 0.01 to 1.0.

[0127] Then, substitute these 100 groups of structural parameters into the Fresnel reflection equation to calculate the reflection coefficient in turn:

[0128]

[0129] where, r p , r s are the amplitude reflection coefficients of the p-wave and s-wave respectively; n 1 , n 2are the refractive indices of the incident medium and the transmission medium, respectively; θ 1 is the angle of incidence, in degrees; σ is the interface roughness, in nm; λ is the wavelength of the incident light, in nm; α T is the temperature coefficient, ranging from 10 -6 to 10 -4 K⁻²; T is the actual temperature, in K; T 0 is the reference temperature, taken as 298 K; ε r is the reflection error term.

[0130] Next, substitute the reflection coefficient into the interface phase delay equation to calculate the phase delay amount δ:

[0131]

[0132] In the formula, λ is the wavelength of the incident light, in nm; n is the refractive index of the coating; d is the thickness of the coating, in nm; θ is the angle of incidence, in degrees; β T is the thermo-optic coefficient, ranging from 10 -6 to 10 -4 K⁻¹; T is the actual temperature, in K; T 0 is the reference temperature, taken as 298 K; γ P is the photoelastic coefficient, ranging from 10 -6 to 10 -4 Pa⁻¹; P is the actual pressure, in Pa; P 0 is the standard atmospheric pressure, taken as 1.013×10 5 Pa; ω is the angular frequency of the light, in rad / s; c is the speed of light in vacuum; η j is the amplitude of the j-th harmonic component; L is the modulation period, in nm; x is the spatial position coordinate; M is the number of harmonics considered, with a value range of 3 - 10; ε δ is the phase error term.

[0133] Then substitute the phase delay amount into the multi-layer film interference equation to calculate the total light reflectivity R:

[0134]

[0135] The definitions of each function are as follows:

[0136] F(θ) = 1 + f 1 sin 2 θ + f 2 sin 4 θ;

[0137] G(T) = 1 + g 1 (T - T 0 ) + g 2 (T - T0 ) 2 ;

[0138]

[0139] In the formula, r 12 , r 23 , r 34 are the reflection coefficients of adjacent interfaces respectively; i is the imaginary unit; f 1 , f 2 is the angle correction coefficient, with a range from -1 to 1; g 1 , g 2 is the temperature correction coefficient, with ranges from 10 -4 to 10 -2 K{-1} and 10 -6 to 10 -4 K{-2}; h 1 , h 2 is the dispersion correction coefficient, with a range from -1 to 1; λ 0 is the central wavelength, with the unit of nm; ε R is the interference error term.

[0140] Finally, substitute the total reflectivity into the absorption loss equation to calculate the absorption loss rate A:

[0141]

[0142] In the formula, I 0 is the incident light intensity, with the unit of W / m 2 ; k is the extinction coefficient; d is the optical path length, with the unit of nm; μ T is the temperature absorption coefficient, with a range from 10 -4 to 10 -2 K{-1}; ΔT is the temperature change, with the unit of K; v P is the pressure absorption coefficient, with a range from 10 -6 to 10 -4 Pa{-1}; ΔP is the pressure change, with the unit of Pa; ξ m is the influence coefficient of the m-th impurity; X m is the concentration of the m-th impurity; N is the number of impurity types considered; ρ l is the nonlinear absorption coefficient; I s is the saturation light intensity, with the unit of W / m 2 ; L is the nonlinear order, with a value range of 2 - 4; ε A is the absorption error term.

[0143] In summary, the purpose of this step is to generate a large number of candidate graphene nanocoating structure parameters based on the established optical property relationships, calculate their corresponding light reflectivities, and provide sufficient data support for subsequent optimization of the structure parameters.

[0144] The specific implementation of step S50 is to use the K-means clustering algorithm to perform clustering analysis on the 100 groups of graphene nanocoating structure parameters generated, and obtain the optimal structure parameters corresponding to 10 cluster centers. First, calculate the Euclidean distance d between every two of the 100 groups of structure parameters ij , and obtain their similarity:

[0145]

[0146] Then, initialize 10 cluster centers μ k , and use the K-means algorithm for iterative optimization:

[0147]

[0148] The condition for terminating the iteration is set as: the change amount of the cluster center position or the number of iterations reaches 1000 times. Evaluate the clustering result with the silhouette coefficient s(i) to judge the goodness of the clustering effect:

[0149] In the formula, a(i) is the average distance between sample i and other samples in the same cluster, and b(i) is the average distance between sample i and the nearest cluster. The value range of the silhouette coefficient is from -1 to 1. Finally, extract the graphene nanocoating structure parameters corresponding to the 10 cluster centers from the clustering result.

[0150] The purpose of this step is to screen out 10 optimal parameter combinations with strong representativeness from a large number of candidate structure parameters through clustering analysis, and provide a targeted test plan for subsequent experimental verification.

[0151] The specific implementation of step S60 is to use plasma-enhanced chemical vapor deposition technology to prepare graphene nanocoating samples corresponding to the 10 cluster centers, and conduct light reflectivity measurement experiments under standard lighting conditions. First, in the plasma-enhanced chemical vapor deposition system, set the flow ratio of reaction gases methane and hydrogen to 1:4, the reaction chamber temperature to 800 degrees Celsius, the pressure to 100 Pa, and the radio frequency power to 100 W. Then, according to the structure parameters (d i , n i , k i ) corresponding to the 10 cluster centers, adjust the deposition process parameters, and deposit graphene nanocoating samples on the surface of the photovoltaic panel.

[0152] Next, a scanning electron microscope, an ellipsometer, and an ultraviolet spectrophotometer are used to measure the detailed structural parameters of these 10 samples to verify whether they meet the expectations. Finally, under the standard light intensity I of 1000 watts per square meter 0 an experiment on measuring the light reflectance R(θ) of these 10 samples is carried out to obtain the light reflectance verification data.

[0153] The purpose of this step is to further confirm the feasibility of the 10 optimal structural parameters selected from a large number of candidate parameters in actual preparation through experimental verification, laying a foundation for subsequent parameter optimization. At the same time, the experimentally measured light reflectance verification data will also be used for comparison with the calculated data to evaluate the optimization effect.

[0154] The specific implementation of step S70 is to use the genetic algorithm to perform 50 generations of optimization iterations on the obtained light reflectance verification data, and finally determine the optimal structural parameters of the graphene nano-coating. First, construct the fitness function f(x) of the genetic algorithm, and use the root mean square error ε between the light reflectance verification data R v (θ) and the calculated data R c (θ) as the fitness evaluation index:

[0155]

[0156] f(x) = -ε;

[0157] Then, set the population size to 100, the chromosome coding length to 30 bits, the crossover probability to 0.8, and the mutation probability to 0.1. Perform 50 generations of optimization iterations. In each generation, select the 20 individuals with the highest fitness for crossover operation, and perform gene mutation on the crossed individuals according to the mutation probability.

[0158] Finally, select the individual with the highest fitness from the last generation of the population as the finally determined optimal structural parameters of the graphene nano-coating. The purpose of this step is to use the global optimization ability of the genetic algorithm to further optimize the structural parameters of the graphene nano-coating on the basis of experimental verification, so that its light reflectance characteristics reach the best. By minimizing the error between the experimental data and the calculated data, it can be ensured that the obtained optimal parameters can truly reflect the optical characteristics of the photovoltaic panel surface.

[0159] The specific implementation of S80 has been described in detail above and will not be elaborated here.

[0160] To better understand and implement the present invention, the following provides Example 2 of a specific application scenario of the present invention: A photovoltaic power generation enterprise adopts the method for optimizing the light reflectance of the graphene nano-coating on the photovoltaic panel proposed by the present invention, and deposits the optimized graphene nano-coating on the surface of the photovoltaic cell.

[0161] First, the enterprise invited a research and development team consisting of optical experts, material technologists, and algorithm engineers to be responsible for the implementation of the specific plan. The research and development team first comprehensively measured the optical characteristic parameters on the surface of the photovoltaic panels. The specific steps are as follows:

[0162] 1. Measurement of incident light intensity. A 10×10 grid array of test points was set on the surface of the photovoltaic panel, and a power meter was used to measure the incident light intensity at each test point. The test results are shown in Table 1, and the average incident light intensity is 1020 watts per square meter.

[0163] Table 1 Measurement data of incident light intensity on the surface of the photovoltaic panel

[0164] Test Point Number Incident Light Intensity (Watt per Square Meter) 1-1 1023 1-2 1025 … … 10-10 1017

[0165] 2. Measurement of light reflectivity. A laser reflectometer was used to measure the light reflectivity at each point in the above 10×10 test point array at intervals of 5 degrees within the range of incident angles from 0 degrees to 90 degrees.

[0166] 3. Measurement of graphene coating thickness. A scanning electron microscope was used to measure the thickness of the graphene nano-coating at each point in the 10×10 test point array. The test results are shown in Table 2, and the average thickness is 65 nanometers.

[0167] Table 2 Measurement data of graphene nano-coating thickness

[0168] Test Point Number Coating Thickness (Nanometer) 1-1 63 1-2 66 … … 10-10 67

[0169] 4. Measurement of graphene coating refractive index. An ellipsometer was used to measure the refractive index of the graphene nano-coating at each point in the 10×10 test point array. The test results are shown in Table 3, and the average refractive index is 2.48.

[0170] Table 3 Measurement data of graphene nano-coating refractive index

[0171] Test Point Number Refractive Index 1-1 2.46 1-2 2.49 … … 10-10 2.50

[0172] 5. Measurement of graphene coating extinction coefficient. A UV spectrophotometer was used to measure the extinction coefficient of the graphene nano-coating at each point in the 10×10 test point array. The test results are shown in Table 4, and the average extinction coefficient is 0.18.

[0173] Table 4 Measurement data of graphene nano-coating extinction coefficient

[0174] Test Point Number Extinction Coefficient 1-1 0.17 1-2 0.18 … … 10-10 0.19

[0175] Through the above series of measurements, the research and development team obtained detailed optical parameter data on the surface of the photovoltaic panel, laying a foundation for subsequent optical modeling and optimization.

[0176] Next, the R & D team used Fourier analysis to perform multivariate decomposition on the obtained light reflectivity measurement data. The specific steps are as follows:

[0177] 1. Construct time - series data. Arrange the light reflectivity measurement data of each test point in the order of the incident angle to form a time series. Taking test point 1 - 1 as an example, its time - series data is as follows:

[0178] R(t) = [0.102, 0.108, 0.115, 0.123, 0.132, 0.142, 0.153, 0.164, 0.175, 0.185];

[0179] 2. Fourier series expansion. Perform Fourier series expansion on the above - mentioned time - series data, and set the truncation number N of the Fourier series to 50. Calculate the cosine coefficient a k and sine coefficient b k .

[0180] 3. Reconstruct the variation function. Substitute the obtained a k and b k into the Fourier series expansion formula to reconstruct the variation function R(t) of the light reflectivity with time.

[0181] 4. Extract components. Separate the steady - state reflection component R s and the dynamic reflection component from the variation function R(t).

[0182] Through the above steps, the R & D team successfully decomposed the original light reflectivity measurement data into two key components, namely the steady - state and dynamic components, laying a foundation for subsequent optical modeling.

[0183] Next, the R & D team used the least - squares method to establish the mapping relationship between the structural parameters of the graphene nano - coating and the light reflectivity. The specific steps are as follows:

[0184] 1. Construct a linear regression model. Use the steady - state reflection component R s as the dependent variable, and d, n, and k as the independent variables to establish a linear regression model:

[0185]

[0186] 2. Solve the normal equations. Use the least - squares method to solve the above - mentioned linear regression model to obtain the coefficients α, β, γ, δ.

[0187] 3. Construct the optical property matrix. According to the obtained coefficients, establish the mapping relationship between the optical properties of the graphene nano - coating and the light reflectivity to form the following optical property matrix:

[0188]

[0189] Through this step, the R & D team established an accurate optical property model, providing a reliable mathematical basis for subsequent parameter optimization.

[0190] Next, the R & D team adopted the Monte Carlo random sampling method to generate 100 different combinations of structural parameters within the value ranges of the graphene nanocoating thickness d from 10 to 100 nanometers, the refractive index n from 1.2 to 3.5, and the extinction coefficient k from 0.01 to 1.0. Then, these 100 groups of parameters were successively substituted into the Fresnel reflection equation, the interface phase delay equation, the multi-layer film interference equation, and the absorption loss equation to calculate the corresponding light reflectivity data. This process can be expressed by the following mathematical formula:

[0191]

[0192] Through the above calculations, the R & D team obtained the light reflectivity data corresponding to 100 groups of structural parameters, laying a data foundation for subsequent parameter optimization.

[0193] Next, the R & D team used the K-means clustering algorithm to classify these 100 groups of structural parameters and extract the optimal parameter combinations corresponding to 10 clustering centers. The specific steps are as follows:

[0194] 1. Calculate the Euclidean distance d between pairwise structural parameters ij :

[0195]

[0196] 2. Initialize 10 clustering centers μ k , and perform iterative optimization using the K-means algorithm:

[0197]

[0198] The iteration termination condition is that the change amount of the clustering center position is less than 0.001 or the number of iterations reaches 1000 times.

[0199] 3. Evaluate the clustering results with the silhouette coefficient s(i):

[0200]

[0201] The closer the silhouette coefficient is to 1, the better the clustering effect.

[0202] 4. Extract the optimal graphene nanocoating structural parameter combination from the 10 clustering centers.

[0203] After K-means clustering analysis, the R & D team finally determined 10 groups of optimal graphene nanocoating structural parameters, and the specific parameters are shown in Table 5.

[0204] Table 5 Structural Parameters of the Optimal Graphene Nanocoatings in 10 Groups

[0205] Parameter Combination Thickness (Nanometer) Refractive Index Extinction Coefficient 1 52 2.37 0.14 2 58 2.41 0.16 3 63 2.44 0.17 4 67 2.48 0.18 5 71 2.51 0.19 6 75 2.54 0.20 7 79 2.57 0.21 8 84 2.60 0.22 9 88 2.63 0.23 10 92 2.66 0.24

[0206] To verify the feasibility of these 10 groups of optimal parameters, the R & D team used the plasma-enhanced chemical vapor deposition process to separately prepare corresponding graphene nanocoating samples on the surface of photovoltaic panels. The deposition process parameters were set as follows: the flow ratio of reaction gases methane to hydrogen was 1:4, the reaction chamber temperature was 800 degrees Celsius, the pressure was 100 Pa, and the radio frequency power was 100 W.

[0207] After preparation, the R & D team conducted a detailed test on the structural parameters of these 10 samples and measured their light reflectance curves under a standard light intensity of 1000 W / m². Through comparative analysis, the R & D team found that these 10 groups of parameters could indeed better cover the common value ranges of the thickness, refractive index, and extinction coefficient of the graphene nanocoating, and the prepared samples also showed good performance in terms of light reflectance. However, for further optimization, the team decided to use the genetic algorithm to deeply analyze these experimental data.

[0208] First, the R & D team constructed the fitness function f(x) of the genetic algorithm, taking the root mean square error ε between the calculated data R c (θ) and the experimentally verified data R v (θ) as the optimization objective:

[0209]

[0210] Then, the population size was set to 100, the chromosome coding length was 30 bits, the crossover probability was 0.8, and the mutation probability was 0.1. 50 generations of optimization iterations were carried out. In each generation, the 20 individuals with the highest fitness were selected for crossover operations, and the crossed individuals were mutated according to the mutation probability; finally, the individual with the highest fitness was selected from the last generation of the population and determined as the optimal structural parameters of the graphene nanocoating. After optimization, the R & D team obtained the following optimal parameters: thickness 71 nm, refractive index 2.51, and extinction coefficient 0.19.

[0211] With the above optimized best structural parameters, the R & D team immediately used the plasma-enhanced chemical vapor deposition process to deposit the optimized graphene nanocoating on the surface of the photovoltaic panel. The specific process parameters were set as follows: the flow ratio of reaction gases methane to hydrogen was 1:4, the reaction chamber temperature was 800 degrees Celsius, the pressure was 100 Pa, and the radio frequency power was 100 W.

[0212] To ensure that the deposited graphene nano - coating meets the requirements of the best parameters, the R & D team used a scanning electron microscope, an ellipsometer, and an ultraviolet spectrophotometer to measure its detailed structural parameters. The test results showed that the thickness of the prepared graphene nano - coating was 72 nanometers, the refractive index was 2.52, and the extinction coefficient was 0.19, which basically matched the optimization target.

[0213] Finally, the R & D team tested the light reflectivity of the optimized photovoltaic panel under a standard light intensity of 1000 watts per square meter. Compared with before optimization, the light reflectivity decreased by about 15%, reaching the expected optimization target.

[0214] It should be noted that the detailed explanations of the variables involved in the description of the present invention are shown in Table 6 below.

[0215] Table 6 Explanation of Variables

[0216]

[0217]

[0218] The above - mentioned is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the light reflectivity of a graphene nanocoating of a photovoltaic panel, characterized in that: The following steps are involved: S10, measuring the surface characteristic parameters of the photovoltaic panel, including: collecting light reflectivity measurement data of multiple test points on the surface of the photovoltaic panel at incident angles of 0 to 90 degrees by a laser reflectometer; measuring the thickness measurement data of the graphene nanocoating by a scanning electron microscope; measuring the refractive index measurement data of the graphene nanocoating by an ellipsometer; measuring the extinction coefficient measurement data of the graphene nanocoating by an ultraviolet spectrophotometer; measuring the incident light intensity measurement data by a laser power meter; S20, performing multivariate decomposition calculation on the light reflectivity measurement data using a Fourier transform method to obtain steady-state reflection component measurement data and dynamic reflection component measurement data; S30, using the least square method to establish a mapping relationship between the graphene nano coating thickness measurement data, the graphene nano coating refractive index measurement data, the graphene nano coating extinction coefficient measurement data and the steady-state reflection component measurement data, and the dynamic reflection component measurement data, to construct an optical property matrix; S40, generating 100 groups of graphene nano-coating structure parameters according to the optical property equation group, and calculating light reflectivity calculation data corresponding to the 100 groups of graphene nano-coating structure parameters; S50, performing cluster analysis on the 100 groups of graphene nanocoating structural parameters using a K-means clustering algorithm to obtain graphene nanocoating structural parameters corresponding to 10 cluster centers; S60, preparing the graphene nano-coating structures corresponding to the 10 cluster centers by chemical vapor deposition, performing experimental verification under a light intensity of 1000 watts per square meter and collecting light reflectivity verification data; S70, using a genetic algorithm to perform 50 generations of optimization iterations on the light reflectivity verification data to obtain optimal graphene nanocoating structure parameters; S80, using plasma enhanced chemical vapor deposition technology, applying the optimal graphene nano-coating structure parameters to the surface of the photovoltaic panel to prepare the graphene nano-coating, and obtaining an optimized photovoltaic panel.

2. The method according to claim 1, characterized in that: The step S10 specifically includes: Step 11: using an optical power meter to set a 10×10 test point grid array on the surface of the photovoltaic panel, measuring the incident light intensity at each test point, and obtaining incident light intensity measurement data, wherein the unit of the incident light intensity measurement data is watts per square meter; Step 12, using a laser reflectometer, measuring the light reflectivity of each test point in the 10 times 10 test point array at an incident angle of 0 to 90 degrees at intervals of 5 degrees to obtain light reflectivity measurement data; Step 13, measuring the thickness of the graphene nano coating at each test point in the 10 times 10 test point array by a scanning electron microscope to obtain graphene nano coating thickness measurement data; Step 14: Using an ellipsometer, measure the refractive index of the graphene nanocoating at each test point in the 10 times 10 test point array to obtain the refractive index measurement data of the graphene nanocoating; Step 15: Use an ultraviolet spectrophotometer to measure the extinction coefficient of the graphene nanocoating at each test point in the 10 times 10 test point array to obtain graphene nanocoating extinction coefficient measurement data.

3. The method according to claim 1, characterized in that The step S20 specifically includes: Step 21, constructing a time series of the light reflectance measurement data to generate time series data for each test point; Step 22, performing Fourier series expansion on the time series data, and setting the number of Fourier series truncation terms to 50; Step 23: Calculate the cosine coefficient and sine coefficient corresponding to each frequency component based on the orthogonality principle; Step 24: Substitute the cosine coefficient and the sine coefficient into the Fourier series expansion to reconstruct a function of light reflectivity changing with time; Step 25: extracting steady-state reflection component measurement data and dynamic reflection component measurement data from the function of light reflectivity changing with time.

4. The method according to claim 1, characterized in that: The step S30 specifically includes: Step 31, constructing a linear regression model, taking the steady-state reflection component measurement data as the dependent variable; Step 32, using the graphene nano coating thickness measurement data, the graphene nano coating refractive index measurement data and the graphene nano coating extinction coefficient measurement data as independent variables; Step 33: Solve the normal equations using the least squares method to obtain the linear regression model coefficients; Step 34, constructing an optical characteristic matrix based on the linear regression model coefficients; Step 35: Calculate the residual of each test point and evaluate the goodness of fit of the model. The criterion for determining the goodness of fit of the model is that the sum of squares of the residuals is less than 0.

01.

5. The method according to claim 1, characterized in that The step S40 specifically includes: Step 41, according to the coefficients in the optical property matrix, based on the Monte Carlo random sampling method, 100 sets of structural parameters are generated within the value range of 10 to 100 nanometers for the graphene nano coating thickness, 1.2 to 3.5 for the refractive index, and 0.01 to 1.0 for the extinction coefficient; Step 42, substituting the 100 sets of structural parameters into the Fresnel reflection equation to calculate the reflection coefficient; Step 43, substituting the reflection coefficient into the interface phase delay equation to calculate the phase delay amount; Step 44, substituting the phase delay into the multilayer film interference equation to calculate the total reflectivity; Step 45: Substitute the total reflectivity into the absorption loss equation to calculate the absorption loss, and obtain the light reflectivity calculation data corresponding to each set of structural parameters.

6. The method according to claim 1, characterized in that The step S50 specifically includes: Step 51, calculating the similarity between the 100 groups of structural parameters according to the Euclidean distance; Step 52: Initialize 10 cluster centers and use K-means clustering algorithm for iterative optimization; Step 53, setting the iteration termination condition as the change in the cluster center position is less than 0.001 or the number of iterations reaches 1000; Step 54, evaluating the silhouette coefficient of the clustering result, wherein the value range of the silhouette coefficient is from -1 to 1; Step 55: extract the graphene nanocoating structural parameters corresponding to the 10 cluster centers.

7. The method according to claim 1, characterized in that The step S60 specifically includes: Step 61: using a plasma enhanced chemical vapor deposition system, preparing graphene nano-coating samples corresponding to 10 cluster centers at a temperature of 800 degrees Celsius and a pressure of 100 Pa; Step 62, measuring the structural parameters of 10 graphene nanocoating samples respectively by scanning electron microscopy, ellipsometer and ultraviolet spectrophotometer; Step 63, performing a reflectivity measurement experiment at a light intensity of 1000 watts per square meter; Step 64, collecting light reflectivity verification data of each graphene nano-coating sample at incident angles of 0 to 90 degrees; Step 65: Calculate the root mean square error between the light reflectance verification data and the light reflectance calculation data.

8. The method according to claim 1, characterized in that The step S70 specifically includes: Step 71, constructing a fitness function of the genetic algorithm, and taking the root mean square error between the light reflectance verification data and the light reflectance calculation data as a fitness evaluation index; Step 72, set the population size to 100, the chromosome code length to 30 bits, the crossover probability to 0.8, and the mutation probability to 0.1; Step 73, perform 50 generations of optimization iterations, and select the 20 individuals with the highest fitness in each generation for crossover operation; Step 74, performing gene mutation on the individuals after the crossover according to the mutation probability; Step 75: Select the individual with the highest fitness from the last generation population as the optimal graphene nanocoating structure parameters.

9. The method according to claim 1, characterized in that: The step S80 specifically includes: Step 81, setting the flow ratio of reaction gas methane to hydrogen to 1:4 in the plasma enhanced chemical vapor deposition system; Step 82, setting the reaction chamber temperature to 800 degrees Celsius, the pressure to 100 Pa, and the radio frequency power to 100 watts; Step 83, adjusting deposition process parameters according to the optimal graphene nanocoating structure parameters; Step 84, depositing a graphene nano coating on the surface of the photovoltaic panel; Step 85: Verify whether the structural parameters of the graphene nanocoating after deposition meet the requirements of the optimal structural parameters by measuring with a scanning electron microscope, an ellipsometer and an ultraviolet spectrophotometer.

10. The method according to claim 1, characterized in that The optical characteristic equations include: Fresnel reflection equation: input the incident angle, the graphene nano-coating refractive index measurement data, the photovoltaic panel substrate material refractive index measurement data, and output the interface reflection coefficient measurement data; Interface phase delay equation: input the graphene nano coating thickness measurement data, the graphene nano coating refractive index measurement data, incident light wavelength measurement data, and output phase delay measurement data; Multilayer film interference equation: input the interface reflection coefficient measurement data, the phase delay measurement data, the graphene nano coating layer number measurement data, and output the total reflectivity measurement data; Absorption loss equation: input the graphene nano coating extinction coefficient measurement data, the graphene nano coating optical path length measurement data, and the incident light intensity measurement data, and output the absorption loss rate measurement data.

Images