A method for extracting model parameters of a photovoltaic cell and related products

Abstract

The present application relates to the field of photovoltaic technology, and particularly relates to a photovoltaic cell model parameter extraction method and related products, the method comprising: establishing a photovoltaic cell mathematical model, and determining model parameters to be optimized in the mathematical model; constructing a fitness function of the photovoltaic cell mathematical model; constructing a parameter optimization model based on a whale optimization algorithm and a differential evolution operator, and taking the fitness function as an objective function; executing the parameter optimization model, and obtaining an optimal parameter set corresponding to the model parameters to be optimized in the mathematical model; obtaining a final photovoltaic cell mathematical model through the optimal parameter set, and verifying errors of the final photovoltaic cell mathematical model; the present application firstly constructs a mathematical model of a photovoltaic cell, and then determines parameters to be optimized in the mathematical model, and then defines a fitness function based on multiple error indicators and physical parameter constraints, and then forms a parameter optimization model suitable for multimodal optimization by combining a whale optimization algorithm and a differential evolution operator, and finally obtains an optimal parameter set of the photovoltaic cell model.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic technology, and in particular to a photovoltaic cell model parameter extraction method and related products. Background Art

[0002] As a renewable energy source with broad development prospects, solar energy has garnered widespread attention. Photovoltaic power generation is the primary application of solar energy and has entered a period of rapid development. The output characteristics of photovoltaic cells exhibit nonlinear characteristics, making accurate modeling of their output power crucial. Accurate photovoltaic cell models facilitate key tasks such as maximum power point tracking, fault detection, and performance evaluation, significantly improving the overall performance of photovoltaic power generation systems. However, current photovoltaic cell models typically exist in the form of implicit functions, containing multiple unknown parameters to be identified and potentially containing multiple local extreme points. Consequently, traditional analytical methods struggle to directly solve these complex models.

[0003] Over the past few decades, the main methods for identifying photovoltaic cell model parameters can be divided into two categories: mathematical analytical methods and metaheuristic algorithms. Mathematical analytical methods typically rely on the accuracy of specific data points and appropriate initial values. However, in practical applications, data accuracy is often difficult to guarantee, and the choice of initial values ​​can significantly influence the results, making it difficult to find a global optimal solution in complex multimodal problems. In contrast, metaheuristic algorithms do not rely on specific data points and are not restricted by initial values. They are more suitable for handling multimodal optimization problems in photovoltaic cell models and have therefore been widely used in complex parameter extraction tasks.

[0004] In recent years, a variety of metaheuristic algorithms, including the Symbiotic Organism Search (SOS) algorithm, the Harris Optimizer (HHO), the Salp Ascidian Algorithm (SSA), and the Supply and Demand Optimization (SDO) algorithm, have been introduced into the parameter extraction of photovoltaic cell models to solve various engineering optimization problems. These algorithms have enhanced the accuracy and robustness of solutions to a certain extent through randomization strategies and group coordination mechanisms. However, in applications, it has been found that the accuracy and robustness of these algorithms in photovoltaic cell parameter extraction tasks have not yet reached the ideal level. The complexity of photovoltaic cell models places higher demands on the computational accuracy and efficiency of the algorithms. Therefore, exploring new, more accurate and efficient algorithms and incorporating adaptive improvement measures to improve parameter extraction accuracy has become an important direction in current photovoltaic cell modeling research. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that it is difficult to accurately extract photovoltaic cell model parameters under complex multi-peak conditions. The purpose is to provide a photovoltaic cell model parameter extraction method and related products, which can improve the accuracy and robustness of photovoltaic cell model parameter extraction and enable the model to maintain high accuracy under different working conditions.

[0006] The present invention is achieved through the following technical solutions:

[0007] A photovoltaic cell model parameter extraction method, comprising:

[0008] Establish a mathematical model of photovoltaic cells and determine the model parameters to be optimized in the mathematical model;

[0009] Constructing the fitness function of the mathematical model of photovoltaic cells;

[0010] Construct a parameter optimization model based on the whale optimization algorithm and differential evolution operator, and use the fitness function as the objective function;

[0011] Execute the parameter optimization model and obtain the optimal parameter set corresponding to the model parameters to be optimized in the mathematical model;

[0012] The final photovoltaic cell mathematical model is obtained through the optimal parameter set, and the error is verified. If the error index is greater than the set value, the optimization parameters in the parameter optimization model are adjusted and the optimal parameter set is obtained again.

[0013] Specifically, the method for establishing a mathematical model of a photovoltaic cell includes:

[0014] Determine the reverse saturation current I0, Among them, I 0ref is the reverse saturation current at the reference temperature, E g is the band gap energy of the semiconductor, T is the absolute temperature, T ref is the reference temperature, n is the ideality factor of the diode, N s is the number of batteries connected in series, is the thermovoltage, k is the Boltzmann constant, and q is the electron charge;

[0015] Determine the diode current I D , Where I0 is the reverse saturation current of the diode, R s is the series resistance, I is the output current of the photovoltaic cell;

[0016] Determine the parallel resistor current I sh , Among them, R sh is the parallel resistance;

[0017] Determine the photocurrent I ph , Among them, I ph0 is the photocurrent at reference temperature and light intensity, α I is the temperature coefficient of the photocurrent, G is the actual light intensity, G ref is the reference light intensity;

[0018] Establishing the mathematical model of photovoltaic cells I=I ph -I D -I sh ;

[0019] The model parameters to be optimized include: photocurrent I ph , diode reverse saturation current I0, reference photocurrent I ph0 , Reference reverse saturation current I 0ref , the temperature coefficient of the photocurrent α I , diode ideality factor n, series resistance R s , parallel resistance R sh .

[0020] Specifically, the method for constructing the fitness function includes:

[0021] Determine physical parameter constraints And determine the penalty function for solutions that violate physical constraints Among them, θ is the parameter model to be optimized, λ is the penalty coefficient, and D is the dimension of the parameter. and is the maximum and minimum value of the jth parameter, θ j is the current value of the jth parameter;

[0022] Determining the Bayesian error term Where N is the number of experimental data points, I model (V i ,θ) is the voltage V i and the calculated current under the parameter θ, I exp (V i ) is the voltage V i The actual current under i is the standard deviation of the ith measurement value;

[0023] Determine the comprehensive fitness function f(θ)=α·(w1·MSE+w2·MAE+w3·RMSE+w4·RE-w5·R 2 +w6·BayesErr, where θ is the parameter model to be optimized, MSE is the mean square error, MAE is the mean absolute error, RMSE is the root mean square error, RE is the relative error, R 2 is the determination coefficient, w1, w2, w3, w4, w5, w6 are weight factors, γ is the regularization coefficient, and α is the adjustment coefficient.

[0024] Specifically, the parameter optimization algorithm includes:

[0025] Initialize the population: Use the whale optimization algorithm as the basic algorithm and initialize it; set the population size to N, randomly generate N candidate solutions in the parameter search space, and form the initial population x = {x1, x2, ..., x N}, where x i =[x i1 , x i2 ,...,x iD ], D is the parameter dimension;

[0026] Fitness calculation: Calculate each candidate solution x according to the fitness function f(·) of the optimization algorithm i The fitness value f(x i );

[0027] Iterative update: set the maximum number of iterations T max , for t = 1 to T max Each iteration of performs the following steps:

[0028] Update control parameters t is the current iteration number; generate random vector r1, random vector r2 and random number p;

[0029] Update candidate solution x i Get the position

[0030] Perform mutation operation v on the updated candidate solution i =x r1 +F·(x r2 -x r3 ) and crossover operations Among them, x r1 、x r2 、x r3 are individuals randomly selected from the population and are different from each other, and x r1 ≠x i , F is the variation factor, u i，d is the dth dimension of the i-th trial vector, C R is the crossover probability, d rand To randomly select an integer in {1, 2, ..., D}, x i，d is the dth dimension of the i-th individual, μ is the chaotic mapping parameter;

[0031] Select the candidate solutions after mutation and crossover operations, Among them, u i is the experimental solution generated after mutation and crossover operations, is the updated position;

[0032] Calculate each The fitness value of And update the global optimal solution

[0033] Specifically, update the candidate solution x i The location methods include:

[0034] Calculate the coefficient vector A = 2ar1-a, the coefficient vector C = 2r2;

[0035] If p<0.5 and |A|<1, then perform the encirclement of prey:

[0036] If p<0.5 and |A|≥1, then perform the encirclement of prey:

[0037] If p≥0.5, perform a spiral update of the position:

[0038] Among them, x best is the current global optimal solution, is the updated position, D1 is the distance calculated in the operation of surrounding the prey, and x rand is an individual randomly selected from the population, D2 is the random distance calculated in the encirclement prey operation, D3 is the distance calculated in the spiral update position, b is a constant defining the spiral shape, and l is a random number.

[0039] Specifically, the variability factor Crossover probability Among them, F min and F max are the minimum and maximum values ​​of the variation factor, and are the minimum and maximum crossover probabilities.

[0040] Specifically, the error verification method includes:

[0041] The optimal parameter set Bring in the mathematical model of photovoltaic cells;

[0042] According to the mathematical model of photovoltaic cells Calculate the voltage V i Calculated current I model (V i );

[0043] Determine the voltage and actual current data obtained from the experimental measurement {(V i , I exp (V i))}, where i = 1, 2, 3, ..., M, where M is the number of data points;

[0044] Calculate the mean square error AMSE, mean absolute error AMAE, and determination coefficient AR by calculating the current and the actual current 2 and mean relative error AMRE;

[0045] Set the corresponding mean square error threshold ∈MSE, mean absolute error threshold eMAE, and determination coefficient threshold eR 2 and mean relative error threshold ∈ MRE;

[0046] Constructing a composite error function

[0047] in, is the weight coefficient of each error term;

[0048] If F total If ≤1, the error verification passes; otherwise, the error verification fails, and the optimization parameters in the parameter optimization model are adjusted.

[0049] A photovoltaic cell model parameter extraction method terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the photovoltaic cell model parameter extraction method as described above is implemented.

[0050] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for extracting photovoltaic cell model parameters as described above is implemented.

[0051] A computer program product includes a computer program / instruction, which implements the photovoltaic cell model parameter extraction method described above when the computer program / instruction is executed by a processor.

[0052] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0053] The present invention first constructs a mathematical model of the photovoltaic cell and identifies the parameters to be optimized. Then, a fitness function is defined based on multiple error indicators and physical parameter constraints. Finally, by combining the whale optimization algorithm with the differential evolution operator, a parameter optimization model adapted to multimodal optimization is formed. Finally, the optimal parameter set of the photovoltaic cell model is obtained.

[0054] This paper significantly improves the accuracy and robustness of photovoltaic cell model parameter extraction by combining an improved whale optimization algorithm with a differential evolution operator. This approach is independent of initial value selection and specific data points, effectively avoiding local extreme value traps. Furthermore, the method incorporates targeted error verification and parameter constraint mechanisms, resulting in a final photovoltaic cell model with excellent adaptability and reliability, capable of adapting to dynamic changes in various operating environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings illustrate exemplary embodiments of the present invention and, together with the description thereof, are used to explain the principles of the present invention. These drawings are included to provide a further understanding of the present invention, and the accompanying drawings are included in and constitute a part of this specification and do not constitute a limitation of the embodiments of the present invention.

[0056] Figure 1 It is a flow chart of a photovoltaic cell model parameter extraction method according to the present invention. DETAILED DESCRIPTION

[0057] To make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the relevant content and are not intended to limit the present invention.

[0058] It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings.

[0059] In the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0060] Example 1

[0061] like Figure 1 As shown, a photovoltaic cell model parameter extraction method includes:

[0062] Establish a mathematical model for photovoltaic cells and determine the parameters to be optimized. Mathematical models for photovoltaic cells are typically used to describe their current-voltage (IV) characteristics, which vary with conditions such as light intensity and temperature. Parameters to be optimized typically include series resistance, shunt resistance, diode reverse saturation current, and photocurrent, all of which have a decisive influence on model accuracy.

[0063] Construct a fitness function for the PV cell mathematical model; this function serves as the objective function during the optimization process, evaluating the performance of different parameter combinations. Fitness functions are typically based on various error metrics, such as mean squared error (MSE) and mean absolute error (MAE), and incorporate physical constraints to ensure that the optimization results adhere to the physical characteristics of the PV cell. By quantifying the magnitude of the error between different parameter sets, this function helps the algorithm identify the parameter combination that best matches actual performance.

[0064] A parameter optimization model based on the whale optimization algorithm and differential evolution operator is constructed, with the fitness function as the objective function. The whale optimization algorithm is an intelligent optimization algorithm that mimics the hunting behavior of whales, primarily searching the parameter space by circling, encircling, and spiraling toward prey. The differential evolution operator is a population-based genetic algorithm operation used to increase global search capabilities and further optimize parameters through mutation and crossover operations on candidate solutions. This embodiment combines the rapid convergence of the whale optimization algorithm with the global search characteristics of the differential evolution algorithm, making the parameter optimization model more robust and accurate for complex multimodal problems.

[0065] Execute the parameter optimization model and obtain the optimal parameter set corresponding to the model parameters to be optimized in the mathematical model; after executing the optimization process, through iterative optimization and fitness evaluation, the algorithm will output an optimal parameter set, that is, the best parameter configuration of the photovoltaic cell model under the current conditions.

[0066] The final PV cell mathematical model is obtained from the optimal parameter set and then subjected to error verification. If the error index exceeds the set value, the optimization parameters in the parameter optimization model are adjusted to obtain the optimal parameter set again. Error verification is a critical step in ensuring model accuracy. The effectiveness of the model is measured by verifying the error between the actual measured values ​​and the model's predicted values. If the error exceeds a preset threshold, the parameters in the optimization model are readjusted and the optimization process is repeated until the error is within the allowable range.

[0067] Example 2

[0068] Based on the physical characteristics of photovoltaic cells, commonly used models include single-diode model, double-diode model, and triple-diode model. To strike a balance between computational complexity and model accuracy, we choose the improved single-diode model to establish the mathematical model of photovoltaic cells. The methods include:

[0069] Determine the reverse saturation current I0, Among them, I 0ref is the reverse saturation current at the reference temperature, E g is the band gap energy of the semiconductor, T is the absolute temperature, T ref is the reference temperature, n is the ideality factor of the diode (dimensionless), Ns is the number of batteries connected in series, is the thermovoltage, k is the Boltzmann constant, 1.380649×10 -23 J / K, q is the electron charge, 1.602176634×10 -19 C; Reverse saturation current is used to describe the small amount of leakage current of the diode when it is reverse biased, which directly affects the dark current characteristics of the photovoltaic cell.

[0070] Determine the current in the photovoltaic cell's internal diode - diode current I D , Where I0 is the reverse saturation current of the diode, R s is the series resistance, I is the output current of the photovoltaic cell; the diode current describes the nonlinear output behavior of the photovoltaic cell under forward bias and is the core part of the photovoltaic cell characteristic modeling.

[0071] Determine the parallel resistor current I sh , Among them, R sh The shunt resistance current is the leakage current of the photovoltaic cell caused by the shunt resistance. The smaller the shunt resistance, the greater the leakage current, resulting in reduced model accuracy. Therefore, this term is crucial for describing the leakage phenomenon and efficiency loss of photovoltaic cells.

[0072] Determine the current generated by the photovoltaic cell after being illuminated by light - the photocurrent I ph , Among them, I ph0 is the photocurrent at reference temperature and light intensity, α I is the temperature coefficient of the photocurrent, G is the actual light intensity, G ref is the reference light intensity; photocurrent represents the effect of light on the battery output current.

[0073] Establishing the mathematical model of photovoltaic cells I=I ph -I D -I sh ; Describes the output characteristics of photovoltaic cells under different voltage and light conditions.

[0074] The model parameters to be optimized include: photocurrent I ph , diode reverse saturation current I0, reference photocurrent I ph0 , Reference reverse saturation current I 0ref , the temperature coefficient of the photocurrent α I , diode ideality factor n, series resistance R s , parallel resistance R sh .

[0075] Example 3

[0076] A fitness function is constructed to measure the difference between the output of the photovoltaic cell model and experimental data. This fitness function serves as the objective function of the optimization algorithm, guiding the algorithm in its search for optimal model parameters. Due to the nonlinearity and complexity of the photovoltaic cell model, a fitness function that accurately reflects the model error is required. Furthermore, to improve algorithm performance, the complexity of the fitness function can be increased by introducing multi-objective optimization, constraints, and penalty terms.

[0077] The methods for constructing the fitness function include:

[0078] The parameters of the photovoltaic cell model have physical meaning and need to meet certain physical constraints. Determine the physical parameter constraints And determine the penalty function for solutions that violate physical constraints Among them, θ is the parameter model to be optimized, λ is the penalty coefficient, and D is the dimension of the parameter. and is the maximum and minimum value of the jth parameter, θ j is the current value of the jth parameter; the penalty function penalizes solutions that violate physical constraints and increases the fitness value, thereby avoiding unreasonable solutions during the optimization process.

[0079] The Bayesian error term is used to evaluate the fit between the model prediction value and the experimental data, and to determine the Bayesian error term Where N is the number of experimental data points, I model (V i ,θ) is the voltage V i and the calculated current under the parameter θ, I exp (V i ) is the voltage V i The actual current under i is the standard deviation of the ith measurement.

[0080] Integrate various error indicators, optimize model fitting and stability, and determine the comprehensive fitness function Among them, θ is the parameter model to be optimized, MSE is the mean square error, MAE is the mean absolute error, RMSE is the root mean square error, RE is the relative error, R 2 is the determination coefficient, w1, w2, w3, w4, w5, w6 are weight factors, γ is the regularization coefficient, and α is the adjustment coefficient.

[0081] Example 4

[0082] Parameter optimization algorithms include:

[0083] Initialize the population: Use the whale optimization algorithm as the basic algorithm and initialize it; set the population size to N, and randomly generate N candidate solutions in the parameter search space. Each candidate solution represents a set of parameters of a photovoltaic cell model. Form the initial population X = {x1, x2, ..., x N}, where x i =[x i1 , x i2 ,...,x iD ], D is the parameter dimension, which includes the model parameters described above, such as photocurrent, reverse saturation current, series resistance, etc.

[0084] Fitness calculation: Calculate each candidate solution x according to the fitness function f(·) of the optimization algorithm i The fitness value f(x i ); The fitness value measures the quality of the current parameter combination. The smaller the fitness value, the better the candidate solution is in terms of accuracy and robustness in fitting the photovoltaic cell model. The fitness function combines multiple error indicators such as physical constraints and Bayesian error, and evaluates the fitness value f(x i ) to guide the selection and update of the algorithm.

[0085] Iterative update: set the maximum number of iterations T max , for t = 1 to T max Each iteration of performs the following steps:

[0086] Update control parameters t is the current iteration number; random vectors r1 and r2 are generated, along with a random number T. As the iteration number t increases, the value of r1 decreases gradually, helping to gradually shift from global search to local search, improving algorithm convergence. The random vector controls the direction and magnitude of candidate solution updates, while the random number p, a random number between 0 and 1, determines how the candidate solutions are updated.

[0087] Update candidate solution x i Get the position According to the current control parameters and random vectors, the positions of candidate solutions are updated and new solutions are generated to obtain better parameter combinations. The specific method is described later.

[0088] In order to enhance the search diversity, the updated candidate solution is mutated. i =x r1 +,·(x r2 -x r3 ), which introduces random perturbations, allowing candidate solutions to have a wider range of exploration in the parameter space, helping to prevent them from falling into local minima.

[0089] Crossover Operation The crossover operation generates new solutions by combining the mutation results with the original solution, thereby improving the global search capability.

[0090] Mutation Factor The mutation factor changes from the initial F min As the number of iterations gradually increases to F max , increasing the mutation intensity as the optimization process progresses. This strategy allows for smaller mutation amplitudes in the early stages of the algorithm to maintain a larger search space; in the later stages, the mutation amplitude is gradually increased to accelerate convergence to the optimal solution and improve local search capabilities.

[0091] Crossover probability As the number of iterations increases, the crossover probability changes from Gradually decrease to In the early stages of the algorithm, a larger crossover probability increases the diversity of solutions and promotes global search; while in the later stages, a smaller crossover probability reduces the diversity of solutions, thereby focusing more on local optimization.

[0092] F min and F max are the minimum and maximum values ​​of the variation factor, and are the minimum and maximum crossover probabilities.

[0093] x r1 、x r2 、x r3 are individuals randomly selected from the population and are different from each other, and x r1 ≠x i , F is the variation factor, controlling the variation range; u i，d is the dth dimension of the i-th trial vector, C R is the crossover probability, which determines the fusion ratio of the mutation value and the original value; d rand To randomly select an integer in {1, 2, ..., D}, x i，d is the dth dimension of the i-th individual, μ is the chaotic mapping parameter;

[0094] Select the candidate solutions after mutation and crossover operations, that is, select the solution with smaller fitness as the updated position, Among them, u i is the experimental solution generated after mutation and crossover operations, is the updated position;

[0095] Calculate each The fitness value of And update the global optimal solution The global optimal solution represents the best set of parameters found in the current iteration.

[0096] This embodiment also gives the updated candidate solution x i The location methods include:

[0097] Calculate the coefficient vector A = 2ar1-a and the coefficient vector C = 2r2. a is a control parameter that decreases with the number of iterations. By dynamically adjusting the size of A, the algorithm can switch between global and local search. The coefficient vector C controls the randomness and variation of the candidate solutions relative to the target solution.

[0098] If p<0.5 and |A|<1, it means that the candidate solution is close to the prey (i.e. the global optimal solution). The candidate solution is made to approach the optimal solution to improve the convergence accuracy, that is, to surround the prey:

[0099] If p < 0.5 and |A| ≥ 1, it means that the candidate solution is far away from the optimal solution, and the algorithm randomly selects a solution x from the population. rand For reference, to simulate the whale's surrounding behavior when it is far away from its prey, execute the following to surround the prey:

[0100] If p ≥ 0.5, a spiral update is performed to simulate the behavior of a whale gradually approaching its prey along a spiral path:

[0101] Among them, x best is the current global optimal solution, is the updated position, D1 is the distance calculated in the operation of surrounding the prey, and x rand is an individual randomly selected from the population, D2 is the random distance calculated in the encirclement prey operation, D3 is the distance calculated in the spiral update position, b is a constant defining the spiral shape, and l is a random number.

[0102] Example 5

[0103] Error verification is performed to evaluate the accuracy of the optimal parameter set and the model's fitting effect. It uses a variety of error indicators to determine whether the model's performance meets the preset standards. The methods include:

[0104] The optimal parameter set Bring in the mathematical model of photovoltaic cells;

[0105] According to the mathematical model of photovoltaic cells Calculate the voltage V i Calculated current I model (V i );

[0106] Determine the voltage and actual current data obtained from the experimental measurement {(V i , I exp (V i))}, where i = 1, 2, 3, ..., M, where M is the number of data points; used for error comparison with the model calculation results.

[0107] Calculate the mean square error AMSE, mean absolute error AMAE, and determination coefficient AR by calculating the current and the actual current 2 and mean relative error AMRE;

[0108] Set the corresponding mean square error threshold ∈MSE, mean absolute error threshold ∈MAE, and determination coefficient threshold ∈R 2 and mean relative error threshold ∈ MRE;

[0109] Constructing a composite error function in, is the weight coefficient of each error term;

[0110] If F total If ≤1, the error verification passes; otherwise, the error verification fails, and the optimization parameters in the parameter optimization model are adjusted.

[0111] Strategies for adjusting parameters to optimize the model include increasing the population size, increasing the maximum number of iterations to increase the mutation factor, and adjusting the crossover probability.

[0112] Example 6

[0113] A photovoltaic cell model parameter extraction method terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the photovoltaic cell model parameter extraction method as described above is implemented.

[0114] The memory can be used to store software programs and modules. The processor executes the software programs and modules stored in the memory to perform various terminal functions and data processing. The memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system and the executable program required for at least one function.

[0115] The data storage area can store data created based on the use of the terminal, etc. In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0116] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for extracting photovoltaic cell model parameters as described above is implemented.

[0117] Without loss of generality, computer-readable media may include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer-readable instruction data structures, program modules, or other data. Computer storage media includes RAM, ROM, EPROM, EEPROM, flash memory or other solid-state storage technologies, CD-ROM, DVD or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that computer storage media is not limited to the aforementioned types. The aforementioned system memory and mass storage devices may be collectively referred to as memory.

[0118] A computer program product includes a computer program / instruction, which implements the photovoltaic cell model parameter extraction method described above when the computer program / instruction is executed by a processor.

[0119] A computer program product consists of a computer program or set of instructions for performing specific tasks or implementing specific functions. These programs or instructions are designed to be executed by a processor, thereby completing a series of predefined steps or operations. The program product may be stored in various forms of computer storage media, such as memory, hard disks, solid-state drives, optical disks, or other digital storage devices. It may exist as compiled binary code or as a script or bytecode executable by an interpreter. Through carefully designed algorithms and logical instructions, the program product enables the processor to process data in a specific order and manner, completing various functions such as data analysis, user interaction, and device control.

[0120] In the description of this specification, the description with reference to the terms "one embodiment / method", "some embodiments / methods", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment / method or example are included in at least one embodiment / method or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment / method or example. Moreover, the specific features, structures, materials or characteristics described may be combined in an appropriate manner in any one or more embodiments / methods or examples. In addition, those skilled in the art may combine and combine different embodiments / methods or examples described in this specification and the features of different embodiments / methods or examples, unless they are contradictory.

[0121] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0122] It should be understood by those skilled in the art that the above embodiments are merely for the purpose of illustrating the present invention clearly, and are not intended to limit the scope of the present invention. For those skilled in the art, other changes or modifications may be made based on the above invention, and these changes or modifications are still within the scope of the present invention.

Claims

1. A photovoltaic cell model parameter extraction method, characterized in that: include: Establish a mathematical model of photovoltaic cells and determine the model parameters to be optimized in the mathematical model; Constructing the fitness function of the mathematical model of photovoltaic cells; Construct a parameter optimization model based on the whale optimization algorithm and differential evolution operator, and use the fitness function as the objective function; Execute the parameter optimization model and obtain the optimal parameter set corresponding to the model parameters to be optimized in the mathematical model; The final photovoltaic cell mathematical model is obtained through the optimal parameter set and the error is verified. If the error index is greater than the set value, the optimization parameters in the parameter optimization model are adjusted and the optimal parameter set is obtained again. Among them, the method of establishing the mathematical model of photovoltaic cells includes: Determine the reverse saturation current I0, Among them, I 0ref is the reverse saturation current at the reference temperature, E g is the band gap energy of the semiconductor, T is the absolute temperature, T ref is the reference temperature, n is the ideality factor of the diode, N s is the number of batteries connected in series, is the thermovoltage, κ is the Boltzmann constant, and q is the electron charge; Determine the diode current I D , Where I0 is the reverse saturation current of the diode, R s is the series resistance, I is the output current of the photovoltaic cell; Determine the parallel resistor current I sh , Among them, R sh is the parallel resistance; Determine the photocurrent I ph , Among them, I ph0 is the photocurrent at reference temperature and light intensity, α I is the temperature coefficient of the photocurrent, G is the actual light intensity, G ref is the reference light intensity; Establishing the mathematical model of photovoltaic cells I=I ph -I D -I sh ; The model parameters to be optimized include: photocurrent I ph , diode reverse saturation current I0, reference photocurrent I ph0 , Reference reverse saturation current I 0ref , the temperature coefficient of the photocurrent α I , diode ideality factor n, series resistance R s , parallel resistance R sh ; Among them, the construction method of the fitness function includes: Determine physical parameter constraints And determine the penalty function for solutions that violate physical constraints Among them, θ is the parameter model to be optimized, λ is the penalty coefficient, and D is the dimension of the parameter. and is the maximum and minimum value of the jth parameter, θ j is the current value of the jth parameter; Determining the Bayesian error term Where N is the number of experimental data points, I model (V i ,θ) is the voltage V i and the calculated current under the parameter θ, I exp (V i ) is the voltage V i The actual current under i is the standard deviation of the ith measurement value; Determine the comprehensive fitness function Among them, θ is the parameter model to be optimized, MSE is the mean square error, MAE is the mean absolute error, RMSE is the root mean square error, RE is the relative error, R 2 is the determination coefficient, w1, w2, w3, w4, W5, and W6 are weight factors, γ is the regularization coefficient, and α is the adjustment coefficient.

2. A photovoltaic cell model parameter extraction method according to claim 1, characterized in that: Parameter optimization algorithms include: Initialize the population: Use the whale optimization algorithm as the basic algorithm and initialize it; set the population size to N, randomly generate N candidate solutions in the parameter search space, and form the initial population X = {x1, x2, ..., x N }, where x i =[x i1 ,x i2 ,...,x iD ], D is the parameter dimension; Fitness calculation: Calculate each candidate solution x according to the fitness function f(·) of the optimization algorithm i The fitness value f(x i ); Iterative update: set the maximum number of iterations T max , for t = 1 to T max Each iteration of performs the following steps: Update control parameters is the current iteration number; generate random vector r1, random vector r2 and random number p; Update candidate solution x i Get the position Perform mutation operation v on the updated candidate solution i =x r1 +F·(x r2 -x r3 ) and crossover operations Among them, x r1 、x r2 、x r3 are individuals randomly selected from the population and are different from each other, and x r1 ≠x i , F is the variation factor, u i，d is the dth dimension of the i-th trial vector, C R is the crossover probability, d rand To randomly select an integer in {1,2,...,D}, x i,d is the dth dimension of the i-th individual, μ is the chaotic mapping parameter; Select the candidate solutions after mutation and crossover operations, Among them, u i is the experimental solution generated after mutation and crossover operations, is the updated position; Calculate each The fitness value of And update the global optimal solution 3. A photovoltaic cell model parameter extraction method according to claim 2, characterized in that: Update candidate solution x i The location methods include: Calculate the coefficient vector A = 2ar1-a, the coefficient vector C = 2r2; If p<0.5 and |A|<1, then surround the prey: If p<0.5 and |A|≥1, then perform the encirclement of prey: If p≥0.5, perform a spiral update of the position: Among them, x best is the current global optimal solution, is the updated position, D1 is the distance calculated in the operation of surrounding the prey, and x rand is an individual randomly selected from the population, D2 is the random distance calculated in the encirclement prey operation, D3 is the distance calculated in the spiral update position, b is a constant defining the spiral shape, and l is a random number.

4. A photovoltaic cell model parameter extraction method according to claim 2, characterized in that: Mutation Factor Crossover probability Among them, F min and F max are the minimum and maximum values ​​of the variation factor, and are the minimum and maximum crossover probabilities.

5. The photovoltaic cell model parameter extraction method according to claim 1, characterized in that: Methods for error verification include: The optimal parameter set Bring in the mathematical model of photovoltaic cells; According to the mathematical model of photovoltaic cells Calculate the voltage V i Calculated current I model (V i ); Determine the voltage and actual current data obtained from the experimental measurement {(V i , I exp (V i ))}, where i=1, 2, 3, ..., M, where M is the number of data points; Calculate the mean square error AMSE, mean absolute error AMAE, and determination coefficient AR by calculating the current and the actual current 2 and mean relative error AMRE; Set the corresponding mean square error threshold ∈MSE, mean absolute error threshold ∈MAE, and determination coefficient threshold ∈R 2 and mean relative error threshold ∈ MRE; Constructing a composite error function in, is the weight coefficient of each error term; If E total If ≤1, the error verification passes; otherwise, the error verification fails, and the optimization parameters in the parameter optimization model are adjusted.

6. A photovoltaic cell model parameter extraction method terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, a photovoltaic cell model parameter extraction method according to any one of claims 1 to 5 is implemented.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a photovoltaic cell model parameter extraction method according to any one of claims 1 to 5 is implemented.

8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, a photovoltaic cell model parameter extraction method according to any one of claims 1 to 5 is implemented.

Images