A photovoltaic array reconstruction method and device based on dandelion optimization algorithm

Abstract

The application discloses a photovoltaic array reconstruction method and device based on a dandelion optimization algorithm, and belongs to the technical field of photovoltaic power generation. The method comprises the following steps: establishing an initial photovoltaic array in an n*n mesh connection configuration; and reconstructing the initial photovoltaic array under a shading condition by using a dandelion optimization DO algorithm to obtain a target photovoltaic array. The DO algorithm is applied to photovoltaic reconstruction, so that the photovoltaic power station can output maximum power under different irradiation conditions, that is, the output power of the photovoltaic array under the shading condition can be improved, and therefore the efficiency of the photovoltaic power generation system is improved. Thus, the technical problem that the output characteristics of the photovoltaic array are seriously affected by cloud coverage is solved, and the operation economy and the stability of grid-connected operation of the photovoltaic power station are significantly improved. In addition, the optimal OAR scheme obtained by the DO algorithm can be used by dispatching staff to change the internal connection condition of the photovoltaic array, so that the current photovoltaic power station can be efficiently operated.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of photovoltaic power generation, and more particularly relates to a photovoltaic array reconstruction method and device based on a dandelion optimization algorithm. BACKGROUND

[0002] In the current world, due to the increasing demand for energy, the reserves of fossil energy are decreasing, and renewable energy is attracting more and more attention and gradually replacing most fossil fuels, one of the most promising energies is solar energy. Photovoltaic cells are the core of photovoltaic power generation systems, and their conversion efficiency determines the practical application ability of the entire system. The application of photovoltaic power generation systems is complex, and a single photovoltaic cell element cannot meet the power supply requirements of the power system. Therefore, it is necessary to connect multiple photovoltaic cell structures together to form a photovoltaic cell array to ensure that the photovoltaic system is maximized.

[0003] In a photovoltaic power generation system, a photovoltaic cell array can be understood as a combination of photovoltaic cell elements. Under the condition that the light intensity is constant, the more the number of photovoltaic cell elements connected in series, the higher the power of the photovoltaic cell array, and the stronger the conversion ability. However, mismatch loss and power loss caused by partial shading can cause the energy output of the photovoltaic array to be significantly reduced, and also cause the service life of the photovoltaic array to be shortened. In seeking the maximum power output of the photovoltaic array, by using reconstruction technology based on heuristic algorithms to compensate for these power losses, the power output of the photovoltaic array under local shading can be significantly improved. Photovoltaic arrays are widely used as the most commonly used devices to obtain solar energy. Photovoltaic cells are nonlinear devices that are the core of photovoltaic power generation, and their output characteristics make the photovoltaic array work at a certain operating voltage to produce the maximum output power. However, some unavoidable destructive factors greatly reduce the efficiency of the photovoltaic array. Partial shading condition (PSC) is one of them, which not only makes the output power curve appear multi-peak, but also causes damage to the photovoltaic panel, which may reduce its service life.

[0004] In actual projects, the shielding of large photovoltaic arrays is mainly caused by clouds, and the shape and position of the clouds change over time. The output characteristics of the photovoltaic array are severely affected by the cloud coverage. SUMMARY

[0005] In view of the above defects or improvement needs of the prior art, the present application provides a photovoltaic array reconfiguration method and device based on a dandelion optimizer, which aims to provide a photovoltaic array reconfiguration (OAR) method based on a dandelion optimizer (DO) to improve the output power of a photovoltaic array under a shading condition, thereby improving the efficiency of a photovoltaic power generation system, and thus solving the technical problem that the output characteristics of a photovoltaic array are seriously affected by cloud coverage.

[0006] To achieve the above object, according to one aspect of the present application, a photovoltaic array reconfiguration method based on a dandelion optimizer is provided, comprising:

[0007] S1: establishing an initial photovoltaic array in an n×n mesh connection configuration;

[0008] S2: reconfiguring the initial photovoltaic array under a shading condition by using a dandelion optimizer (DO) algorithm to obtain a target photovoltaic array; specifically as follows:

[0009] S21: initializing the position of a population, setting boundary values, population quantity and iteration times related system parameters;

[0010] S22: calculating the fitness value of each solution, and selecting the best fitness value;

[0011] S23: dandelion rising stage, updating the position according to the iteration formula of the rising stage;

[0012] S24: dandelion descending stage, updating the position by using the iteration formula of the descending stage;

[0013] S25: dandelion landing stage, updating the position by using the iteration formula of the landing stage;

[0014] S26: when the iteration times reach a threshold value, outputting the optimal result, otherwise repeating S22-S25 until the iteration times reach the threshold value, and outputting the corresponding optimal result; taking the optimal result as the target photovoltaic array.

[0015] In one embodiment, the initial photovoltaic array in S1 is represented as:

[0016] ;

[0017] ;

[0018] wherein, is the output voltage of the initial photovoltaic array; is the output current of the initial photovoltaic array; is aLine maximum output voltage; is a Line b Column node current.

[0019] In one embodiment, the iteration formula of the rising phase is represented as:

[0020] ;

[0021] In sunny days, the wind speed can be considered as a lognormal distribution , , represents the position of the dandelion seed in the iteration process; represents the position randomly selected in the search space during the iteration t , , represents a lognormal distribution subject to and ; ; y represents a standard normal distribution N (0, 1); is an adaptive parameter for adjusting the search step size, , is a random number conforming to the standard normal distribution; and represent the lift component coefficient of the dandelion due to the action of the separated vortex flow, , , , The value range of ];

[0022] In the case of rainy days, the updated position is represented as: ; is used to regulate the local search domain of the dandelion, , .

[0023] In one embodiment, the iteration formula of the descending phase is represented as:

[0024] ;

[0025] In the formula, represents Brownian motion, which is a random number from the standard normal distribution; represents the average position of the population in the i th iteration, .

[0026] In one embodiment, the iteration formula of the landing phase is represented as: ;

[0027] wherein, represents the optimal position of the dandelion seed in the i iteration; represents the Levy function , is a random number between [0, 2]; is a fixed constant; and is a random number between [0, 1], , is fixed as 1.5; is a linear increasing function between [0, 2], .

[0028] In one of the embodiments, the method further comprises:

[0029] S3: analyzing and evaluating the performance parameters of the target photovoltaic array by using three evaluation indexes of mismatch loss, fill factor and standard deviation.

[0030] In one of the embodiments, the mismatch loss is represented as: ; the maximum output power of the target photovoltaic array without shading, is the maximum output power of the target photovoltaic array with shading;

[0031] The fill factor is represented as: ; and is the voltage and current at the local maximum power point; and is the open-circuit voltage and short-circuit current of the target photovoltaic array;

[0032] The maximum output power corresponding to the standard deviation is: , and represent the output current and output voltage of the a-th row, respectively.

[0033] According to another aspect of the present application, there is provided a photovoltaic array reconstruction device based on a dandelion optimization algorithm, for executing the photovoltaic array reconstruction method, comprising:

[0034] The establishing module is configured to establish an initial photovoltaic array in an n x n mesh connection configuration;

[0035] The reconstruction module is configured to reconstruct the initial photovoltaic array under shading conditions by using the dandelion optimization DO algorithm to obtain a target photovoltaic array, and is specifically configured to:

[0036] Initialize the positions of the population, set the boundary value, the number of populations and the iteration number related system parameters;

[0037] Calculate the fitness value of each solution, and select the best fitness value;

[0038] In the dandelion ascending stage, the position is updated according to the iteration formula of the ascending stage;

[0039] In the dandelion descending stage, the position is updated by using the iteration formula of the descending stage;

[0040] In the dandelion landing stage, the position is updated by using the iteration formula of the landing stage;

[0041] When the iteration number reaches the threshold value, the optimal result is output, otherwise, steps S22-S25 are repeated until the iteration number reaches the threshold value, and the corresponding optimal result is output; the optimal result is taken as the target photovoltaic array.

[0042] According to another aspect of the present application, an electronic device is provided, comprising a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method when executing the computer program.

[0043] According to another aspect of the present application, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method.

[0044] Overall, compared with the prior art, the above technical solutions conceived by the present application can achieve the following beneficial effects:

[0045] 1. The present application provides a photovoltaic array reconfiguration (OAR) method based on a dandelion optimizer (DO) algorithm, which applies the DO algorithm to photovoltaic reconfiguration, so that the photovoltaic power station can output maximum power under different irradiation conditions, i.e. can improve the output power of the photovoltaic array under shading conditions, thereby improving the efficiency of the photovoltaic power generation system; thereby solving the technical problem that the output characteristics of the photovoltaic array are seriously affected by cloud coverage, and significantly improving the operation economy and stability of grid-connected operation of the photovoltaic power station.

[0046] 2. The optimal OAR scheme obtained by the DO algorithm can be used by the dispatching staff to change the internal connection of the photovoltaic array, so as to ensure that the current photovoltaic power station can operate efficiently. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is a photovoltaic array reconfiguration model diagram of a TCT structure in an embodiment of the present application;

[0048] Figure 2 is a flow chart of OAR based on DO algorithm in an embodiment of the present application;

[0049] Figure 3 is a diagram of the initial photovoltaic array in the case of shading of light within 9 minutes in an embodiment of the present application;

[0050] Figure 4 is a diagram of the photovoltaic array corresponding to the shading of light within 9 minutes based on the DO algorithm optimization in an embodiment of the present application;

[0051] Figure 5a and Figure 5b are the output curves of various algorithms and photovoltaic arrays before and after optimization within 9 minutes respectively. I - U curve and P - U curve. DETAILED DESCRIPTION

[0052] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0053] The present application proposes a photovoltaic array reconfiguration (OAR) method based on dandelion optimizer (DO) to improve the output power of photovoltaic array and the efficiency of photovoltaic power generation system. The method comprises the following steps:

[0054] The present application establishes a photovoltaic array reconfiguration model of 9x9 mesh connection configuration, and the method is as follows:

[0055] A 9x9 TCT connected photovoltaic array is used in this work, as shown in Figure 1 The TCT configuration is the most widely used connection, which is proved to be the most stable topology of photovoltaic array, and can exhibit good performance when facing PSC. It is worth noting that this configuration technique does not change the original position of the photovoltaic array, but changes its electrical connection. The total output voltage of the photovoltaic array can be written as the following equation:

[0056] (1)

[0057] (2)

[0058] wherein, is the photovoltaic array output voltage; is the photovoltaic array output current; is the maximum output voltage of a row a; is the node current of row a and column b.

[0059] The present application uses three evaluation criteria to measure the simulation results of the proposed DO method.

[0060] (1) mismatch loss:

[0061] (3)

[0062] wherein, is the maximum output power of the photovoltaic model without shading, is the maximum output power of the photovoltaic model with shading.

[0063] (2) fill factor

[0064] The fill factor is a key standard for indicating the power loss of the photovoltaic system under PSC, and the formula is as follows:

[0065] (4)

[0066] wherein, and are the voltage and current at the local maximum power point; and are the open-circuit voltage and short-circuit current of the photovoltaic array, usually given by the manufacturer.

[0067] (3) standard deviation

[0068] The present application uses the standard deviation (STD) to evaluate the reconfiguration stability of the heuristic algorithm.

[0069] When the photovoltaic array works normally under PSC, they will run at a position deviating from their maximum power point, resulting in a decrease in output power. In order to reduce the impact of mismatch loss caused by PSC, a reconstruction method is used to make the irradiation of the photovoltaic array uniform, so as to maximize the output power, which can be represented as:

[0070] (5)

[0071] wherein, and represent the output current and output voltage of each row, respectively.

[0072] The DO algorithm simulates the flight process of dandelion seeds, which are one of the most representative plants that rely on wind for seed propagation. Under appropriate conditions, its seeds can fly tens of kilometers with the wind. When dandelion seeds fly, they form two vortices that generate upward resistance. When the seeds fall at a lower speed, the two vortices above become larger and symmetrical. A symmetrical vortex ensures the stable descent of the seeds; that is, the pappus is level with the ground, and the fruit points downward. To fly a long distance, dandelion seeds need to maintain a relatively stable height. The separated vortices are maintained at a fixed distance below the dandelion crown. Strangely, the porosity of the dandelion crown seems to be precisely adjusted to stabilize the vortex ring. The pappus is composed of slender filaments that radiate outward from the central handle, similar to the spokes on a bicycle wheel. This consistency is the key to the stability of the separated vortex above the dandelion seed, thus helping the seed to maintain stability during long-distance flight. Wind speed and weather are two main factors affecting the spread of dandelion seeds. Wind speed is used to determine whether the seeds fly far or short. Weather controls whether dandelion seeds can fly, affecting the ability of dandelions to grow in nearby or distant spaces. Dandelion seeds undergo three stages of propagation, as follows. In the ascending stage, dandelion seeds produce vortices above them, which rise under the influence of drag in sunny and windy weather. Conversely, if the weather is rainy, there will be no vortices above the seeds. In this case, only a local search can be performed. In the descending stage, when the seeds rise to a certain height, they will descend steadily. In the landing stage, dandelion seeds eventually fall randomly in a place under the influence of wind and weather, growing new dandelions. Dandelions evolve their population in three stages by passing seeds to the next generation. The DO algorithm operates as follows.

[0073] (1) Initialization

[0074] Similar to other natural-inspired heuristic algorithms, the DO algorithm implements population evolution and iterative optimization based on population initialization. In the proposed DO algorithm, it is assumed that each dandelion seed represents a candidate solution, and its population is represented as:

[0075] population= (6)

[0076] wherein pop denotes the population size, Dim is the dimension of the variable; each candidate solution is randomly generated given the upper bound ( UB ) and the lower bound ( LB ) of the problem, and the i th candidate solution is:

[0077] (7)

[0078] where, is an integer between 1 and pop rand represents a random number between (0, 1). LB and UB is represented as:

[0079] LB [ ], UB [ ](8)

[0080] In the initialization process, DO considers the individual with the best fitness value as the initial elite, which is roughly considered as the best place for dandelion seeds to grow. Taking the minimum value as an example, the mathematical expression of the initial elite is given as:

[0081] , = (9)

[0082] where, denotes two indices with equal values.

[0083] (2) Ascending phase

[0084] In the ascending phase, dandelion seeds need to reach a certain height to be carried away from their parent plants. Under the influence of wind speed, air humidity, etc., dandelion seeds will rise to different heights. The weather here is divided into the following two cases.

[0085] The first case is in sunny weather, where the wind speed can be considered as a lognormal distribution . Under this distribution, random numbers are more distributed along the Y-axis, which increases the chances of dandelion seeds being spread to distant areas. Therefore, in this case, DO emphasizes exploration. In the search space, dandelion seeds are randomly blown to different positions by the wind. The ascending height of dandelion seeds is determined by the wind speed. The stronger the wind, the higher the dandelion flies, and the farther the seeds are scattered. Under the influence of wind speed, the vortex above the dandelion seed is constantly adjusted, causing them to rise in a spiral shape. The corresponding mathematical expression in this case is:

[0086] (10)

[0087] where, denotes the position of the dandelion seed in the iteration process; denotes the randomly selected position in the search space during the iteration t . Equation (11) provides an expression for randomly generating positions.

[0088] (11) ​​

[0089] denotes subject to =0 and =1 lognormal distribution, the mathematical formula is:

[0090] (12)

[0091] wherein, y denotes the standard normal distribution N (0, 1). is an adaptive parameter mathematical expression for adjusting the search step size is:

[0092] (13)

[0093] wherein, is a random parameter between [0, 1], which tends to 0 in the nonlinear decreasing process. This fluctuation makes the algorithm focus on global search in the early stage, and turn to local search in the later stage, which is beneficial to ensure accurate convergence after global search is completed. and denote the lift component coefficient of dandelion due to the action of separated vortex. The force variable is calculated by equation (14).

[0094] , , (14)

[0095] wherein, the value range of ].

[0096] The second case, in rainy days, due to the influence of air resistance, humidity and other factors, dandelion seeds can not be properly lifted with the wind. In this case, dandelion seeds are developed in their local area, and the corresponding mathematical expression is:

[0097] (15)

[0098] wherein, is used to normalize the local search domain of dandelion, and equation (16) is used to calculate the definition domain.

[0099] , (16)

[0100] wherein, show "downward convex" oscillation, which is beneficial to the algorithm with large step size in the early stage and small step size in the later stage. At the end of iteration, the parameter gradually tends to 1 to ensure that the population finally converges to the optimal search area.

[0101] In summary, the mathematical expression of the dandelion seed in the rising stage is:

[0102] (17)

[0103] where, is a random number following the standard normal distribution.

[0104] First, in sunny weather, the dandelion seed is updated according to the randomly selected position information, emphasizing the exploration process. The vortex above the seed acts on the movement vector by multiplying x and y components to correct the direction of the dandelion moving in the spiral. In the second case, the dandelion seed is widely used in the local population. The normal distribution of random numbers is dynamically controlled to develop and explore. In order to make the algorithm more global search-oriented, the cutoff point is set to 1.5. This setting makes the dandelion seed traverse the entire search space as much as possible in the first stage, providing the correct direction for the iterative optimization of the next stage.

[0105] (3) The descending stage

[0106] In this stage, the proposed DO algorithm also emphasizes exploration. The dandelion seed stabilizes after rising to a certain distance. In DO, Brownian motion is used to simulate the trajectory of the dandelion. Since Brownian motion follows a normal distribution at each change, individuals are more likely to traverse more search populations during iterative updates. In order to reflect the stability of the dandelion's descent, the average position information after the rising stage is adopted. This helps the entire individual to develop towards a promising population. The corresponding mathematical expression is:

[0107] (18)

[0108] where, represents Brownian motion, which is a random number from the standard normal distribution. represents the average position of the population in the i th iteration, and its mathematical expression is:

[0109] (19)

[0110] The above formula shows the regeneration process of the dandelion seed during the descending process. The average position information of the population is crucial for the iterative update of the individual, which directly determines the evolution direction of the individual. This irregular movement makes it highly likely for the search agent to escape from the local extremum during the iterative update process, thus prompting the population to search in the vicinity of the global optimum.

[0111] (4) Landing stage:

[0112] According to the first two stages, the dandelion seeds are randomly selected to fall anywhere. As the iteration proceeds, the algorithm has the potential to converge to the global optimal solution. Thus, the optimal solution obtained is the approximate location where the dandelion seeds are most likely to survive. To precisely converge to the global optimum, the search agents borrow the elite information of the current optimal solution and develop in their vicinity. As the population evolves, the global optimal solution can eventually be found. This behavior is represented in equation (20).

[0113] (20)

[0114] where, represents the best position of the dandelion seeds in the i th iteration; represents the Levy function, which is calculated using equation (21):

[0115] (21)

[0116] where, is a random number between [0, 2]; is a fixed constant of 0.01; and are random numbers between [0, 1]. The mathematical expression for is:

[0117] (22)

[0118] where, is fixed at 1.5. is a linearly increasing function between [0, 2], calculated using equation (23).

[0119] (23)

[0120] To precisely converge to the global optimum, a linearly increasing function is applied to the individuals to avoid over-development. This stage uses the Levy flight coefficient to simulate the step length of the individuals' movement. The reason is that under the Gaussian distribution, the Levy flight coefficient can be crossed by the agents to other positions with a high probability, developing more local search domains in a limited number of iterations.

[0121] As shown in Figure 1, the algorithm flow of DO is as follows: Figure 2

[0122] (1) Initialize the positions of the population, set the boundary values, population size, and number of iterations, etc.

[0123] (2) Calculate the fitness value of each solution and select the best fitness value.

[0124] (3) Dandelion ascent stage and position update according to equations (10)-(17);​

[0125] (4) Dandelion descent phase, update the position using equations (18)-(19);

[0126] (5) Dandelion landing phase, update the position using (20)-(23);

[0127] (6) Determine whether the number of iterations is reached, if so, output the optimal result, otherwise repeat steps 2-5.

[0128] 4. The ASW-280M type solar photovoltaic panel is adopted in the application, and the specific parameters are shown in Table 1.

[0129] Table 1 Specific parameters of photovoltaic modules

[0130]

[0131] In order to verify the effectiveness of the reconstruction algorithm, different sizes of photovoltaic arrays and different shading methods are often simulated. In this section, the most commonly used 9x9 photovoltaic array is used to verify the effect of DO, and 9 minutes of moving cloud shadow is used to cooperate with the simulation experiment of this paper. The simulation tool used is MATLAB 2021b. And GA, PSO and other reconstruction methods are constructed for performance comparison. The running time, iteration number and total number of DO algorithm are set to 20, 200 and 20. Figure 3 is the shading condition of the photovoltaic array in 9 minutes.

[0132] The present application aims to reconfigure the shaded photovoltaic array by using heuristic algorithms, change its connection mode according to the simulation results, increase the output power, and improve the system operation benefit. The present application introduces a 25-megawatt photovoltaic power station with 20 identical subsystems to evaluate the performance of the present application, wherein each subsystem is composed of a 9x9 TCT configured photovoltaic array. The introduced partially shaded photovoltaic array simulates the slow movement of clouds for nine minutes, and different colors represent different irradiance. The working temperature is set to 25℃, and the irradiance distribution of the photovoltaic array at each minute is as shown in Figure 4 .

[0133] When the DO algorithm is used to optimize and reconfigure the photovoltaic array, the internal connection of the photovoltaic array is changed according to the best optimization result of the DO algorithm simulation. The maximum and minimum optimization output power obtained by each heuristic algorithm simulation, and the mismatch loss and fill factor of each method are shown in Table 2. Among all the algorithms, the and obtained by the DO algorithm are higher than those of other algorithms. Moreover, the STD of the DO algorithm is smaller than that of other algorithms, which indicates that the DO algorithm has higher stability. It is obvious that the mismatch loss and ffThe DO algorithm performs better in all algorithms. The optimal mismatch loss obtained by the DO algorithm is 38.13%, 3.75% and 0.51% lower than that before optimization, GA and PSO, respectively. The optimal power output obtained by the DO algorithm is 20.70%, 1.15% and 0.13% higher than that before optimization, GA and PSO, respectively. ff

[0134] Figure 5a Figure 5b The output I U curves and P U curves of various algorithms and photovoltaic arrays before and after optimization for 9 minutes. It is observed that the more shadows, the smoother the output characteristic curve. The output I U curve usually has many inflection points, and the output P U curve has many peaks. Taking the 4th minute as an example, there are 6 power peaks before reconfiguration. DO reduces the number of energy peaks to one, which completely eliminates local peaks. Therefore, compared with other methods of the present application, DO achieves significant reconstruction effect.

[0135] An OAR technology based on DO is proposed. The method is realized by discretizing the original DO algorithm and combining photovoltaic reconstruction methods. DO can provide optimal solution in real time, thereby avoiding local optimal solution. By quantitatively comparing the output power, mismatch loss, fill factor, standard deviation and output characteristic curve of DO, GA and PSO, the superiority of DO is proved.

[0136] Table 2 Comparison of simulation results of various algorithms

[0137]

[0138] Those skilled in the art will readily understand that the above description is only preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.​​​​​​

Claims

1. A photovoltaic array reconfiguration method based on dandelion optimization algorithm, characterized in that, The method comprises: S1: establishing an initial photovoltaic array in an n*n mesh connection configuration; S2: reconstructing the initial photovoltaic array under a shading condition by using a dandelion optimization DO algorithm to obtain a target photovoltaic array; specifically as follows: S21: initializing the position of a population, setting boundary values, population quantity, and iteration times related system parameters; S22: calculating the fitness value of each solution, selecting the best fitness value, and taking the best fitness value as an initial elite, which is considered as the most suitable position for dandelion seed growth; the dandelion evolves through seed transmission to the next generation in three stages: S23: dandelion ascending stage, vortex is generated above the dandelion seed, and the vortex rises under the action of drag force in sunny and windy weather; no vortex is generated above the seed in rainy weather, and only local search can be performed; the position is updated according to the iteration formula of the ascending stage; is the iteration t+ 1is the position of the dandelion seed during the iteration is the iteration t is the position of the dandelion seed during the iteration is an adaptive parameter for adjusting the search step size; and is the lift component coefficient of the dandelion due to the detached eddy action; is the iteration t is the randomly selected position in the search space during the iteration is subject to = 0 and = 1 lognormal distribution; is the local search domain for the dandelion; is a random number conforming to the standard normal distribution; S24: the dandelion descending stage, when the dandelion seed rises to a certain height, it will steadily descend; the position is updated by using the iteration formula of the descending stage; the iteration formula of the descending stage is: ; is a random number from a standard normal distribution, which is a Brownian motion; is the average position of the population in the i th iteration; S25: dandelion landing stage, dandelion seeds finally grow into new dandelions in a random place under the influence of wind and weather; position update is performed using the iteration formula of the landing stage; the iteration formula of the landing stage is: ; is the optimal position of the dandelion seed in the i th iteration; is a Levy function; is a linear increasing function between [0, 2]; S26: when the iteration times reach a threshold value, the optimal result is output, otherwise, steps S22-S25 are repeated until the iteration times reach the threshold value, and the corresponding optimal result is output; the optimal result is taken as the target photovoltaic array.

2. The dandelion optimization algorithm based photovoltaic array reconfiguration method of claim 1, wherein, The initial photovoltaic array in S1 is represented as: ； ； wherein, is the output voltage of the initial photovoltaic array; is the output current of the initial photovoltaic array; is a the row maximum output voltage; is a the row b the column node current.

3. The dandelion optimization algorithm based photovoltaic array reconfiguration method of claim 1, wherein, The method further comprises: S3: analyzing and evaluating the performance parameters of the target photovoltaic array by using three evaluation indexes of mismatch loss, fill factor, and standard deviation.

4. The photovoltaic array reconstruction method based on a dandelion optimization algorithm according to claim 3, characterized in that, The mismatch loss is expressed as: ; Pmax, no shading is the maximum output power of the target photovoltaic array without shading, Pmax, shading is the maximum output power of the target photovoltaic array with shading; The fill factor is expressed as: ; and are the voltage and current at the local maximum power point; and are the open circuit voltage and short circuit current of the target photovoltaic array; The maximum output power corresponding to the standard deviation is: , and represent the output current and output voltage of the a-th row, respectively.

5. A photovoltaic array reconfiguration device based on dandelion optimization algorithm, characterized by, A device for executing the photovoltaic array reconstruction method according to any one of claims 1-4, comprising: an establishing module for establishing an initial photovoltaic array in an n*n mesh connection configuration; a reconstruction module for reconstructing the initial photovoltaic array under a shading condition by using a dandelion optimization DO algorithm to obtain a target photovoltaic array, specifically for: initializing the position of a population, setting boundary values, population quantity, and iteration times related system parameters; calculating the fitness value of each solution, and selecting the best fitness value; in the dandelion ascending stage, updating the position according to the iteration formula of the ascending stage; in the dandelion descending stage, updating the position by using the iteration formula of the descending stage; in the dandelion landing stage, updating the position by using the iteration formula of the landing stage; when the iteration times reach a threshold value, the optimal result is output, otherwise, steps S22-S25 are repeated until the iteration times reach the threshold value, and the corresponding optimal result is output; the optimal result is taken as the target photovoltaic array. 6.An electronic device comprising a memory and a processor, the memory storing a computer program, wherein, The processor executes the computer program to implement the steps of the method according to any one of claims 1-4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method according to any one of claims 1-4.

Images