Photovoltaic array multi-level aging identification and processing method, device, equipment and medium

Abstract

The present invention belongs to the field of photovoltaic abnormality identification technology, and specifically relates to a photovoltaic array multi-level aging identification processing method, device, equipment and medium. The method obtains the voltage change rate and current change rate of the maximum power point of the photovoltaic array, inputs the first fault identification model based on the support vector machine, and obtains the abnormal aging degree of the photovoltaic array. During the training process of the first fault identification model, the training set data is obtained based on the photovoltaic array simulation model, and the model is provided with a simulated resistor to simulate different aging degrees, and the parameters of the support vector machine are optimized using a dynamic adaptive cuckoo search algorithm. When a mild, moderate or severe abnormality is identified, the maximum power point voltage change rate and current change rate of the photovoltaic string in the abnormally aged photovoltaic array are further obtained, and the aging degree of each photovoltaic string is obtained using the second fault identification model. According to the aging degree and the switch matrix, dynamic adaptive compensation operation and maintenance measures with different duty ratios of the compensation strings corresponding to the multi-level faults are taken.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic anomaly identification, and specifically relates to a photovoltaic array multi-level aging identification processing method, device, equipment and medium. Background Art

[0002] As the global demand for renewable energy continues to grow, photovoltaic power generation, as a clean and renewable form of energy, has become increasingly important. As the core component of the photovoltaic power generation system, the performance of the photovoltaic array directly affects the power generation efficiency and operational stability of the entire system. However, in the actual operation process, the photovoltaic array will be affected by many factors, such as environmental factors, material aging, mechanical stress, etc., which will cause the performance of the photovoltaic modules to gradually decline and show abnormal aging phenomena to varying degrees. Abnormal aging of the photovoltaic array will not only lead to reduced power generation efficiency, but may also cause a series of operation and maintenance problems. Traditional operation and maintenance methods often rely on manual inspections and regular testing, which is not only inefficient, but also difficult to accurately judge the degree of aging of photovoltaic modules, resulting in high operation and maintenance costs and poor results. Summary of the invention

[0003] The purpose of the present invention is to provide a photovoltaic array multi-level aging identification processing method, device, equipment and medium to solve the problem of how to efficiently identify the aging degree of photovoltaic arrays in the prior art.

[0004] In order to achieve the above object, the present invention adopts the following technical solution:

[0005] In a first aspect, the present invention provides a photovoltaic array multi-level aging identification and processing method, comprising the following steps:

[0006] Obtain the voltage change rate and current change rate of the maximum power point of the photovoltaic array;

[0007] The voltage change rate and current change rate of the maximum power point of the photovoltaic array are input into a pre-trained first fault identification model, and the first fault identification model outputs the abnormal aging degree of the photovoltaic array; operation and maintenance are performed according to the abnormal aging degree of the photovoltaic array; wherein the first fault identification model is obtained based on support vector machine training; during the training process of the first fault identification model, a dynamic adaptive cuckoo search algorithm is used to optimize the penalty factor and kernel function parameters of the support vector machine; the training set data of the first fault identification model is obtained based on a photovoltaic array simulation model, and the photovoltaic array simulation model is provided with a simulated resistor for simulating the abnormal aging degree of the photovoltaic array.

[0008] Furthermore, the abnormal aging degree of the PV array includes normal, mildly abnormal, moderately abnormal, and severely abnormal;

[0009] Operation and maintenance are performed according to the abnormal aging degree of the photovoltaic array, including:

[0010] When the abnormal aging degree output by the first fault identification model is a slight abnormality, a moderate abnormality or a severe abnormality, obtaining a voltage change rate and a current change rate of a maximum power point of each photovoltaic string;

[0011] The voltage change rate and current change rate of the maximum power point of each photovoltaic string are input into a pre-trained second fault identification model, and the second fault identification model outputs the abnormal aging degree of each photovoltaic string; wherein the abnormal aging degree of each photovoltaic string includes no aging, slight aging, moderate aging and severe aging;

[0012] When the PV strings are not aged, no operation is performed; when the PV strings are slightly aged, the PV strings and the compensation strings are used alternately, and the PV strings are continuously monitored; when the PV strings are moderately aged or severely aged, the PV strings are removed, and the compensation strings are cut into the position of the removed PV strings in real time for replacement.

[0013] Furthermore, the compensation string is connected in parallel with the photovoltaic string, and the compensation string is connected to the busbar through a switching switch.

[0014] Furthermore, the compensation string is connected to the busbar through a switching switch, including:

[0015] When the photovoltaic strings are not aged, the duty cycle of the switches corresponding to each compensation string is 0;

[0016] When the photovoltaic strings are slightly aged, the switching duty ratios of the compensation strings and the slightly aged photovoltaic strings are equal, and the compensation strings and the slightly aged photovoltaic strings are switched into the busbar in turn;

[0017] When the photovoltaic strings are moderately aged or severely aged, the moderately aged or severely aged photovoltaic strings are removed; the compensation strings are alternately cut into the busbar, or all the compensation strings are cut into the busbar;

[0018] When each compensation string is switched into the busbar alternately, the switching duty ratio of each compensation string is determined according to the following formula:

[0019]

[0020] in, represents the switching duty cycle of compensation string i; represents the output power of compensation string i; represents the output power of compensation string j; n Indicates the number of compensation strings.

[0021] Further, in the step of inputting the voltage change rate and the current change rate of the maximum power point of the photovoltaic array into a pre-trained first fault identification model, and the first fault identification model outputting the abnormal aging degree of the photovoltaic array, the first fault identification model is trained in the following manner:

[0022] Acquire training set data for training a first fault identification model;

[0023] Build a support vector machine model and initialize the penalty factor and kernel function parameters of the support vector machine model;

[0024] Based on the dynamic adaptive cuckoo search algorithm, M bird nests are randomly generated, each of which represents a set of penalty factors and kernel function parameters. The fitness value corresponding to each bird nest is calculated according to the fitness function using the training set data. The fitness function is defined as the average value of the sum of the absolute values ​​of the errors between the diagnosis results of the support vector machine model and the true results.

[0025] Update the nest position according to Levy flight and nest host mechanism; calculate the fitness value of the updated nest, update the current optimal fitness and optimal nest position; determine whether the iteration termination condition is met; when the iteration terminates, output the global optimal fitness value and the corresponding optimal nest position; take the penalty factor and kernel function parameters corresponding to the optimal nest position as the optimal parameters, and substitute the optimal parameters into the support vector machine model.

[0026] Further, obtaining training set data for training the first fault identification model includes:

[0027] A photovoltaic array simulation model is constructed in a simulation platform; wherein the photovoltaic array simulation model includes a plurality of simulated photovoltaic strings and a plurality of simulated collection busbars; each simulated photovoltaic string is provided with switches corresponding to the number of simulated collection busbars, which are used to connect the simulated collection busbars respectively; no simulated resistor is provided on one of the simulated collection busbars, and corresponding simulated resistors are provided in series on the remaining simulated collection busbars, and the simulated resistors are used to simulate different aging degrees of the simulated photovoltaic strings;

[0028] The switches on the simulated photovoltaic strings are closed respectively with different simulated collecting buses to obtain the grid connection point voltage and grid connection point current corresponding to the photovoltaic array simulation model under different simulated resistance conditions;

[0029] Calculate the voltage change rate and current change rate at the maximum power point of the photovoltaic array simulation model according to the grid connection point voltage and grid connection point current; determine the aging abnormality degree of the photovoltaic array simulation model according to the voltage change rate and a preset first threshold value;

[0030] The voltage change rate and current change rate at the maximum power point of the photovoltaic array simulation model, as well as the corresponding aging abnormality degree, are used as training set data.

[0031] Furthermore, according to Levi flight and nest host mechanism, the nest location is updated, including:

[0032]

[0033] In the formula, t is the current iteration number; For the t At the iteration, i The location information of the bird's nest; For the t+ In the first iteration, i The location information of the bird's nest; M is the total number of nests; λ is the Levy index; L ( λ ) is the random search path that satisfies Levy flight; α represents the step length control amount;

[0034]

[0035] In the formula, μ and ν represents random numbers that follow a normal distribution;

[0036] The step length control amount α is determined as follows:

[0037]

[0038] In the formula, d , e , f is the step size control factor, d > e > f ; N The maximum number of iterations to terminate; t is the current iteration number.

[0039] In a second aspect, the present invention provides a photovoltaic array multi-level aging fault identification device, comprising:

[0040] A data acquisition module, used to obtain the voltage change rate and current change rate of the maximum power point of the photovoltaic array;

[0041] The abnormal identification module is used to input the voltage change rate and current change rate of the maximum power point of the photovoltaic array into a pre-trained first fault identification model, and the first fault identification model outputs the abnormal aging degree of the photovoltaic array; operation and maintenance are performed according to the abnormal aging degree of the photovoltaic array; wherein the first fault identification model is obtained based on support vector machine training; during the training process of the first fault identification model, a dynamic adaptive cuckoo search algorithm is used to optimize the penalty factor and kernel function parameters of the support vector machine; the training set data of the first fault identification model is obtained based on a photovoltaic array simulation model, and the photovoltaic array simulation model is provided with a simulated resistor for simulating the abnormal aging degree of the photovoltaic array.

[0042] According to a third aspect of the present invention, an electronic device is provided, comprising a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to implement the photovoltaic array multi-level aging identification and processing method as described above.

[0043] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the multi-level aging identification and processing method for photovoltaic arrays as described above is implemented.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] The photovoltaic array multi-level aging identification and processing method provided by the present invention is that the first fault identification model is obtained based on support vector machine SVM training, and the penalty factor and kernel function parameters of SVM are optimized by dynamically adapting the cuckoo search algorithm, thereby improving the recognition accuracy and generalization ability of the model. It can quickly and accurately identify the abnormal aging degree of the photovoltaic array without relying on traditional manual inspections and regular testing, greatly improving the operation and maintenance efficiency. The photovoltaic array multi-level aging fault identification device, electronic device and computer-readable storage medium provided by the present invention also solve the problems raised in the background technology section.

[0046] The multi-level aging identification and processing method for photovoltaic arrays provided by the present invention can identify the abnormal aging degree of the photovoltaic array as a whole, and can also be further refined to the abnormal aging degree of each photovoltaic string, thereby realizing multi-level aging fault identification. According to different aging degrees, corresponding operation and maintenance measures are taken, such as continuing to monitor and remove aging strings, and cutting in compensation strings based on different duty cycles of string performance, etc., which can extend the service life of the photovoltaic array and reduce operation and maintenance costs.

[0047] In the present invention, the real-time switching-in and switching-out operations based on the adaptive real-time switching of different duty ratios of the compensation string performance enable the photovoltaic array to operate stably when an aging fault occurs, thereby improving the reliability and stability of the system.

[0048] The present invention constructs a photovoltaic array simulation model, sets simulation resistors and a switch matrix to simulate the operation status of the photovoltaic array under different aging degrees, reduces the number of system simulation resistors to be set, and simplifies the simulation topology. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0050] Figure 1 A schematic diagram of a process flow of a photovoltaic array multi-level aging identification and processing method according to an embodiment of the present application;

[0051] Figure 2 A topological structure diagram of a 3*3 photovoltaic array simulation model in an embodiment of the present application;

[0052] Figure 3 This is a flow chart of improving the support vector machine based on the dynamic adaptive cuckoo search algorithm in an embodiment of the present application;

[0053] Figure 4 This is an access topology diagram of the compensation string in the embodiment of the present application;

[0054] Figure 5 The output characteristic diagram of the photovoltaic array when aging failure occurs in multiple levels is obtained by simulation in the embodiment of the present application; (a) the photovoltaic array is simulated U - I Characteristic comparison and analysis; (b) Simulated photovoltaic array U - P Comparative analysis of characteristics;

[0055] Figure 6 The output characteristic diagram of the photovoltaic array when a multi-level abnormal aging failure occurs is obtained experimentally in the embodiment of the present application; (a) the experimental photovoltaic array U - I Characteristic comparison and analysis; (b) Experimental photovoltaic array U - P Comparative analysis of characteristics;

[0056] Figure 7 This is a curve diagram of the fault identification fitness evolution in the embodiment of the present application;

[0057] Figure 8 Graph showing the fault recognition accuracy of the traditional SVM model in the embodiment of the present application; (a) fault diagnosis result of the traditional SVM training set; (b) fault diagnosis result of the traditional SVM test set;

[0058] Fig. 9Graph showing the fault recognition accuracy of the first fault recognition model in the embodiment of the present application; (a) fault diagnosis results of the DACS-SVM training set; (b) fault diagnosis of the DACS-SVM test set;

[0059] Fig.10 This is a diagram showing the effect of treating a mild aging fault in an embodiment of the present application; (a) Mild fault U - I Characteristic comparison analysis; (b) Minor fault U - P Comparative analysis of characteristics;

[0060] Fig.11 This is a diagram showing the effect of treating a moderate aging fault in an embodiment of the present application; (a) moderate fault U - I Characteristic comparison analysis; (b) Moderate failure U - P Comparative analysis of characteristics;

[0061] Fig.12 This is a diagram showing the effect of severe fault processing in the embodiment of the present application; (a) severe fault U - I Characteristic comparison analysis; (b) severe fault U - P Comparative analysis of characteristics;

[0062] Fig.13 This is a structural block diagram of a photovoltaic array multi-level abnormal aging fault identification device according to an embodiment of the present invention;

[0063] Fig.14 The present invention is a structural block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0065] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.

[0066] Glossary

[0067] DACS, DACS (Dynamic Adaptation Cuckoo Search) algorithm, that is, dynamic adaptation cuckoo search algorithm.

[0068] SVM, SVM (Support Vector Machine) is a machine learning algorithm.

[0069] Example 1

[0070] like Figure 1 As shown, a photovoltaic array multi-level aging identification and processing method includes the following steps:

[0071] S100, obtaining the voltage change rate and current change rate of the maximum power point of the photovoltaic array.

[0072] Specifically, step S100 obtains the voltage change rate and current change rate of the maximum power point of the photovoltaic array as the simplest feature vector for subsequent abnormal aging identification.

[0073] S200. Input the voltage change rate and current change rate of the maximum power point of the photovoltaic array into a pre-trained first fault identification model, and the first fault identification model outputs the abnormal aging degree of the photovoltaic array; perform operation and maintenance processing according to the abnormal aging degree of the photovoltaic array; wherein, the first fault identification model is obtained based on support vector machine training; during the training process of the first fault identification model, a dynamic adaptive cuckoo search algorithm is used to optimize the penalty factor and kernel function parameters of the support vector machine; the training set data of the first fault identification model is obtained based on a photovoltaic array simulation model, and the photovoltaic array simulation model is provided with a simulated resistor for simulating the abnormal aging degree of the photovoltaic array.

[0074] Specifically, the abnormal aging degree of the photovoltaic array includes normal, slightly abnormal, moderately abnormal and severely abnormal.

[0075] Specifically, in step S200, the first fault identification model is trained in the following manner:

[0076] S210: Obtain training set data for training a first fault identification model.

[0077] Specifically, the training set data includes normal, slightly abnormal, moderately abnormal and severely abnormal conditions of the photovoltaic array, as well as the voltage change rate and current change rate of the maximum power point corresponding to different conditions.

[0078] Specifically, the voltage change rate and current change rate of the maximum power point under abnormal conditions can be calculated as follows:

[0079] (1)

[0080] (2)

[0081] In the formula, and They are the voltage change rate and current change rate at the maximum power point of the photovoltaic array respectively;

[0082] s is the total number of components contained in each photovoltaic string in the photovoltaic array; p is the number of PV strings connected in parallel in the PV array;

[0083] , They are the voltage change and current change at the maximum power point of the photovoltaic array respectively;

[0084] U oc,ref is the open circuit voltage of the components in the reference PV string;

[0085] I sc,ref is the short-circuit current of the reference PV string;

[0086] U m,normal and I m,normal They are respectively the maximum power point voltage and maximum power point current of the photovoltaic array under normal conditions;

[0087] U m,fault and I m,fault They are the maximum power point voltage and maximum power point current of the photovoltaic array when aging abnormality occurs.

[0088] Preferably, after obtaining the training set data, the training set data is normalized.

[0089] Optionally, test set data is also obtained to test the performance of the model after the first fault identification model training is completed.

[0090] Specifically, obtaining training set data for training the first fault identification model includes:

[0091] S2101. Construct a photovoltaic array simulation model in a simulation platform; the photovoltaic array simulation model includes a plurality of simulated photovoltaic strings and a plurality of simulated collection busbars; each simulated photovoltaic string is provided with switches corresponding to the number of simulated collection busbars, which are used to connect the simulated collection busbars respectively; no simulated resistor is provided on one of the simulated collection busbars, and corresponding simulated resistors are provided in series on the remaining simulated collection busbars, and the simulated resistors are used to simulate different aging degrees of the simulated photovoltaic strings;

[0092] S2102, closing the switches on the simulated photovoltaic strings and the different simulated collecting buses respectively, to obtain the grid connection point voltage and grid connection point current corresponding to the photovoltaic array simulation model under different simulated resistance conditions;

[0093] S2103, calculating the voltage change rate and current change rate at the maximum power point of the photovoltaic array simulation model according to the grid connection point voltage and grid connection point current;

[0094] S2104, determining the abnormal degree of aging of the photovoltaic array simulation model according to the voltage change rate and a preset first threshold;

[0095] S2105. The voltage change rate and current change rate at the maximum power point of the photovoltaic array simulation model, as well as the corresponding aging abnormality degree, are used as training set data.

[0096] like Figure 2 As shown, the grid-connected point voltage and grid-connected point current are the voltage and current collected at the grid-connected point of the photovoltaic array.

[0097] Specifically, the grid-connected point voltage is the voltage at the connection point of the entire photovoltaic array and the AC power grid. By measuring the grid-connected point voltage, the maximum power point voltage of the photovoltaic array during normal operation and the maximum power point voltage of the photovoltaic array when aging abnormalities occur can be obtained; the grid-connected point current is the current at the connection point of the entire photovoltaic array and the AC power grid. By measuring the grid-connected point current, the maximum power point current of the photovoltaic array during normal operation and the maximum power point current of the photovoltaic array when aging abnormalities occur can be obtained.

[0098] As an example, the simulation platform may be a Matlab / Simulink platform.

[0099] As an example, Figure 2 As shown, the photovoltaic array simulation model can be a 3×3 simulation model, that is, the photovoltaic array simulation model includes a first simulated photovoltaic string, a second simulated photovoltaic string and a third simulated photovoltaic string, and a first simulated collection bus, a second simulated collection bus and a first simulated collection bus, no simulated resistor is set on the first simulated collection bus, and simulated resistors R are set on the second simulated photovoltaic string and the third simulated photovoltaic string respectively. s1 and R s2 ;

[0100] A switch S3, a switch S6 and a switch S9 are provided on the first simulated photovoltaic string, the switch S3 corresponds to the first simulated collection bus, the switch S6 corresponds to the second simulated collection bus, and the switch S9 corresponds to the third simulated collection bus;

[0101] The second simulated photovoltaic string is provided with a switch S2, a switch S5 and a switch S8, the switch S2 corresponds to the first simulated collection bus, the switch S5 corresponds to the second simulated collection bus, and the switch S8 corresponds to the third simulated collection bus;

[0102] The third simulated photovoltaic string is provided with a switch S1, a switch S4 and a switch S7, the switch S1 corresponds to the first simulated collection bus, the switch S4 corresponds to the second simulated collection bus, and the switch S7 corresponds to the third simulated collection bus.

[0103] The switches S1 to S9 above form a switch array. s1 and R s2 Located between the corresponding simulation busbar and the grid connection point.

[0104] By default, all switches in the switch array are in the open state. You can close different switches as needed to simulate different abnormal aging degrees of the photovoltaic array. For example:

[0105] When the first simulated photovoltaic string, the second simulated photovoltaic string and the third simulated photovoltaic string are all working normally, switches S1, S2 and S3 are closed, and the simulated resistors are on standby. At this time, the simulated photovoltaic array is normal.

[0106] When a moderate abnormal aging fault occurs in the photovoltaic array, if the components in the second simulated photovoltaic string are abnormally aged, switches S1, S8 and S3 are closed, and the simulated resistance R is adjusted. s1 At the corresponding resistance value, the simulated resistor R s1 Connected in series in the second simulated photovoltaic string to simulate a moderately abnormal aging fault;

[0107] When a severe abnormal aging fault occurs in the photovoltaic array, if the aging components are located in the second simulated photovoltaic string and the third simulated photovoltaic string, the switches S1, S8 and S6 are closed, and the simulated resistance R is adjusted. s1 , R s2 At the corresponding resistance value, the simulated resistor R s1 In series in the second simulated photovoltaic string, the simulated resistance R s2 Connected in series in the third simulated PV string, it simulates a severe abnormal aging failure.

[0108] In this solution, the simulated resistor R s1 and R s2 The resistance value of the simulation resistor R s1 , R s2 The resistance value is adjusted to simulate the aging of the PV strings.

[0109] As an example: the voltage change rate at the maximum power point of the photovoltaic array is greater than 0.5% and less than 5%, and the aging degree is classified as mild abnormality. At this time, the corresponding resistance value of the simulated resistor connected in series with any simulated photovoltaic string increases to 0~15Ω; the voltage change rate at the maximum power point of the photovoltaic array is 5%~10%, and the aging degree is classified as moderate abnormality. At this time, the corresponding resistance value of the simulated resistor connected in series with any simulated photovoltaic string increases to 15~30Ω; the voltage change rate at the maximum power point of the photovoltaic array is greater than 10%, and the aging degree is classified as severe abnormality; at this time, the corresponding resistance values ​​of the simulated resistors connected in series with any two simulated photovoltaic strings increase to 15~30Ω and 0~15Ω respectively. It should be noted that the above photovoltaic array simulation model is only an example for the aging simulation of the photovoltaic array (such as the resistance value adjustment of the simulated resistor), and those skilled in the art can flexibly set it according to actual needs.

[0110] The PV array simulation model set up in the above scheme dynamically matches different simulated resistors to the running PV strings by controlling the switch array to simulate different levels of abnormal aging of the PV array. According to the simulated resistors and switch arrays, the number of simulated resistors is reduced and the simulation topology is simplified.

[0111] S220, constructing a support vector machine model, and initializing the penalty factor and kernel function parameters of the support vector machine model.

[0112] It should be noted that the support vector machine model (SVM) in this solution is a supervised learning model.

[0113] S230. Based on the dynamic adaptive cuckoo search algorithm, M bird nests are randomly generated, each bird nest represents a set of penalty factors and kernel function parameters; using the training set data, the fitness value corresponding to each bird nest is calculated according to the fitness function; wherein the fitness function is defined as the average value of the sum of the absolute values ​​of the errors between the diagnosis results of the support vector machine model and the true results.

[0114] Specifically, Figure 3 As shown, before the algorithm starts, the algorithm parameters are initialized, including setting the number of bird nests. M , terminate the number of iterations N , the fixed probability P of the nest host discovering foreign bird eggs, the probability r of each bird egg being discovered, and the step length control parameter α , the upper and lower limits of the bird's nest position [ x min , x max ].

[0115] Specific, randomly generated M Bird nests, represented by a matrix: In the formula, xi For the i The location of the bird's nest is composed of the penalty factor and kernel function parameters of SVM.

[0116] S240, updating the nest position according to Levy flight and nest host mechanism; calculating the fitness value of the updated nest, updating the current optimal fitness and optimal nest position; judging whether the iteration termination condition is met; when the iteration terminates, outputting the global optimal fitness value and the corresponding optimal nest position; taking the penalty factor and kernel function parameters corresponding to the optimal nest position as the optimal parameters, and substituting the optimal parameters into the support vector machine model.

[0117] Specifically, according to Levi's flight and nest host mechanism, the nest position is updated, including:

[0118] (3)

[0119] In the formula, t is the current iteration number; For the t At the iteration, i The location information of the bird's nest; For the t+ In the first iteration, i The location information of the bird's nest; λ is the Levy index; L ( λ ) is the random search path that satisfies the Levy flight; α represents the step size control parameter.

[0120] Random search path that satisfies Levy flight L ( λ ) is determined as follows:

[0121] (4)

[0122] In the formula, μ and ν Represents random numbers that follow a normal distribution.

[0123] Specifically, the step size control parameter α is determined as follows:

[0124] (5)

[0125] In the formula, d , e , f is the step size control factor, d > e > f ; N The maximum number of iterations to terminate; t is the current iteration number.

[0126] After the training is completed, the first fault identification model is the DACS-SVM model.

[0127] Preferably, operation and maintenance processing is performed according to the abnormal aging degree of the photovoltaic array, including: when the abnormal aging degree output by the first fault identification model is a slight abnormality, a moderate abnormality or a severe abnormality, obtaining the voltage change rate and the current change rate of the maximum power point of each photovoltaic string; inputting the voltage change rate and the current change rate of the maximum power point of each photovoltaic string into a pre-trained second fault identification model, and the second fault identification model outputs the abnormal aging degree of each photovoltaic string; wherein the abnormal aging degree of each photovoltaic string includes no aging, slight aging, moderate aging and severe aging; when the photovoltaic string is not aged, no operation is performed; when the photovoltaic string is slightly aged, it is used alternately with the compensation string, and the aging status of the photovoltaic string is continuously monitored; when the photovoltaic string is moderately aged or severely aged, the photovoltaic string is removed, and the compensation string is cut into the position of the removed photovoltaic string in real time for replacement.

[0128] Specifically, the second fault identification model is trained using a support vector machine model. The training method is basically the same as that of the first fault identification model. The difference is that the training set used for training the second fault identification model is based on the simulated aging data of each photovoltaic string in the photovoltaic array simulation model. For example: Figure 2 As shown, in the photovoltaic array simulation model, at the photovoltaic string collection point, the string voltage and string current of each simulated photovoltaic string are collected; according to the string voltage and string current of each simulated photovoltaic string, the string voltage change rate and the string true current change rate of each simulated photovoltaic string at the maximum power point are calculated, and the abnormal aging degree of the photovoltaic string is determined according to the magnitude relationship between the string voltage change rate and the preset second threshold. The string voltage change rate and the string true current change rate at the maximum power point, as well as the abnormal aging degree of the corresponding photovoltaic string, are used as the training set of the second fault identification model. Different switches can be closed according to demand to simulate different abnormal aging degrees of the photovoltaic string. For example, the switch S3 can be closed to simulate the operation of the first simulated photovoltaic string in a normal state, the switch S6 can be closed to simulate the operation of the first simulated photovoltaic string in an abnormally old state, and the switch S9 can also be closed to simulate the operation of the first simulated photovoltaic string in an abnormally old state. In order to achieve the simulation of the aging degree, the resistance value of the simulated resistor can be increased or decreased.

[0129] For the second simulated photovoltaic string and the third simulated photovoltaic string, the above method can also be used to switch on and off the switches and adjust the simulated resistors, which will not be described in detail here.

[0130] Specifically, in order to achieve the replacement of moderately aged or severely aged photovoltaic strings, the present solution also provides a fault handling circuit, including a plurality of compensation strings, which are connected in parallel with the photovoltaic strings. When the moderately aged or severely aged photovoltaic strings are removed, the compensation strings are connected to the busbar through the switching switches.

[0131] As an example, the compensation string is further explained:

[0132] Figure 4 The PV strings 1 to 3 are connected in parallel, as well as compensation strings 1 and 2. When the PV strings are working normally, switches S(1, 1), S(1, 2), and S(1, 3) are closed, and the two compensation strings are in standby mode.

[0133] When a moderate abnormal aging fault occurs in the photovoltaic array, if the components in photovoltaic string 2 are abnormally aged, S(1, 2) is disconnected, and switch S(2, 1) or S(2, 2) is closed to connect compensation string 1 or compensation string 2; optionally, compensation string 1 and compensation string 2 can be replaced in turn to ensure stable output; at the same time, the two groups of compensation strings are replaced in turn, with similar usage frequency, extending the service life. When a severe abnormal aging fault occurs, if the aged components are located in photovoltaic string 2 and photovoltaic string 3, S(1, 2) and S(1, 3) are disconnected at the same time, and S(2, 1) and S(2, 2) are closed to make photovoltaic string 2 and photovoltaic string 3 stop running, and the two groups of compensation strings are respectively switched into the positions of photovoltaic string 2 and photovoltaic string 3 for operation.

[0134] Preferably, each compensation string can be controlled to be connected to the busbar through the switching switch in the following manner, including:

[0135] 1. When the photovoltaic strings are not aged, the duty cycle of the switching switches corresponding to each compensation string is 0;

[0136] 2. When the photovoltaic strings are slightly aged, the switching duty ratios of the compensation strings and the slightly aged photovoltaic strings are equal, and the compensation strings and the slightly aged photovoltaic strings are switched into the busbar in turn;

[0137] 3. When the PV strings are moderately aged or severely aged, the moderately aged or severely aged PV strings shall be removed; the compensation strings shall be alternately cut into the busbar, or all the compensation strings shall be cut into the busbar;

[0138] When each compensation string is switched into the busbar alternately, the switching duty ratio of each compensation string is determined according to the following formula:

[0139]

[0140] in, represents the switching duty cycle of compensation string i; represents the output power of compensation string i; represents the output power of compensation string j; n Indicates the number of compensation strings.

[0141] As an example, there are two compensation strings in this solution:

[0142] During normal operation: the switch duty ratios of the two compensation strings are ;

[0143] When one PV string is slightly aged, the slightly aged PV string continues to be used, alternating with the two compensation strings. The switching duty ratios of the slightly aged PV string and the two compensation strings are , can slow down the aging speed of the strings;

[0144] When one PV string is moderately or severely aged, the moderately or severely aged PV string is removed and two compensation strings are used alternately. The strategy is: monitor the voltage and current parameters of each compensation string every hour, dynamically calculate the output power of each compensation string in real time, determine the switch duty cycle of the corresponding compensation string based on the output power of the compensation string, adaptively adjust the use time of the compensation string, and the switch duty cycle of compensation string 1 , compensate the switching duty cycle of string 1 , ensuring that the usage frequency of the compensation strings is similar, balancing the differences in the consumption of the compensation strings, improving the overall string output consistency of the system, and enhancing the economic efficiency of the operation and maintenance of the compensation strings.

[0145] When two PV strings are moderately or severely aged at the same time, the moderately or severely aged PV strings are removed, and the switch duty ratios of the PV strings corresponding to the two compensation strings are both .

[0146] The following is a specific example to illustrate the multi-level aging identification and processing method for photovoltaic arrays.

[0147] Step 1: In the constructed 3×3 photovoltaic array simulation model, the photovoltaic string parameters are: maximum power P m =298.34W, maximum power point current I m =8.38A, maximum power point voltage U m =35.62V, short-circuit current of PV string I sc =8.85A, the open circuit voltage of the components in the photovoltaic string U oc =44.96V.

[0148] Step 2: Set the simulated resistance R of the first simulated photovoltaic string s1 The simulated resistance R of the first simulated photovoltaic string is set to 10Ω. The simulated resistance R of the second simulated photovoltaic string and the third simulated photovoltaic string are not connected in series. The switches S7, S2, and S3 are closed to simulate a mild abnormal aging fault of the photovoltaic array. s1 The simulated resistance R of the first simulated photovoltaic string is set to 20Ω, and the switches S7, S2, and S3 are closed. The simulated resistance R of the second simulated photovoltaic string and the third simulated photovoltaic string are not connected in series. s1 is 10Ω, and the simulated resistance R s2 The third simulated PV string has no simulated resistance in series, and switches S3, S5, and S7 are closed to simulate a severe abnormal aging fault of the PV array.

[0149] Setting the solar radiation intensity S =700 W / m 2 , Ambient temperature T =25℃, the simulation obtains the output characteristics of the photovoltaic array UI and UP under four conditions: normal, mildly abnormal, moderately abnormal and severely abnormal. Figure 5 As shown. S =700 W / m 2 , T =25℃, based on the experimental test sampling of photovoltaic panels, photovoltaic IV testers, illuminance meters, etc., the photovoltaic array UI and UP output characteristics are obtained under four conditions: normal, slightly abnormal, moderately abnormal, and severely abnormal. Figure 6 In this step, based on the simulated PV array UI and UP output characteristics, the simulated voltage change rate and the simulated current change rate of the maximum power point are calculated; based on the measured PV array UI and UP output characteristics, the measured voltage change rate and the measured current change rate of the maximum power point are calculated, combined with the simulated aging abnormality and the measured aging abnormality, as fault characteristic data.

[0150] Step 3: Set the irradiance range to 0~1200W / m 2 , the photovoltaic array was in four states: normal, slightly abnormal, moderately abnormal and severely abnormal, and 150 sets of samples were obtained for each state, totaling 600 sets.

[0151] DACS-SVM parameter settings: set the number of bird nests to 25, the upper limit of iterations to 250, the lower limit of the fitness function to 0.05, the upper limit to 0.1, and SVM uses the RBF kernel function.

[0152] Traditional SVM model parameter setting: Kernel function selects radial basis function, penalty parameter C=20.0, kernel parameters γ =0.05.

[0153] The fault feature data is divided into training samples and test samples in a 2:1 ratio, and Δ U m ’ , Δ I m ’ , input into the traditional SVM model and DACS-SVM model, and compare and analyze the fault diagnosis results.

[0154] Based on the DACS-SVM fault identification model, the different fitness evolution curves obtained after the cycle end conditions are met are as follows: Figure 7 As shown, the optimal solution of output parameters is obtained. When the DACS-SVM model is at fitness 1, fitness 2, and fitness 3, the fitness function reaches the prediction error requirement after 205 iterations, 147 iterations, and 28 iterations respectively. Therefore, by searching the SVM parameters through the DACS algorithm, it can be obtained that when the RBF kernel function is used, the penalty factor C and kernel function parameters γ They are fitness 3 corresponding to C =10.0, γ =0.01 when the model classification accuracy is the highest.

[0155] The fault diagnosis results of the traditional SVM and DACS-SVM algorithm models are shown in Figure 8 and Fig. 9 As shown. Based on the DACS-SVM model, the preprocessed fault sample data is input into the model for training and learning. After model verification, the fault diagnosis accuracy of the traditional SVM sample training set and test set are 94.53% and 93.16% respectively; the fault diagnosis accuracy of the DACS-SVM sample training set and test set are 99.44% and 98.89% respectively, taking into account both diagnostic accuracy and convergence speed, and the fault identification effect is better.

[0156] like Figure 4 As shown in the figure, the designed photovoltaic array has an abnormal aging fault in photovoltaic string 2, and one component in photovoltaic string 2 has abnormal aging, causing the maximum power point voltage to drop by 5%~10%. At this time, switch S(1, 2) is disconnected, S(2, 1) is closed, the original photovoltaic string 2 is cut off, and compensation string 1 or compensation string 2 is replaced in rotation. Since photovoltaic strings 1 and 3 have no faults, switches S(1,1) and S(1, 3) are also closed.

[0157] Fig.10 , Fig.11 and Fig.12 for S =1000 W / m 2 ,T =25℃, the output characteristics of the photovoltaic array before and after reconstruction under mild, moderate and severe abnormalities.

[0158] When the photovoltaic array has a mild abnormal aging fault, the system power generation efficiency is slightly affected, and the photovoltaic strings with aging components operate for a short time; when the photovoltaic array has a moderate or severe abnormal aging fault, the switch is driven to operate, and the compensation strings are switched in turn or fixedly to make up for the output during the replacement of the aging components; after the aging components are replaced, the compensation strings are cut out. Directly cutting off the aged photovoltaic strings without taking any treatment methods will greatly reduce the output characteristics. When the photovoltaic array continues to operate with mild abnormal, moderate abnormal and severe abnormal aging, the maximum output power of the photovoltaic array is 315.84 W, 281.33 W, and 245.06 W, respectively, and the working efficiency is 90.17%, 80.32%, and 69.96%, respectively. After the local reconstruction fault processing, the maximum output power of the photovoltaic array is 350.27 W, and the working efficiency is increased by 9.83%, 19.68%, and 30.04%, respectively. The effectiveness of the local reconstruction of the switch array is verified.

[0159] Example 2

[0160] like Fig.13 As shown, based on the same inventive concept as the above embodiment, the present invention also provides a photovoltaic array multi-level aging fault identification device, comprising:

[0161] A data acquisition module is used to obtain the voltage change rate and current change rate of the maximum power point of each photovoltaic string in the photovoltaic array;

[0162] The abnormal identification module is used to input the voltage change rate and current change rate of the maximum power point of each photovoltaic string into a pre-trained fault identification model, and the fault identification model outputs the abnormal aging degree of each photovoltaic string; operation and maintenance are performed according to the abnormal aging degree of the photovoltaic array; wherein the fault identification model is obtained based on support vector machine training; during the training process of the fault identification model, the penalty factor and kernel function parameters of the support vector machine are optimized using a dynamic adaptive cuckoo search algorithm; the training set data of the fault identification model is obtained based on a photovoltaic array simulation model, and the photovoltaic array simulation model is provided with a simulated resistor for simulating the abnormal aging degree of the photovoltaic array.

[0163] Example 3

[0164] like Fig.14 As shown, the present invention also provides an electronic device 100 for implementing a photovoltaic array multi-level aging identification and processing method;

[0165] The electronic device 100 includes a memory 101 , at least one processor 102 , a computer program 103 stored in the memory 101 and executable on the at least one processor 102 , and at least one communication bus 104 .

[0166] The memory 101 can be used to store a computer program 103 . The processor 102 implements the steps of a photovoltaic array multi-level aging identification processing method in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101 .

[0167] The memory 101 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 may include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0168] At least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.

[0169] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a photovoltaic array multi-level aging identification and processing method, and the processor 102 can execute the plurality of instructions to implement:

[0170] Obtain the voltage change rate and current change rate of the maximum power point of each photovoltaic string in the photovoltaic array;

[0171] The voltage change rate and current change rate of the maximum power point of each photovoltaic string are input into a pre-trained fault identification model, and the fault identification model outputs the abnormal aging degree of each photovoltaic string; operation and maintenance are performed according to the abnormal aging degree of the photovoltaic array; wherein, the fault identification model is obtained based on support vector machine training; during the training process of the fault identification model, a dynamic adaptive cuckoo search algorithm is used to optimize the penalty factor and kernel function parameters of the support vector machine; the training set data of the fault identification model is obtained based on a photovoltaic array simulation model, and the photovoltaic array simulation model is provided with a simulated resistor for simulating the abnormal aging degree of the photovoltaic array.

[0172] Example 4

[0173] If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0174] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0175] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0176] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0177] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0178] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A photovoltaic array multi-level aging identification and processing method, characterized in that: The steps include: Obtain the voltage change rate and current change rate of the maximum power point of the photovoltaic array; Inputting the voltage change rate and the current change rate of the maximum power point of the photovoltaic array into a pre-trained first fault identification model, the first fault identification model outputs the abnormal aging degree of the photovoltaic array; Operation and maintenance processing is performed according to the abnormal aging degree of the photovoltaic array; wherein the first fault identification model is obtained based on support vector machine training; during the training process of the first fault identification model, a dynamic adaptive cuckoo search algorithm is used to optimize the penalty factor and kernel function parameters of the support vector machine; the training set data of the first fault identification model is obtained based on a photovoltaic array simulation model, and the photovoltaic array simulation model is provided with a simulated resistor for simulating the abnormal aging degree of the photovoltaic array; The abnormal aging degree of PV array includes normal, mildly abnormal, moderately abnormal and severely abnormal; Operation and maintenance are performed according to the abnormal aging degree of the photovoltaic array, including: when the photovoltaic array is slightly abnormal, moderately abnormal or severely abnormal, the voltage change rate and current change rate of the maximum power point of each photovoltaic string are input into the pre-trained second fault recognition model to obtain the abnormal aging degree of each photovoltaic string; the abnormal aging degree of each photovoltaic string includes no aging, slight aging, moderate aging and severe aging; a compensation string is set in parallel with the photovoltaic string, and the compensation string is connected to the busbar through the switching switch; when the photovoltaic string is not aged, the compensation string of each compensation string is connected to the busbar through the switching switch; The corresponding switching duty ratios are all 0; when the photovoltaic strings are slightly aged, the switching duty ratios corresponding to the compensation strings and the slightly aged photovoltaic strings are equal, and the compensation strings and the slightly aged photovoltaic strings are switched into the busbar in turn; when the photovoltaic strings are moderately aged or severely aged, the moderately aged or severely aged photovoltaic strings are removed; the compensation strings are switched into the busbar alternately, or all the compensation strings are switched into the busbar; when the compensation strings are switched into the busbar alternately, the switching duty ratios of the compensation strings are determined according to the following formula: in, represents the switching duty cycle of compensation string i; represents the output power of compensation string i; represents the output power of compensation string j; n Indicates the number of compensation strings.

2. The photovoltaic array multi-level aging identification and processing method according to claim 1 is characterized in that: In the step of inputting the voltage change rate and the current change rate of the maximum power point of the photovoltaic array into a pre-trained first fault identification model, and the first fault identification model outputting the abnormal aging degree of the photovoltaic array, the first fault identification model is trained in the following manner: Acquire training set data for training a first fault identification model; Build a support vector machine model and initialize the penalty factor and kernel function parameters of the support vector machine model; Based on the dynamic adaptive cuckoo search algorithm, M bird nests are randomly generated, and each bird nest represents a set of penalty factors and kernel function parameters; Using the training set data, the fitness value corresponding to each bird's nest is calculated according to the fitness function; wherein the fitness function is defined as the average value of the sum of the absolute values ​​of the errors between the diagnosis results of the support vector machine model and the true results; Update the nest position according to Levy flight and nest host mechanism; calculate the fitness value of the updated nest, update the current optimal fitness and optimal nest position; determine whether the iteration termination condition is met; when the iteration terminates, output the global optimal fitness value and the corresponding optimal nest position; take the penalty factor and kernel function parameters corresponding to the optimal nest position as the optimal parameters, and substitute the optimal parameters into the support vector machine model.

3. The photovoltaic array multi-level aging identification and processing method according to claim 2 is characterized in that: Acquiring training set data for training the first fault identification model includes: A photovoltaic array simulation model is constructed in a simulation platform; wherein the photovoltaic array simulation model includes a plurality of simulated photovoltaic strings and a plurality of simulated collection busbars; each simulated photovoltaic string is provided with switches corresponding to the number of simulated collection busbars, which are used to connect the simulated collection busbars respectively; no simulated resistor is provided on one of the simulated collection busbars, and corresponding simulated resistors are provided in series on the remaining simulated collection busbars, and the simulated resistors are used to simulate different aging degrees of the simulated photovoltaic strings; The switches on the simulated photovoltaic strings are closed respectively with different simulated collecting buses to obtain the grid connection point voltage and grid connection point current corresponding to the photovoltaic array simulation model under different simulated resistance conditions; Calculate the voltage change rate and current change rate at the maximum power point of the photovoltaic array simulation model according to the grid connection point voltage and grid connection point current; determine the aging abnormality degree of the photovoltaic array simulation model according to the voltage change rate and a preset first threshold value; The voltage change rate and current change rate at the maximum power point of the photovoltaic array simulation model, as well as the corresponding aging abnormality degree, are used as training set data.

4. The photovoltaic array multi-level aging identification and processing method according to claim 3 is characterized in that: Updated nest locations based on Levi flight and nest host mechanisms, including: In the formula, t is the current iteration number; For the t At the iteration, i The location information of the bird's nest; For the t+ In the first iteration, i The location information of the bird's nest; M is the total number of nests; λ is the Levy index; L ( λ ) is the random search path that satisfies the Levy flight; α represents the step length control amount; In the formula, μ and ν represents random numbers that follow a normal distribution; The step size control amount α is determined as follows: In the formula, d , e , f is the step size control factor, d > e > f ; N The maximum number of iterations to terminate; t is the current iteration number.

5. A photovoltaic array multi-level aging fault identification device, used to implement the photovoltaic array multi-level aging identification processing method according to claim 1, characterized in that: include: A data acquisition module, used to obtain the voltage change rate and current change rate of the maximum power point of the photovoltaic array; An abnormality identification module, used for inputting the voltage change rate and the current change rate of the maximum power point of the photovoltaic array into a pre-trained first fault identification model, and the first fault identification model outputs the abnormal aging degree of the photovoltaic array; Operation and maintenance are performed according to the abnormal aging degree of the photovoltaic array; wherein the first fault identification model is obtained based on support vector machine training; During the training process of the first fault identification model, a dynamic adaptive cuckoo search algorithm is used to optimize the penalty factor and kernel function parameters of the support vector machine; the training set data of the first fault identification model is obtained based on a photovoltaic array simulation model, and the photovoltaic array simulation model is provided with a simulated resistor for simulating the abnormal aging degree of the photovoltaic array.

6. An electronic device, characterized in that: The invention comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the photovoltaic array multi-level aging identification and processing method as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by the processor, the photovoltaic array multi-level aging identification and processing method according to any one of claims 1 to 4 is implemented.

Images