A distributed photovoltaic intelligent monitoring method, system, device and medium

Abstract

The present invention belongs to the technical field of big data analysis and management of photovoltaic power operation and maintenance, and specifically relates to a distributed photovoltaic intelligent monitoring method, system, equipment and medium. Meteorological data and pre-processed photovoltaic operation data are obtained, and the pre-processed photovoltaic operation data include photovoltaic operation power generation and component temperature; the meteorological data, photovoltaic operation power generation and component temperature are input into a pre-trained prediction model to obtain a photovoltaic power generation prediction value; wherein the pre-trained prediction model is obtained based on the FA algorithm and the GAT algorithm. The method of the present invention takes into account photovoltaic operation data including photovoltaic operation power generation and component temperature, and predicts photovoltaic power generation in combination with real-time meteorological data, thereby improving the accuracy of photovoltaic power generation prediction; by combining the efficient FA algorithm and the highly accurate GAT algorithm, the efficiency and accuracy of power generation prediction are further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power operation and maintenance big data analysis and management, and specifically relates to a distributed photovoltaic intelligent monitoring method, system, equipment and medium. Background Art

[0002] With the continuous development of intelligent and efficient power industry, the rapid popularization of distributed photovoltaic power generation technology has triggered an urgent need for photovoltaic equipment status monitoring, output prediction and fault alarm diagnosis. The reliable operation of photovoltaic power stations requires photovoltaic components, inverters and other equipment to operate in a fault-free state, and the status monitoring of photovoltaic equipment is very important. At present, distributed photovoltaic monitoring-related software often faces challenges in real-time and accuracy. Existing distributed photovoltaic power generation prediction models may produce large errors when facing volatile data, resulting in reduced accuracy of photovoltaic power generation prediction; at the same time, if the parameters in the prediction model are not selected properly, the solution may fall into the local optimum, which directly affects the prediction effect and prediction efficiency. Summary of the invention

[0003] The purpose of the present invention is to provide a distributed photovoltaic intelligent monitoring method, system, device and medium to solve the problem that the existing photovoltaic monitoring methods have low efficiency and accuracy in predicting photovoltaic power generation.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0005] In a first aspect, the present invention provides a distributed photovoltaic intelligent monitoring method, comprising the following steps:

[0006] Acquiring meteorological data and pre-processed photovoltaic operation data, wherein the pre-processed photovoltaic operation data includes photovoltaic operation power generation and component temperature;

[0007] Input meteorological data, photovoltaic power generation and component temperature into the pre-trained prediction model to obtain the photovoltaic power generation prediction value;

[0008] The pre-trained prediction model is obtained based on the FA algorithm and the GAT algorithm.

[0009] Based on the above technical solution, further, the pre-processed photovoltaic operation data is obtained by the following steps:

[0010] Obtain photovoltaic operation data;

[0011] The photovoltaic operation data are processed by outlier processing, missing value filling and data format normalization to obtain preprocessed photovoltaic operation data.

[0012] Based on the above technical solution, further, the pre-trained prediction model is obtained by the following steps:

[0013] The loss function MSE of photovoltaic output prediction is defined as the objective function;

[0014] Set FA algorithm parameters, including: number of flames, moving step length, and light absorption coefficient;

[0015] Define the nodes in the GAT algorithm graph as flames, and calculate the light intensity of the flames and the attraction between the flames;

[0016] In each iteration of the GAT algorithm, the flames in the image will move according to the attraction between the flames and the light intensity of the flames. The parameters of the GAT algorithm are updated according to the movement of the flames. When the MSE of the loss function of the photovoltaic output prediction is less than the set threshold, the optimal hyperparameter combination is obtained, and the prediction model is obtained.

[0017] Obtain training set data and train the prediction model to obtain a pre-trained prediction model.

[0018] Based on the above technical solution, further, the step of defining the nodes in the graph of the GAT algorithm as flames and calculating the light intensity of the flames and the attraction between the flames specifically includes:

[0019] The nodes in the graph of the GAT algorithm are defined as flames, and the Euclidean distance is used to represent the distance between two flames in the graph. The formula is as follows:

[0020]

[0021] Where: x a,k is the coordinate of flame a in dimension k in the figure; x b,k is the coordinate of the flame b in dimension k; n is the number of dimensions;

[0022] The relative light intensity of flame a to flame b is defined as follows:

[0023]

[0024] Among them: Brightness a is the absolute light intensity of flame a; γ is the light absorption coefficient; Distance ab is the distance between flame a and flame b;

[0025] The attractive force between flame a and flame b is as follows:

[0026]

[0027] Among them: Attraction0 is the attraction when the distance between the two flames is 0.

[0028] Based on the above technical solution, further, in each iteration of the GAT algorithm, according to the attraction between the flames in the image and the light intensity of the flames, the flames in the image will move, and the step of updating the parameters of the GAT algorithm according to the movement of the flames specifically includes:

[0029] Each flame represents a set of hyperparameter configurations of the GAT algorithm. According to the attraction between the flames and the light intensity of the flames, the flame with smaller light intensity will move towards the position of the flame with stronger light intensity. The flame position update formula is as follows:

[0030]

[0031] in: Respectively represent the positions of flame a and flame b at time t; represents the position of flame a at time t+1; α is the moving step size, which is used to control the step size of the flame movement; rand is a random number in the range of [0,1];

[0032] After the flame position is updated, the parameters of the GAT algorithm are updated accordingly.

[0033] Based on the above technical solution, further comprising:

[0034] The pre-processed photovoltaic operation data also includes the power generation of photovoltaic operation;

[0035] Obtain the photovoltaic power generation and the photovoltaic rated maximum power generation, calculate the ratio of photovoltaic power generation to photovoltaic rated maximum power generation, and obtain the photovoltaic power generation ratio;

[0036] The photovoltaic power generation identification result is obtained according to the proportion of photovoltaic operation power generation, and an alarm notification is issued according to the photovoltaic power generation identification result.

[0037] Based on the above technical solution, further, the step of obtaining a photovoltaic power generation identification result according to the photovoltaic operation power generation ratio and issuing an alarm notification according to the photovoltaic power generation identification result specifically includes:

[0038] When the photovoltaic power generation ratio is greater than or equal to 90% and less than or equal to 100%, the photovoltaic power generation judgment result is that the component is normal, and no alarm notification is issued at this time;

[0039] When the photovoltaic power generation accounts for 70% or more and less than 90%, the photovoltaic power generation judgment result is dust accumulation on the components, and the alarm level is level 1;

[0040] When the photovoltaic power generation accounts for 40% or more and less than 70%, the photovoltaic power generation judgment result is component shielding, and the alarm level is level 2;

[0041] When the photovoltaic power generation accounts for less than 40%, the photovoltaic power generation judgment result is component failure, and the alarm level is level 3;

[0042] The alarm notification includes an alarm level and a photovoltaic power generation determination result.

[0043] In a second aspect, the present invention provides a distributed photovoltaic intelligent monitoring system, comprising:

[0044] A data acquisition module, used to acquire meteorological data and pre-processed photovoltaic operation data, wherein the pre-processed photovoltaic operation data includes photovoltaic power generation and component temperature;

[0045] The power generation prediction module is used to input meteorological data, photovoltaic power generation and component temperature into a pre-trained prediction model to obtain the photovoltaic power generation prediction value;

[0046] The pre-trained prediction model is obtained based on the FA algorithm and the GAT algorithm.

[0047] In a third aspect, the present invention provides an electronic device, including a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the distributed photovoltaic intelligent monitoring method.

[0048] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the distributed photovoltaic intelligent monitoring method is implemented.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The present invention provides a distributed photovoltaic intelligent monitoring method, system, device and medium, the method comprising:

[0051] Acquiring meteorological data and pre-processed photovoltaic operation data, wherein the pre-processed photovoltaic operation data includes photovoltaic operation power generation and component temperature;

[0052] Input meteorological data, photovoltaic power generation and component temperature into the pre-trained prediction model to obtain the photovoltaic power generation prediction value;

[0053] The pre-trained prediction model is obtained based on the FA algorithm and the GAT algorithm.

[0054] The method of the present invention takes into account photovoltaic operation data including photovoltaic power generation and component temperature, and predicts photovoltaic power generation in combination with real-time meteorological data, thereby improving the accuracy of photovoltaic power generation prediction; by combining the efficient FA algorithm and the highly accurate GAT algorithm, the efficiency and accuracy of power generation prediction are further improved; therefore, the present invention can predict the power generation of photovoltaic equipment in real time, accurately and efficiently. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings constituting a part of the present invention 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 accompanying drawings:

[0056] Figure 1 This is a flow chart of a distributed photovoltaic intelligent monitoring method of the present invention;

[0057] Figure 2 This is a structural diagram of a distributed photovoltaic intelligent monitoring system of the present invention;

[0058] Figure 3 The present invention is a structural block diagram of an electronic device. DETAILED DESCRIPTION

[0059] 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 invention can be combined with each other without conflict.

[0060] 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 invention belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present invention.

[0061] Example 1

[0062] See also Figure 1 As shown, the present invention provides a distributed photovoltaic intelligent monitoring method, which specifically includes the following steps:

[0063] S1. Obtain meteorological data and pre-processed photovoltaic operation data

[0064] Obtain meteorological data such as total radiation, scattered radiation, direct radiation, air pressure, ambient temperature, etc. through the weather forecast website interface;

[0065] An intelligent data collection device is installed on the back of each photovoltaic panel to collect photovoltaic operation data and preprocess the photovoltaic operation data, including: data outlier processing, missing value filling and data format normalization, and finally the preprocessed photovoltaic operation data is obtained, including: photovoltaic operation power generation, power generation power, voltage, current, component temperature and other data.

[0066] S2. Input the meteorological data, photovoltaic power generation and component temperature into the pre-trained prediction model to obtain the photovoltaic power generation prediction value.

[0067] This embodiment uses the FA firefly algorithm to optimize the hyperparameters of the GAT algorithm model based on the attention mechanism graph neural network to obtain a pre-trained prediction model, specifically including:

[0068] The loss function MSE of photovoltaic output prediction is set as the objective function;

[0069] Set the parameters of the FA algorithm such as the number of flames, moving step length, light absorption coefficient, etc.

[0070] The nodes in the GAT algorithm graph are defined as flames, and the Euclidean distance is used to represent the distance between two flames in the graph. The formula is as follows:

[0071]

[0072] Where: x a,k is the coordinate of flame a in dimension k in the figure; x b,k is the coordinate of the flame b in dimension k; n is the number of dimensions;

[0073] The relative light intensity of flame a to flame b is defined as follows:

[0074]

[0075] Among them: Brightness a is the absolute light intensity of flame a; γ is the light absorption coefficient; Distance ab is the distance between flame a and flame b;

[0076] The attractive force formula between flame a and flame b in the figure is as follows:

[0077]

[0078] Among them: Attraction0 is the attraction when the distance between the two flames is 0;

[0079] According to the light intensity of the flame and the attraction between the two flames, the individual flame will move in the direction of the flame with greater light intensity than itself;

[0080] Each flame represents a set of hyperparameter configurations of the GAT algorithm, including learning rate, number of hidden layers, and number of attention heads. Different parameter values ​​represent the positions of different flames. In each iteration of the GAT algorithm, according to the attraction between the flames in the image and the light intensity of the flames, the flame with smaller light intensity will move towards the position of the flame with stronger light intensity. The flame position update formula is as follows:

[0081]

[0082] in: Respectively represent the positions of flame a and flame b at time t; represents the position of the flame a at time t+1; α is the moving step size, which is used to control the step size of the flame movement; rand is a random number in the range of [0,1], which helps the algorithm avoid falling into a local optimal solution;

[0083] The position of the flame is updated after it moves, so the parameters of the GAT algorithm are updated to minimize the error of the photovoltaic output prediction. When the loss function MSE of the photovoltaic output prediction is less than the set threshold or no longer changes significantly, the model is considered to have converged and the optimal hyperparameter combination is obtained, thus obtaining the prediction model;

[0084] The prediction model is trained using the training set data to ensure that the prediction model can learn the complex characteristics and trends of photovoltaic power generation, and finally obtain a pre-trained prediction model.

[0085] Based on the above embodiment, further, it also includes: obtaining the power generation of photovoltaic operation and the rated maximum power generation of photovoltaic, calculating the proportion of photovoltaic operation power generation, obtaining the photovoltaic power generation identification result, and issuing an alarm notification according to the photovoltaic power generation identification result, specifically including:

[0086] Get the power generation of photovoltaic operation and the rated maximum power generation of photovoltaic, calculate the ratio of the power generation of photovoltaic operation to the rated maximum power generation of photovoltaic, and obtain the proportion of the power generation of photovoltaic operation:

[0087] When the photovoltaic power generation ratio is greater than or equal to 90% and less than or equal to 100%, the photovoltaic power generation judgment result is that the component is normal, and no alarm notification is issued at this time;

[0088] When the photovoltaic power generation accounts for 70% or more and less than 90%, the photovoltaic power generation judgment result is dust accumulation on the components, and the alarm level is level 1;

[0089] When the photovoltaic power generation accounts for 40% or more and less than 70%, the photovoltaic power generation judgment result is component shielding, and the alarm level is level 2;

[0090] When the photovoltaic power generation accounts for less than 40%, the photovoltaic power generation judgment result is component failure, and the alarm level is level 3;

[0091] The content of the alarm notification includes: power station name, alarm equipment, equipment number, alarm level, alarm type, photovoltaic power generation identification result, occurrence time, and display the alarm notification. The alarm notification is saved to the database, and the alarm result can be exported.

[0092] Based on the above embodiment, it further includes: visual display of pre-processed photovoltaic operation data, real-time monitoring and viewing of photovoltaic operation power generation, power generation power, voltage, current, component temperature and other data, and exporting to csv format file by time filtering for local viewing and storage.

[0093] Based on the above embodiment, further, it also includes: obtaining a photovoltaic power generation power prediction value and a corresponding photovoltaic power generation power actual value, and visually displaying a photovoltaic power generation power prediction value curve and a photovoltaic power generation power actual value curve.

[0094] Based on the above embodiment, further, it also includes using MSE, MAE and R 2 Evaluate the prediction effect of the pre-trained prediction model and visualize the evaluation results, including:

[0095] When using MSE to evaluate the prediction effect of the pre-trained prediction model, the predicted value of photovoltaic power generation and the actual value of photovoltaic power generation are obtained and the root mean square error is calculated. The formula is as follows:

[0096]

[0097] Where: m is the number of samples, y is i represents the true value of the ith photovoltaic power generation, Represents the predicted value of the ith photovoltaic power generation; the smaller the MSE value of the evaluation result, the more accurate the prediction effect of the pre-trained prediction model;

[0098] When using MAE to evaluate the prediction effect of the pre-trained prediction model, the predicted value of photovoltaic power generation and the actual value of photovoltaic power generation are obtained and the mean absolute error is calculated. The formula is as follows:

[0099]

[0100] The smaller the MAE value of the evaluation result, the more accurate the prediction effect of the pre-trained prediction model is;

[0101] Using R 2When evaluating the prediction effect of the pre-trained prediction model, the predicted value of photovoltaic power generation, the actual value of photovoltaic power generation, and the average value of the actual value of photovoltaic power generation are obtained and the determination coefficient is calculated. The formula is as follows:

[0102]

[0103] in: is the average value of the real value of photovoltaic power generation;

[0104] Evaluation results R 2 When the value is 1, it means that the pre-trained prediction model fits the data perfectly, and the predicted value is exactly the same as the true value;

[0105] Evaluation results R 2 When the value of is between 0 and 1, the closer it is to 1, the more accurate the prediction effect of the pre-trained prediction model is;

[0106] Evaluation results R 2 When the value of is 0, it means that the prediction effect of the pre-trained prediction model is the same as the ordinary average effect;

[0107] Evaluation results R 2 When the value of is less than 0, it means that the fitting effect of the pre-trained prediction model is poor, which is lower than the normal average effect.

[0108] Example 2

[0109] See also Figure 2 As shown, the present invention provides a distributed photovoltaic intelligent monitoring system, which specifically includes:

[0110] The data acquisition module is used to acquire meteorological data and pre-processed photovoltaic operation data.

[0111] An intelligent data collection device is installed on the back of each photovoltaic panel to collect photovoltaic operation data. A set of master controllers and wireless routers are installed in each photovoltaic station. The photovoltaic operation data is sent to a distributed photovoltaic intelligent monitoring system on the main server via 4G / 5G wireless transmission. After receiving the photovoltaic operation data, the system pre-processes the photovoltaic operation data, including: data outlier processing, missing value filling and data format normalization, and finally obtains pre-processed photovoltaic operation data, including: photovoltaic operation power generation, power generation power, voltage, current, component temperature and other data; the pre-processed photovoltaic operation data is stored in the database;

[0112] Obtain meteorological data such as total radiation, scattered radiation, direct radiation, air pressure, ambient temperature, etc. through the weather forecast website interface.

[0113] The power generation prediction module is used to input meteorological data, photovoltaic power generation and component temperature into a pre-trained prediction model to obtain the photovoltaic power generation prediction value;

[0114] This embodiment uses the FA firefly algorithm to optimize the hyperparameters of the GAT algorithm model based on the attention mechanism graph neural network to obtain a pre-trained prediction model, specifically including:

[0115] The loss function MSE of photovoltaic output prediction is set as the objective function;

[0116] Set the parameters of the FA algorithm such as the number of flames, moving step length, light absorption coefficient, etc.

[0117] The nodes in the GAT algorithm graph are defined as flames, and the Euclidean distance is used to represent the distance between two flames in the graph. The formula is as follows:

[0118]

[0119] Where: x a,k is the coordinate of flame a in dimension k in the figure; x b,k is the coordinate of the flame b in dimension k; n is the number of dimensions;

[0120] The relative light intensity of flame a to flame b is defined as follows:

[0121]

[0122] Among them: Brightness a is the absolute light intensity of flame a; γ is the light absorption coefficient; Distance ab is the distance between flame a and flame b;

[0123] The attractive force formula between flame a and flame b in the figure is as follows:

[0124]

[0125] Among them: Attraction0 is the attraction when the distance between the two flames is 0;

[0126] According to the light intensity of the flame and the attraction between the two flames, the individual flame will move in the direction of the flame with greater light intensity than itself;

[0127] Each flame represents a set of hyperparameter configurations of the GAT algorithm, including learning rate, number of hidden layers, and number of attention heads. Different parameter values ​​represent the positions of different flames. In each iteration of the GAT algorithm, according to the attraction between the flames in the image and the light intensity of the flames, the flame with smaller light intensity will move towards the position of the flame with stronger light intensity. The flame position update formula is as follows:

[0128]

[0129] in: Respectively represent the positions of flame a and flame b at time t; represents the position of the flame a at time t+1; α is the moving step size, which is used to control the step size of the flame movement; rand is a random number in the range of [0,1], which helps the algorithm avoid falling into a local optimal solution;

[0130] The position of the flame is updated after it moves, so the parameters of the GAT algorithm are updated to minimize the error of the photovoltaic output prediction. When the loss function MSE of the photovoltaic output prediction is less than the set threshold or no longer changes significantly, the model is considered to have converged and the optimal hyperparameter combination is obtained, thus obtaining the prediction model;

[0131] The prediction model is trained using the training set data to ensure that the prediction model can learn the complex characteristics and trends of photovoltaic power generation, and finally obtain a pre-trained prediction model.

[0132] Based on the above embodiment, further, an alarm notification module is included, which is used to obtain the power generation of photovoltaic operation and the rated maximum power generation of photovoltaic, calculate the proportion of photovoltaic operation power generation, obtain the photovoltaic power generation identification result, and issue an alarm notification according to the photovoltaic power generation identification result, specifically including:

[0133] Get the power generation of photovoltaic operation and the rated maximum power generation of photovoltaic, calculate the ratio of the power generation of photovoltaic operation to the rated maximum power generation of photovoltaic, and obtain the proportion of the power generation of photovoltaic operation:

[0134] When the photovoltaic power generation ratio is greater than or equal to 90% and less than or equal to 100%, the photovoltaic power generation judgment result is that the component is normal, and no alarm notification is issued at this time;

[0135] When the photovoltaic power generation accounts for 70% or more and less than 90%, the photovoltaic power generation judgment result is dust accumulation on the components, and the alarm level is level 1;

[0136] When the photovoltaic power generation accounts for 40% or more and less than 70%, the photovoltaic power generation judgment result is component shielding, and the alarm level is level 2;

[0137] When the photovoltaic power generation accounts for less than 40%, the photovoltaic power generation judgment result is component failure, and the alarm level is level 3;

[0138] The content of the alarm notification includes: power station name, alarm equipment, equipment number, alarm level, alarm type, photovoltaic power generation identification result, occurrence time, and display the alarm notification. The alarm notification is saved to the database, and the alarm result can be exported.

[0139] Based on the above embodiment, further, it also includes a real-time monitoring module for visually displaying the pre-processed photovoltaic operation data, which can monitor and view the photovoltaic operation power generation, power generation power, voltage, current, component temperature and other data in real time, and can filter and export them to csv format files by time for local viewing and storage.

[0140] Based on the above embodiment, further, the power generation prediction module also includes obtaining the photovoltaic power generation power prediction value and the corresponding photovoltaic power generation power actual value, and visually displaying the photovoltaic power generation power prediction value curve and the photovoltaic power generation power actual value curve.

[0141] Based on the above embodiment, further, the power generation prediction module also includes using MSE, MAE and R 2 Evaluate the prediction effect of the pre-trained prediction model and visualize the evaluation results, including:

[0142] When using MSE to evaluate the prediction effect of the pre-trained prediction model, the predicted value of photovoltaic power generation and the actual value of photovoltaic power generation are obtained and the root mean square error is calculated. The formula is as follows:

[0143]

[0144] Where: m is the number of samples, y is i represents the true value of the ith photovoltaic power generation, Represents the predicted value of the ith photovoltaic power generation; the smaller the MSE value of the evaluation result, the more accurate the prediction effect of the pre-trained prediction model;

[0145] When using MAE to evaluate the prediction effect of the pre-trained prediction model, the predicted value of photovoltaic power generation and the actual value of photovoltaic power generation are obtained and the mean absolute error is calculated. The formula is as follows:

[0146]

[0147] The smaller the MAE value of the evaluation result, the more accurate the prediction effect of the pre-trained prediction model is;

[0148] Using R 2 When evaluating the prediction effect of the pre-trained prediction model, the predicted value of photovoltaic power generation, the actual value of photovoltaic power generation, and the average value of the actual value of photovoltaic power generation are obtained and the determination coefficient is calculated. The formula is as follows:

[0149]

[0150] in: is the average value of the real value of photovoltaic power generation;

[0151] Evaluation results R 2 When the value is 1, it means that the pre-trained prediction model fits the data perfectly, and the predicted value is exactly the same as the true value;

[0152] Evaluation results R 2 When the value of is between 0 and 1, the closer it is to 1, the more accurate the prediction effect of the pre-trained prediction model is;

[0153] Evaluation results R 2 When the value is 0, it means that the prediction effect of the pre-trained prediction model is the same as the ordinary average effect;

[0154] Evaluation results R 2 When the value of is less than 0, it means that the fitting effect of the pre-trained prediction model is poor, which is lower than the normal average effect.

[0155] Example 3

[0156] See also Figure 3 As shown, the present invention also provides an electronic device 100 for implementing a distributed photovoltaic intelligent monitoring method; 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.

[0157] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of a distributed photovoltaic intelligent monitoring method described in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area, wherein the program storage area can 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 can store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 can 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.

[0158] The 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.

[0159] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a distributed photovoltaic intelligent monitoring method, and the processor 102 can execute the plurality of instructions to implement:

[0160] Acquiring meteorological data and pre-processed photovoltaic operation data, wherein the pre-processed photovoltaic operation data includes photovoltaic operation power generation and component temperature;

[0161] Input meteorological data, photovoltaic power generation and component temperature into the pre-trained prediction model to obtain the photovoltaic power generation prediction value;

[0162] The pre-trained prediction model is obtained based on the FA algorithm and the GAT algorithm.

[0163] Example 4

[0164] 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 process 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. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0165] Those skilled in the art will appreciate 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. Moreover, 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.

[0166] 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 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0167] 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 including 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.

[0168] 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.

[0169] 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 distributed photovoltaic intelligent monitoring method, characterized in that: The following steps are involved: Acquiring meteorological data and pre-processed photovoltaic operation data, wherein the pre-processed photovoltaic operation data includes photovoltaic operation power generation and component temperature; Input meteorological data, photovoltaic power generation and component temperature into the pre-trained prediction model to obtain the photovoltaic power generation prediction value; Wherein, the pre-trained prediction model is obtained based on the FA algorithm and the GAT algorithm; The pre-trained prediction model is obtained by the following steps: The loss function MSE of photovoltaic output prediction is defined as the objective function; Set FA algorithm parameters, including: number of flames, moving step length, and light absorption coefficient; Define the nodes in the GAT algorithm graph as flames, and calculate the light intensity of the flames and the attraction between the flames; In each iteration of the GAT algorithm, the flames in the image will move according to the attraction between the flames and the light intensity of the flames. The parameters of the GAT algorithm are updated according to the movement of the flames. When the MSE of the loss function of the photovoltaic output prediction is less than the set threshold, the optimal hyperparameter combination is obtained, and the prediction model is obtained. Obtain training set data and train the prediction model to obtain a pre-trained prediction model; The steps of defining the nodes in the graph of the GAT algorithm as flames and calculating the light intensity of the flames and the attraction between the flames specifically include: The nodes in the graph of the GAT algorithm are defined as flames, and the Euclidean distance is used to represent the distance between two flames in the graph. The formula is as follows: Where: x a,k is the coordinate of flame a in dimension k in the figure; x b,k is the coordinate of the flame b in dimension k; n is the number of dimensions; The relative light intensity of flame a to flame b is defined as follows: Among them: Brightness a is the absolute light intensity of flame a; γ is the light absorption coefficient; Distance ab is the distance between flame a and flame b; The attractive force between flame a and flame b is as follows: Among them: Attraction0 is the attraction when the distance between the two flames is 0; In each iteration of the GAT algorithm, the flames in the image will move according to the attraction between the flames and the light intensity of the flames, and the step of updating the parameters of the GAT algorithm according to the movement of the flames specifically includes: Each flame represents a set of hyperparameter configurations of the GAT algorithm. According to the attraction between the flames and the light intensity of the flames, the flame with smaller light intensity will move towards the position of the flame with stronger light intensity. The flame position update formula is as follows: in: Respectively represent the positions of flame a and flame b at time t; represents the position of flame a at time t+1; α is the moving step size, which is used to control the step size of the flame movement; rand is a random number in the range of [0,1]; After the flame position is updated, the parameters of the GAT algorithm are updated accordingly.

2. A distributed photovoltaic intelligent monitoring method according to claim 1, characterized in that: The preprocessed photovoltaic operation data is obtained by the following steps: Obtain photovoltaic operation data; The photovoltaic operation data are processed by outlier processing, missing value filling and data format normalization to obtain preprocessed photovoltaic operation data.

3. A distributed photovoltaic intelligent monitoring method according to claim 1, characterized in that: Also includes: The pre-processed photovoltaic operation data also includes the power generation of photovoltaic operation; Obtain the photovoltaic power generation and the photovoltaic rated maximum power generation, calculate the ratio of photovoltaic power generation to photovoltaic rated maximum power generation, and obtain the photovoltaic power generation ratio; The photovoltaic power generation identification result is obtained according to the proportion of photovoltaic operation power generation, and an alarm notification is issued according to the photovoltaic power generation identification result.

4. A distributed photovoltaic intelligent monitoring method according to claim 3, characterized in that: The step of obtaining a photovoltaic power generation identification result according to the photovoltaic operation power generation ratio and issuing an alarm notification according to the photovoltaic power generation identification result specifically includes: When the photovoltaic power generation ratio is greater than or equal to 90% and less than or equal to 100%, the photovoltaic power generation judgment result is that the component is normal, and no alarm notification is issued at this time; When the photovoltaic power generation accounts for 70% or more and less than 90%, the photovoltaic power generation judgment result is dust accumulation on the components, and the alarm level is level 1; When the photovoltaic power generation accounts for 40% or more and less than 70%, the photovoltaic power generation judgment result is component shielding, and the alarm level is level 2; When the photovoltaic power generation accounts for less than 40%, the photovoltaic power generation judgment result is component failure, and the alarm level is level 3; The alarm notification includes an alarm level and a photovoltaic power generation determination result.

5. A distributed photovoltaic intelligent monitoring system, characterized in that: include: A data acquisition module, used to acquire meteorological data and pre-processed photovoltaic operation data, wherein the pre-processed photovoltaic operation data includes photovoltaic power generation and component temperature; The power generation prediction module is used to input meteorological data, photovoltaic power generation and component temperature into a pre-trained prediction model to obtain the photovoltaic power generation prediction value; Wherein, the pre-trained prediction model is obtained based on the FA algorithm and the GAT algorithm; The pre-trained prediction model is obtained by the following steps: The loss function MSE of photovoltaic output prediction is defined as the objective function; Set FA algorithm parameters, including: number of flames, moving step length, and light absorption coefficient; Define the nodes in the GAT algorithm graph as flames, and calculate the light intensity of the flames and the attraction between the flames; In each iteration of the GAT algorithm, the flames in the image will move according to the attraction between the flames and the light intensity of the flames. The parameters of the GAT algorithm are updated according to the movement of the flames. When the MSE of the loss function of the photovoltaic output prediction is less than the set threshold, the optimal hyperparameter combination is obtained, and the prediction model is obtained. Obtain training set data and train the prediction model to obtain a pre-trained prediction model; The steps of defining the nodes in the graph of the GAT algorithm as flames and calculating the light intensity of the flames and the attraction between the flames specifically include: The nodes in the graph of the GAT algorithm are defined as flames, and the Euclidean distance is used to represent the distance between two flames in the graph. The formula is as follows: Where: x a,k is the coordinate of flame a in dimension k in the figure; x b,k is the coordinate of the flame b in dimension k; n is the number of dimensions; The relative light intensity of flame a to flame b is defined as follows: Among them: Brightness a is the absolute light intensity of flame a; γ is the light absorption coefficient; Distance ab is the distance between flame a and flame b; The attractive force between flame a and flame b is as follows: Among them: Attraction0 is the attraction when the distance between the two flames is 0; In each iteration of the GAT algorithm, the flames in the image will move according to the attraction between the flames and the light intensity of the flames, and the step of updating the parameters of the GAT algorithm according to the movement of the flames specifically includes: Each flame represents a set of hyperparameter configurations of the GAT algorithm. According to the attraction between the flames and the light intensity of the flames, the flame with smaller light intensity will move towards the position of the flame with stronger light intensity. The flame position update formula is as follows: in: Respectively represent the positions of flame a and flame b at time t; represents the position of flame a at time t+1; α is the moving step size, which is used to control the step size of the flame movement; rand is a random number in the range of [0,1]; After the flame position is updated, the parameters of the GAT algorithm are updated accordingly.

6. An electronic device, characterized in that: It comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement a distributed photovoltaic intelligent monitoring method as described 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, a distributed photovoltaic intelligent monitoring method as described in any one of claims 1 to 4 is implemented.

Images