Regional photovoltaic power prediction method, system, equipment and medium

Abstract

The invention relates to the technical field of power prediction, in particular to a region-level photovoltaic power prediction method, system and device and a medium, and the method comprises the steps: obtaining the meteorological data of a photovoltaic module region and the operation data of each photovoltaic power station in real time; and inputting the meteorological data of the photovoltaic module area and the operation data of each photovoltaic power station into the trained prediction model for calculating the scattering and absorption of the solar radiation by the atmosphere according to the input meteorological data, estimating the effective solar radiation amount reaching the ground, predicting the photovoltaic power according to the operation data of each photovoltaic power station, and predicting the photovoltaic power according to the operation data of each photovoltaic power station. Obtaining a power prediction result; the prediction model adopts a generative adversarial network model. The method can improve the precision of power prediction.

Description

Technical Field

[0001] The invention relates to the technical field of power prediction, and in particular to a method, system, equipment and medium for regional photovoltaic power prediction. Background Art

[0002] Photovoltaic power output has obvious intermittent fluctuation characteristics. Large-scale photovoltaic power access will have a certain impact on the safe and stable operation of the power grid. For photovoltaic bases, accurate regional-level centralized light power forecasting is one of the key technologies to solve this problem. In addition, with the gradual development of spot power trading, forecast accuracy will also become the basic guarantee for fulfilling trading commitments.

[0003] At present, the conventional light resource platform of photovoltaic power stations is the station-level irradiation station, which has several problems: first, there are too few points to reflect the actual situation of a large area. Generally, there is only horizontal irradiation data, and occasionally there is inclined irradiation data, but no direct irradiation and scattered irradiation data, so it is impossible to conduct accurate analysis of light resources. Most irradiance meters are not calibrated, and the data accuracy is not high. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a regional photovoltaic power prediction method, system, device and medium to address the technical problem of low photovoltaic data prediction accuracy in view of the deficiencies in the above-mentioned prior art.

[0005] The objective of the present invention is achieved by the following technical solutions: In a first aspect, the present invention provides a regional photovoltaic power prediction method, comprising: Obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station are input into the trained prediction model, which is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model. As a further improvement of the present invention, the prediction model includes a generation module and a discrimination module; The generation module is used to convert the input operation data and meteorological data into generated data with similar characteristics, perform preliminary feature extraction on the generated data, simulate the time characteristics and spatial characteristics of photovoltaic power through the convolution layer, and generate a predicted power value sequence through the output layer based on the time characteristics and spatial characteristics; The discrimination module is used to perform convolution and pooling operations on the input data, extract the features of the data, use the fully connected layer to perform classification judgment, and determine whether the input data is real photovoltaic power data or predicted data generated in the generation module.

[0006] As a further improvement of the present invention, the prediction model uses historical operation data and corresponding meteorological data as data sets, and the historical operation data and corresponding meteorological data are pre-processed data, and the pre-processing includes data cleaning and data normalization processing.

[0007] As a further improvement of the present invention, the training process of the prediction model specifically includes: Initialize the parameters of the generation module and the discrimination module; The generation module is fixed and the discrimination module is trained, specifically including: randomly extracting a batch of real photovoltaic power data and corresponding meteorological data from the data set as positive samples; randomly generating a batch of noise vectors from the prior distribution, inputting them into the generation module, generating a batch of false photovoltaic power data and corresponding meteorological data as negative samples; randomly generating a batch of noise vectors from the prior distribution, inputting them into the generation module, generating a batch of false photovoltaic power data and corresponding meteorological data as negative samples; optimizing the algorithm to update the parameters of the discrimination module; The discriminant module is fixed and the generation module is trained, specifically including: randomly generating a batch of noise vectors from the prior distribution, inputting them into the generator, generating a batch of false photovoltaic power data and corresponding meteorological data; inputting the generated false data into the discriminant module, calculating the loss function of the generation module; using the optimization algorithm to update the parameters of the generation module, so that the generation module can generate more realistic photovoltaic power data; When the loss function values ​​of the generation module and the discrimination module change less than the set threshold in several consecutive training rounds, the training is stopped.

[0008] As a further improvement of the present invention, the loss function corresponding to the generation module is:

[0009]

[0010] In the formula, is the adversarial loss function of the generation module, G is the generation module, D is the discrimination module, From the prior distribution The random noise sampled in To generate data for the module, The probability of the real data for the discriminant module to judge the generated data; Fit the loss function to the data, ) is the real data distribution, is the square of the Euclidean distance between the generated data and the true data.

[0011] As a further improvement of the present invention, the loss function corresponding to the discrimination module is:

[0012]

[0013] In the formula, To discriminate the module against the loss function, is the probability that the discriminant module judges that the generated module is real data, is the regularization loss function of the discriminant module, is the regularization coefficient, is the discriminant module parameter.

[0014] As a further improvement of the present invention, during the training process of the prediction model, the total loss function is:

[0015] In the formula, , , are weight coefficients, is the adversarial loss function of the generation module, Fitting a loss function to the data of the generative module, To discriminate the module against the loss function, is the regularization loss function of the discriminative module.

[0016] In a second aspect, the present invention provides a regional photovoltaic power prediction system, which is used to implement the above-mentioned regional photovoltaic power prediction method, including: Data acquisition module, used to obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The power prediction module is used to input the meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station into the trained prediction model, and is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model.

[0017] In a third aspect, the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, which, when executed by a computing device, enable the computing device to perform the above-mentioned regional photovoltaic power prediction method.

[0018] In a fourth aspect, this embodiment provides a computing device, including: One or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for executing the above-mentioned regional-level photovoltaic power prediction method.

[0019] The beneficial effect of the present invention is that the regional photovoltaic power prediction method provided by the present invention provides a rich source of information for the prediction model by acquiring the meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station in real time. Meteorological data include multiple factors such as solar radiation intensity, temperature, humidity, wind speed, cloud cover, etc. These factors directly affect the scattering and absorption of solar radiation by the atmosphere, and then determine the effective solar radiation reaching the ground. The operating data of each photovoltaic power station reflects the actual working state of the photovoltaic module, such as power generation efficiency, equipment aging degree, etc. Combining these two types of data, the prediction model can more comprehensively understand the factors affecting photovoltaic power, thereby improving the accuracy of the prediction. Using the generative adversarial network model as the prediction model, in the photovoltaic power prediction, the generator can generate the predicted photovoltaic power value according to the input meteorological data and power station operation data. The discriminator is responsible for judging the closeness of the generated power value to the true value, thereby guiding the generator to continuously optimize the prediction result. Compared with the traditional prediction model, the generative adversarial network model can better handle nonlinear relationships and uncertainties, and has higher accuracy for the prediction task of photovoltaic power, which is affected by multiple factors and has certain uncertainties. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 is a schematic flow chart of a method for predicting regional photovoltaic power in an embodiment of the present invention; Figure 2 is a schematic diagram of the structure of a regional photovoltaic power prediction system in an embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose and technical solution of the present invention clearer and easier to understand, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments, wherein the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0024] Example 1 like Figure 1 As shown, this embodiment provides a regional photovoltaic power prediction method, and the specific implementation method is as follows.

[0025] Obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station are input into the trained prediction model, which is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model.

[0026] The prediction model includes a generation module and a discrimination module.

[0027] The generation module is used to convert the input operation data and meteorological data into generated data with similar characteristics, perform preliminary feature extraction on the generated data, simulate the time characteristics and spatial characteristics of photovoltaic power through the convolution layer, and generate a predicted power value sequence through the output layer based on the time characteristics and spatial characteristics. The task of the generation module is to generate photovoltaic power prediction values ​​based on the input random noise and some auxiliary information (such as time series characteristics, meteorological characteristics, etc.).

[0028] This embodiment uses a deep neural network (DNN) as a generation module, which includes multiple fully connected layers, and each layer is connected using a nonlinear activation function (such as ReLU or Sigmoid). The input layer receives random noise and auxiliary information, and the output layer generates a photovoltaic power prediction value.

[0029] The discriminant module is used to perform convolution and pooling operations on the input data, extract the features of the data, and use the fully connected layer for classification and judgment to determine whether the input data is real photovoltaic power data or predicted data generated in the generation module. The task of the discriminant module is to determine whether the input data is real photovoltaic power data or false data generated by the generator.

[0030] The discriminant module is usually composed of a multi-layer neural network, and its structure can be similar to or different from that of the generator. The output of the discriminant module is a value between 0 and 1, indicating the probability that the input data is real data. For example, this embodiment uses a convolutional neural network (CNN) as a discriminant module, extracts the features of the data by performing convolution and pooling operations on the input data, and then uses a fully connected layer for classification judgment.

[0031] In the prediction model of this embodiment, the parameters of the generation module and the discrimination module are set, such as the number of network layers, the number of neurons in each layer, the activation function, the learning rate, the optimization algorithm, etc. The selection of these parameters will affect the performance and training speed of the model. Therefore, the best parameter combination is determined through training and parameter adjustment.

[0032] This embodiment uses the Adam optimization algorithm as the optimizer, sets the learning rate to 0.001, the number of network layers to 5, and the number of neurons in each layer is adjusted according to the dimension and complexity of the data.

[0033] Among them, the prediction model uses historical operation data and corresponding meteorological data as data sets, and the historical operation data and corresponding meteorological data are pre-processed data, and the pre-processing includes data cleaning and data normalization. In this embodiment, the data set needs to be divided into a training set, a validation set, and a test set. The training set is used to train the model, the validation set is used to adjust the parameters of the model and select the best model, and the test set is used to evaluate the final performance of the model. It can usually be divided according to a certain ratio, such as 70% of the data as a training set, 20% of the data as a validation set, and 10% of the data as a test set.

[0034] Data cleaning specifically includes: cleaning the collected data to remove outliers and erroneous data. Outliers may be obviously unreasonable data points caused by sensor failure, data transmission errors, or other reasons. Specifically, when a data point differs from the mean of other data points by more than a few standard deviations, it can be considered an outlier and deleted. At the same time, check the integrity of the data to ensure that there are no missing values. If there are missing values, interpolation methods or other data filling techniques can be used to supplement the missing data.

[0035] The normalization process includes: normalizing the cleaned data and mapping the data values ​​to a specific range, usually between [0, 1] or [-1, 1]. Normalization can speed up the training of the model, improve the performance of the model, and avoid the impact of large data on model training. Common normalization methods include minimum-maximum normalization, mean-standard deviation normalization, etc. For example, using the minimum-maximum normalization method, the data values ​​are mapped to between [0, 1].

[0036] The training of the prediction model is an adversarial process, which requires alternating training of the generator and the discriminator. The specific steps are as follows: Initialize the parameters of the generation module and the discrimination module; use a random initialization method, such as normal distribution initialization or uniform distribution initialization. The purpose of initializing the parameters is to break the symmetry of the model and prevent the model from falling into a local optimal solution.

[0037] The generation module is fixed and the discrimination module is trained, specifically including: randomly extracting a batch of real photovoltaic power data and corresponding meteorological data from the data set as positive samples; randomly generating a batch of noise vectors from the prior distribution, inputting them into the generation module, generating a batch of false photovoltaic power data and corresponding meteorological data as negative samples; randomly generating a batch of noise vectors from the prior distribution, inputting them into the generation module, generating a batch of false photovoltaic power data and corresponding meteorological data as negative samples; optimizing the algorithm to update the parameters of the discrimination module; Among them, the loss function corresponding to the generation module is:

[0038]

[0039] In the formula, is the adversarial loss function of the generation module, G is the generation module, D is the discrimination module, From the prior distribution The random noise sampled in To generate data for the module, The probability of the real data for the discriminant module to judge the generated data; Fit the loss function to the data, ) is the real data distribution, is the square of the Euclidean distance between the generated data and the true data.

[0040] The discriminant module is fixed and the generation module is trained, specifically including: randomly generating a batch of noise vectors from the prior distribution, inputting them into the generator, generating a batch of false photovoltaic power data and corresponding meteorological data; inputting the generated false data into the discriminant module, calculating the loss function of the generation module; using the optimization algorithm to update the parameters of the generation module, so that the generation module can generate more realistic photovoltaic power data; The loss function corresponding to the discriminant module is:

[0041]

[0042] In the formula, To discriminate the module against the loss function, is the probability that the discriminant module judges that the generated module is real data, is the regularization loss function of the discriminant module, is the regularization coefficient, is the discriminant module parameter.

[0043] When the loss function values ​​of the generation module and the discrimination module change less than the set threshold in several consecutive training rounds, the training is stopped.

[0044] During the training process of the prediction model, the total loss function is:

[0045] In the formula, , , are weight coefficients, is the adversarial loss function of the generation module, Fitting a loss function to the data of the generative module, To discriminate the module against the loss function, is the regularization loss function of the discriminative module.

[0046] During the training process, it is necessary to monitor the model's training progress and performance indicators, such as the loss function value, the quality of generated data, the accuracy of the discriminator, etc.

[0047] Use visualization tools, such as TensorBoard, to observe the loss function curve and the distribution of generated data during training, so as to find problems in time and adjust the parameters of the model. Observe whether the loss function curves of the generation module and the discriminant module gradually decrease, whether the distribution of generated data is getting closer to the distribution of real data, and whether the accuracy of the discriminator fluctuates within a reasonable range.

[0048] Use the test set to evaluate the trained model and calculate the values ​​of the selected evaluation indicators. Compare the model's prediction results with the true values, analyze the model's error distribution, prediction trends, etc. to understand the model's performance and limitations. Observe the relationship between the predicted values ​​and the true values ​​by drawing a scatter plot; calculate the error histogram to understand the error distribution; analyze the model's prediction performance under different weather conditions to determine the model's applicability.

[0049] If the performance of the prediction model is not ideal, the model can be verified and adjusted by increasing the number of model layers, adjusting the learning rate, using different optimization algorithms, adding data enhancement techniques, etc. Then train and evaluate again until the performance of the model meets the requirements.

[0050] In this embodiment, appropriate evaluation indicators are also selected to measure the performance of the model. For photovoltaic power prediction tasks, indicators such as root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (R) can be used.

[0051] RMSE and MAE measure the error between the predicted value and the true value, and R measures the linear correlation between the predicted value and the true value.

[0052] The prediction accuracy and reliability of the model were evaluated by calculating the RMSE and MAE values ​​on the test set, as well as the R value between the predicted value and the true value.

[0053] Use Docker containers to package the model into a deployable image, then deploy it to a cloud server and provide services to the outside world through an API interface.

[0054] By acquiring the meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station in real time, a rich source of information is provided for the prediction model. Meteorological data includes multiple factors such as solar radiation intensity, temperature, humidity, wind speed, cloud cover, etc. These factors directly affect the scattering and absorption of solar radiation by the atmosphere, and thus determine the effective solar radiation reaching the ground. The operating data of each photovoltaic power station reflects the actual working status of the photovoltaic module, such as power generation efficiency, equipment aging, etc. Combining these two types of data, the prediction model can have a more comprehensive understanding of the factors affecting photovoltaic power, thereby improving the accuracy of the prediction. Compared with traditional prediction models, the generative adversarial network model can better handle nonlinear relationships and uncertainties, and has higher accuracy for prediction tasks such as photovoltaic power, which is affected by multiple factors and has certain uncertainties.

[0055] Accurate photovoltaic power forecasting can provide an important reference for the operation and management of power stations. By acquiring data in real time and making forecasts, power station managers can understand the future photovoltaic power output in advance, so as to formulate reasonable power generation plans and scheduling strategies. The operation data of each photovoltaic power station can not only be used for power forecasting, but also provide clues for equipment maintenance and fault diagnosis. By analyzing the changing trend of operation data, managers can timely discover abnormal conditions of equipment and take corresponding maintenance measures. For example, if the power generation efficiency of a photovoltaic module suddenly drops, it may be caused by damage to the module or excessive dust accumulation. By performing maintenance in time, the further expansion of the fault can be avoided and the reliability and power generation efficiency of the power station can be improved.

[0056] Example 2 like Figure 2 As shown, this embodiment provides a regional photovoltaic power prediction system, which is used to implement the regional photovoltaic power prediction method described in Example 1, including: Data acquisition module, used to obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The power prediction module is used to input the meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station into the trained prediction model, and is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model.

[0057] Example 3 In another embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor 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 gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding functions; the processor described in the embodiment of the present invention can be used for the operation of the regional photovoltaic power prediction method, including: Obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station are input into the trained prediction model, which is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model.

[0058] Example 4 In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device, and can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that more specific examples (non-exhaustive list) of the computer-readable storage medium here include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0059] Computer readable storage media also include data signals propagated in baseband or as part of a carrier wave, which carry readable program codes. Such propagated data signals can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0060] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0061] One or more instructions stored in a computer-readable storage medium may be loaded and executed by a processor to implement the corresponding steps of the regional photovoltaic power prediction method in the above embodiment; one or more instructions in the computer-readable storage medium are loaded and executed by the processor as follows: Obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station are input into the trained prediction model, which is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model.

[0062] Figure 3 A schematic diagram of a computer device provided by an embodiment of the present invention.

[0063] See also Figure 3 , the terminal device is a computer device, and the computer device 60 of this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, the regional photovoltaic power prediction method in the embodiment is implemented. To avoid repetition, it is not described one by one here. Alternatively, when the computer program 63 is executed by the processor 61, the functions of each model / unit in the computing system of the regional photovoltaic power prediction method in the embodiment are implemented. To avoid repetition, it is not described one by one here.

[0064] The computer device 60 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will appreciate that Figure 3 This is only an example of the computer device 60 and does not constitute a limitation of the computer device 60. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0065] The processor 61 may be a central processing unit (CPU), or other general-purpose processors, central processing units, graphics 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, data processing logic based on quantum computing, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0066] The memory 62 may be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 60.

[0067] Furthermore, the memory 62 may include both an internal storage unit of the computer device 60 and an external storage device. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0068] Any reference to a memory, database or other medium used in the embodiments provided in this application may include at least one of a non-volatile and a volatile memory. Non-volatile memory may include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory, an optical memory, a high-density embedded non-volatile memory, a resistive random access memory (ReRAM), a magnetic random access memory (MRAM), a ferroelectric random access memory (FRAM), a phase change memory (PCM), a graphene memory, etc. Volatile memory may include a random access memory (RAM) or an external cache memory, etc. As an illustration and not limitation, RAM may be in various forms, such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), etc.

[0069] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

Claims

1. A regional photovoltaic power prediction method, characterized in that: include: Obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station are input into the trained prediction model, which is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model.

2. The regional photovoltaic power prediction method according to claim 1, characterized in that: The prediction model includes a generation module and a discrimination module; The generation module is used to convert the input operation data and meteorological data into generated data with similar characteristics, perform preliminary feature extraction on the generated data, simulate the time characteristics and spatial characteristics of photovoltaic power through the convolution layer, and generate a predicted power value sequence through the output layer based on the time characteristics and spatial characteristics; The discrimination module is used to perform convolution and pooling operations on the input data, extract the features of the data, use the fully connected layer to perform classification judgment, and determine whether the input data is real photovoltaic power data or predicted data generated in the generation module.

3. The regional photovoltaic power prediction method according to claim 2, characterized in that: The prediction model uses historical operation data and corresponding meteorological data as data sets, and the historical operation data and corresponding meteorological data are pre-processed data, and the pre-processing includes data cleaning and data normalization processing.

4. The regional photovoltaic power prediction method according to claim 2, characterized in that: The training process of the prediction model specifically includes: Initialize the parameters of the generation module and the discrimination module; The generation module is fixed and the discrimination module is trained, specifically including: randomly extracting a batch of real photovoltaic power data and corresponding meteorological data from the data set as positive samples; randomly generating a batch of noise vectors from the prior distribution, inputting them into the generation module, generating a batch of false photovoltaic power data and corresponding meteorological data as negative samples; randomly generating a batch of noise vectors from the prior distribution, inputting them into the generation module, generating a batch of false photovoltaic power data and corresponding meteorological data as negative samples; optimizing the algorithm to update the parameters of the discrimination module; The discriminant module is fixed and the generation module is trained, specifically including: randomly generating a batch of noise vectors from the prior distribution, inputting them into the generator, generating a batch of false photovoltaic power data and corresponding meteorological data; inputting the generated false data into the discriminant module, calculating the loss function of the generation module; using the optimization algorithm to update the parameters of the generation module, so that the generation module can generate more realistic photovoltaic power data; When the loss function values ​​of the generation module and the discrimination module change less than the set threshold in several consecutive training rounds, the training is stopped.

5. The regional photovoltaic power prediction method according to claim 4, characterized in that: The loss function corresponding to the generation module is: In the formula, is the adversarial loss function of the generation module, G is the generation module, D is the discrimination module, From the prior distribution The random noise sampled in To generate data for the module, The probability of the real data for the discriminant module to judge the generated data; Fit the loss function to the data, ) is the real data distribution, is the square of the Euclidean distance between the generated data and the true data.

6. The regional photovoltaic power prediction method according to claim 4, characterized in that: The loss function corresponding to the discriminant module is: In the formula, To discriminate the module against the loss function, is the probability that the discriminant module judges that the generated module is real data, is the regularization loss function of the discriminant module, is the regularization coefficient, is the discriminant module parameter.

7. The regional photovoltaic power prediction method according to claim 5 or 6, characterized in that: During the training process of the prediction model, the total loss function is: In the formula, , , are weight coefficients, is the adversarial loss function of the generation module, Fitting a loss function to the data of the generative module, To discriminate the module against the loss function, is the regularization loss function of the discriminative module.

8. A regional photovoltaic power prediction system, used to implement the regional photovoltaic power prediction method according to any one of claims 1 to 7, characterized in that: include: Data acquisition module, used to obtain meteorological data of the photovoltaic module area and operation data of each photovoltaic power station in real time; The power prediction module is used to input the meteorological data of the photovoltaic module area and the operating data of each photovoltaic power station into the trained prediction model, and is used to calculate the scattering and absorption of solar radiation by the atmosphere based on the input meteorological data, estimate the effective solar radiation reaching the ground, and predict the photovoltaic power according to the operating data of each photovoltaic power station to obtain the power prediction result; the prediction model adopts a generative adversarial network model.

9. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions, which, when executed by a computing device, enable the computing device to perform the regional photovoltaic power prediction method according to any one of claims 1 to 7.

10. A computing device, characterized in that: include: One or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for executing the regional photovoltaic power prediction method described in any one of claims 1 to 7.

Images