Photovoltaic power prediction method, system and device and storage medium

Abstract

The invention belongs to the field of photovoltaic power prediction, and discloses a photovoltaic power prediction method, which comprises the following steps of: calculating historical solar effective radiation intensity according to photovoltaic power generation power, photoelectric conversion efficiency of a photovoltaic cell and the total area of a photovoltaic cell assembly; as the photovoltaic power generation power, the photoelectric conversion efficiency of the photovoltaic cell and the total area data of the photovoltaic cell assembly are all historical data true values, the accuracy of obtaining the historical solar effective radiation intensity is ensured; historical solar effective radiation intensity and main factors influencing the solar effective radiation intensity are utilized to train a solar effective radiation intensity prediction model, and the precision of the model is improved due to the fact that the historical solar effective radiation intensity has accuracy; and predicting the solar effective radiation intensity by using the prediction model, and finally outputting a photovoltaic power prediction result according to the solar effective radiation intensity so as to realize accurate prediction of the photovoltaic power. According to the method, on the basis of real historical data, historical solar effective radiation intensity is deduced and calculated, and accuracy and reliability are improved.

Description

Technical Field

[0001] The present invention belongs to the field of photovoltaic power prediction, specifically the field of photovoltaic data acquisition technology, and in particular relates to a photovoltaic power prediction method, system, device and storage medium. Background Art

[0002] Accurately predicting photovoltaic power can not only provide real-time and accurate power output expectations for the grid-connected operation of photovoltaic power stations, optimize scheduling strategies, reduce the phenomenon of abandoned light, and improve energy utilization efficiency, but also provide key inputs for the energy management system (EMS) of the power grid, which helps to achieve the balance of supply and demand of the power grid and enhance the flexibility and reliability of the power grid. In the context of the development of smart grids, this capability is of immeasurable value in achieving large-scale grid connection of renewable energy and promoting the transformation of energy structure.

[0003] At present, the number of stations conducting solar radiation observation is far less than the number of photovoltaic power stations. This means that if the output of a photovoltaic power station is to be predicted, the solar radiation value in this area cannot be directly obtained. The existing solar radiation estimation methods are mostly for irradiance calculation on tilted photovoltaic modules, and do not take into account the situation where dust on the surface of photovoltaic cell modules, modules are blocked by cloud shadows or other unknown obstacles. It is not the effective irradiance actually shining on the surface of photovoltaic modules. Since the solar radiation value input value for photovoltaic power prediction is obtained by estimation, it is quite different from the actual value, resulting in low accuracy of the final photovoltaic power prediction result.

[0004] It can be seen that the existing photovoltaic power prediction measures are not very accurate in the final photovoltaic power prediction results because the solar radiation value is obtained by estimation, which is quite different from the actual value. Summary of the invention

[0005] The purpose of the present invention is to provide a photovoltaic power prediction method, system, device and storage medium to solve the technical problem that the current photovoltaic power prediction measures are not accurate enough because the solar radiation value is obtained by estimation and is greatly different from the actual value.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A photovoltaic power prediction method, comprising: Obtain historical solar effective radiation intensity based on photovoltaic power generation, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell modules; The pre-trained solar effective radiation intensity prediction model is used to predict the solar effective radiation intensity prediction data, and the photovoltaic power prediction result is obtained according to the solar effective radiation intensity prediction data; The solar effective radiation intensity prediction model is trained based on the historical solar effective radiation intensity and the main factors affecting the solar effective radiation intensity.

[0007] Furthermore, the specific relationship between the solar effective radiation intensity and the photovoltaic power is as follows:

[0008] Among them, P is the photovoltaic power generation power; A2 is the total area of ​​the photovoltaic cell assembly; η is the photoelectric conversion efficiency of the photovoltaic cell; I is the effective solar radiation intensity.

[0009] Furthermore, the solar effective radiation intensity prediction model adopts an LSTM network structure.

[0010] Furthermore, the training process of the solar effective radiation intensity prediction model is as follows: The historical solar effective radiation intensity and the main factors screened are used as a data set, and the data set is divided into a training set and a test set; The training set is used to build and train the initial model for predicting the effective solar radiation intensity, and the test set is used to test the model accuracy; The initial solar effective radiation intensity prediction model is iterated cyclically until the model accuracy meets the preset accuracy requirements, and the solar effective radiation intensity prediction model is output.

[0011] Furthermore, the main factors affecting the effective solar radiation intensity were screened using the Pearson correlation analysis method, and the specific formula is as follows:

[0012] Where R represents the Pearson correlation coefficient; x represents the factor to be analyzed; l represents the photovoltaic power time series.

[0013] Furthermore, the main factors include: wind direction, wind speed, temperature, air pressure, relative humidity, and precipitation.

[0014] Furthermore, before obtaining the historical solar effective radiation intensity, the historical data of the photovoltaic power station are preprocessed; The preprocessing process specifically includes: data cleaning and data normalization of historical data of photovoltaic power stations; the historical data of photovoltaic power stations include photovoltaic power generation power, photoelectric conversion efficiency of photovoltaic cells and total area of ​​photovoltaic cell modules.

[0015] A photovoltaic power prediction system, used to implement the steps of the photovoltaic power prediction method, comprising: A data acquisition module is used to obtain historical solar effective radiation intensity based on photovoltaic power generation, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell modules; A power prediction module is used to predict the solar effective radiation intensity prediction data using a pre-trained solar effective radiation intensity prediction model, and obtain a photovoltaic power prediction result based on the solar effective radiation intensity prediction data; The solar effective radiation intensity prediction model is trained based on historical data of photovoltaic power stations, historical solar effective radiation intensity and main factors affecting solar effective radiation intensity.

[0016] A device comprising: Memory for storing computer programs; A processor is used to implement the steps of the photovoltaic power prediction method when executing the computer program.

[0017] A computer-readable storage medium stores a computer program, which is used to implement the steps of the photovoltaic power prediction method when executed by a processor.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a photovoltaic power prediction method. The prediction method first calculates the historical solar effective radiation intensity according to the photovoltaic power generation power, the photoelectric conversion efficiency of the photovoltaic cell and the total area of ​​the photovoltaic cell assembly. Since the photovoltaic power generation power, the photoelectric conversion efficiency of the photovoltaic cell and the total area of ​​the photovoltaic cell assembly are all real values ​​of the historical data, the accuracy of the historical solar effective radiation intensity is guaranteed. Then, the solar effective radiation intensity prediction model is trained by using the historical solar effective radiation intensity and the main factors affecting the solar effective radiation intensity. Since the accuracy of the historical solar effective radiation intensity is high, the accuracy of the prediction model is also improved. Finally, the solar effective radiation intensity is predicted by using the solar effective radiation intensity prediction model, and the photovoltaic power prediction result is output according to the solar effective radiation intensity, so as to realize the accurate prediction of the photovoltaic power. The method uses the real historical data as the basis, derives and calculates the historical solar effective radiation intensity, and improves the accuracy and reliability of the prediction.

[0019] Preferably, in the present invention, the relationship between photovoltaic power generation power and the effective solar radiation intensity, the photoelectric conversion efficiency of photovoltaic cells and the total area of ​​photovoltaic cell modules is utilized to provide a theoretical basis for obtaining historical data to calculate the effective solar radiation intensity. The use of this method makes data analysis and model training more scientific and accurate, thereby improving the prediction accuracy.

[0020] Preferably, in the present invention, an LSTM (Long Short-Term Memory Network) network structure is used as a prediction model for the effective solar radiation intensity, which can effectively handle the long-term dependency problem in time series data, capture the complex dynamic relationship between photovoltaic power generation and the effective solar radiation intensity, and further improve the prediction accuracy.

[0021] Preferably, in the present invention, the training process of the solar effective radiation intensity prediction model includes the division of the data set, the construction and training of the model, and iterative optimization until the preset accuracy requirements are met, thereby ensuring the stability and accuracy of the model, so that the final output prediction model has a higher practical application value.

[0022] Preferably, in the present invention, the Pearson correlation analysis method is used to screen the main factors affecting the effective solar radiation intensity, which can quantify the correlation between each factor and the effective solar radiation intensity, remove redundant information, and retain factors that have a significant impact on the prediction results, thereby improving the efficiency and accuracy of the prediction model.

[0023] Preferably, in the present invention, the main factors affecting the effective solar radiation intensity include wind direction, wind speed, temperature, air pressure, relative humidity and precipitation, etc. These factors cover the main aspects of meteorological conditions, provide a comprehensive reference for the input of the model, and help to improve the comprehensiveness and accuracy of the prediction.

[0024] Preferably, in the present invention, before obtaining the historical effective solar radiation intensity, the historical data of the photovoltaic power station is preprocessed, including data cleaning and data normalization, so as to remove noise and outliers in the data, unify the dimension and distribution range of the data, improve the data quality, and lay a solid foundation for subsequent model training and prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A flow chart of a photovoltaic power prediction method provided by an embodiment of the present invention; Figure 2 A network structure diagram of a long short-term memory network LSTM used in a solar effective radiation intensity prediction model provided in an embodiment of the present invention; Figure 3 A diagram of the training process of the solar effective radiation intensity prediction model provided in an embodiment of the present invention; Figure 4 A comparison chart of training set prediction results provided by an embodiment of the present invention; Figure 5 A comparison chart of test set prediction results provided by an embodiment of the present invention; Figure 6 A flow chart of a photovoltaic power prediction method provided by the present invention; Figure 7 A schematic structural diagram of a photovoltaic power prediction system provided by the present invention. DETAILED DESCRIPTION

[0026] Example 1 The present invention provides a photovoltaic power prediction method, such as Figure 6 As shown, the following steps are included: S1: Obtain historical solar effective radiation intensity based on photovoltaic power generation, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell modules; The historical data of the photovoltaic power station at least includes photovoltaic power generation, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell modules; the specific relationship between solar effective radiation intensity and photovoltaic power is as follows:

[0027] Among them, P is the photovoltaic power generation power; A2 is the total area of ​​the photovoltaic cell assembly; η is the photoelectric conversion efficiency of the photovoltaic cell; I is the effective solar radiation intensity.

[0028] S2: Use the pre-trained solar effective radiation intensity prediction model to predict the solar effective radiation intensity prediction data, and obtain the photovoltaic power prediction result according to the solar effective radiation intensity prediction data; The solar effective radiation intensity prediction model is trained based on the historical solar effective radiation intensity and the main factors affecting the solar effective radiation intensity.

[0029] Among them, the above-mentioned solar effective radiation intensity prediction model adopts the LSTM network structure.

[0030] The specific training process of the solar effective radiation intensity prediction model is as follows: The historical solar effective radiation intensity and the main factors screened are used as a data set, and the data set is divided into a training set and a test set; The training set is used to build and train the initial model for predicting the effective solar radiation intensity, and the test set is used to test the model accuracy; The initial solar effective radiation intensity prediction model is iterated cyclically until the model accuracy meets the preset accuracy requirements, and the solar effective radiation intensity prediction model is output.

[0031] The main factors affecting the effective solar radiation intensity were screened using the Pearson correlation analysis method. The specific formula is as follows:

[0032] Where R represents the Pearson correlation coefficient; x represents the factor to be analyzed; l represents the photovoltaic power time series.

[0033] The above main factors include: wind direction, wind speed, temperature, air pressure, relative humidity, and precipitation.

[0034] In the present invention, before obtaining the historical effective solar radiation intensity, the historical data of the photovoltaic power station is preprocessed; the preprocessing process specifically includes: data cleaning and data normalization processing of the historical data of the photovoltaic power station.

[0035] like Figure 7 As shown, the present invention also provides a photovoltaic power prediction system, including: a data acquisition module, used to obtain historical solar effective radiation intensity based on photovoltaic power generation power, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell assemblies; a power prediction module, used to use a pre-trained solar effective radiation intensity prediction model to predict solar effective radiation intensity prediction data, and obtain a photovoltaic power prediction result based on the solar effective radiation intensity prediction data; wherein, the solar effective radiation intensity prediction model is trained based on historical data of photovoltaic power stations, historical solar effective radiation intensity and main factors affecting solar effective radiation intensity.

[0036] The present invention also provides a device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the photovoltaic power prediction method when executing the computer program.

[0037] When the processor executes the computer program, the above-mentioned photovoltaic power prediction steps are implemented, for example: obtaining historical solar effective radiation intensity based on photovoltaic power generation power, photoelectric conversion efficiency of photovoltaic cells and total area of ​​photovoltaic cell modules; using a pre-trained solar effective radiation intensity prediction model to predict solar effective radiation intensity prediction data, and obtaining a photovoltaic power prediction result based on the solar effective radiation intensity prediction data; wherein, the solar effective radiation intensity prediction model is trained based on historical data of photovoltaic power stations, historical solar effective radiation intensity and main factors affecting solar effective radiation intensity.

[0038] Alternatively, when the processor executes the computer program, it implements the functions of each module in the above system, for example: a data acquisition module, which is used to obtain historical solar effective radiation intensity based on photovoltaic power generation power, photoelectric conversion efficiency of photovoltaic cells and total area of ​​photovoltaic cell assemblies; a power prediction module, which is used to use a pre-trained solar effective radiation intensity prediction model to predict solar effective radiation intensity prediction data, and obtain photovoltaic power prediction results based on the solar effective radiation intensity prediction data; wherein, the solar effective radiation intensity prediction model is trained based on historical data of photovoltaic power stations, historical solar effective radiation intensity and main factors affecting solar effective radiation intensity.

[0039] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete preset functions, and the instruction segments are used to describe the execution process of the computer program in the photovoltaic power prediction device. For example, the computer program can be divided into a data acquisition module and a power prediction module; the specific functions of each module are as follows: a data acquisition module, which is used to obtain historical solar effective radiation intensity based on photovoltaic power generation power, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell components; a power prediction module, which is used to use a pre-trained solar effective radiation intensity prediction model to predict solar effective radiation intensity prediction data, and obtain photovoltaic power prediction results based on the solar effective radiation intensity prediction data; wherein, the solar effective radiation intensity prediction model is trained based on historical data of photovoltaic power stations, historical solar effective radiation intensity and main factors affecting solar effective radiation intensity.

[0040] The photovoltaic power prediction device can be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The photovoltaic power prediction device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above are examples of photovoltaic power prediction devices and do not constitute a limitation on photovoltaic power prediction devices. More components than the above may be included, or some components may be combined, or different components may be included. For example, the photovoltaic power prediction device may also include input and output devices, network access devices, buses, etc.

[0041] 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 gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The processor is the control center of the photovoltaic power prediction, and uses various interfaces and lines to connect various parts of the entire photovoltaic power prediction device.

[0042] The memory can be used to store the computer program and / or module, and the processor implements various functions of the photovoltaic power prediction device by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0043] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also 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 volatile solid-state storage devices.

[0044] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the photovoltaic power prediction method are implemented.

[0045] If the integrated module / unit of the photovoltaic power prediction system 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.

[0046] Based on such understanding, the present invention implements all or part of the processes in the above photovoltaic power prediction method, and can also be completed by instructing related hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above photovoltaic power prediction method can be implemented. 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 preset intermediate form, etc.

[0047] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0048] It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0049] The present invention will be further described below in conjunction with the embodiments and drawings: Example 2 As described in the background technology, the number of existing stations for solar radiation observation is far less than the number of photovoltaic power stations. This means that if the output of a photovoltaic power station is to be predicted, the solar radiation value in this area cannot be directly obtained. Most of the existing solar radiation estimation methods are based on irradiance calculations on tilted photovoltaic modules, and do not take into account dust on the surface of photovoltaic cell modules, modules being blocked by cloud shadows or other unknown obstacles. It is not the effective irradiance actually shining on the surface of photovoltaic modules. Since the solar radiation value input value for photovoltaic power prediction is obtained by estimation, it is quite different from the actual value, resulting in low accuracy of the final photovoltaic power prediction result.

[0050] In order to achieve the above-mentioned purpose, the present invention provides a photovoltaic power prediction method, which collects real data to deduce and calculate the effective solar radiation intensity, and then uses the effective solar radiation intensity combined with the main factors affecting the effective solar radiation intensity to construct and train a solar effective radiation intensity prediction model, thereby achieving accurate and reliable prediction of photovoltaic power.

[0051] The photovoltaic power prediction method provided in this embodiment is based on the following ideas: First, there is a strong correlation between the two data, solar effective radiation intensity and photovoltaic power generation, which are non-independent data. This will make the model perform well on the training set when training the photovoltaic power generation prediction model, but when encountering new and unseen data, the generalization ability is poor, which may lead to prediction errors; therefore, the historical true value is used to obtain the solar effective radiation intensity, and then the model is constructed and trained to ensure the accuracy of the final prediction result, that is, the relationship between solar effective radiation intensity and photovoltaic power is used to solve the problem that the solar radiation value cannot be directly obtained. Second, considering that too much consideration of non-critical influencing factors will increase the difficulty of training the photovoltaic power prediction model, the main factors affecting photovoltaic power can be analyzed through Pearson correlation analysis, thereby solving the problem of overfitting in the training process.

[0052] like Figure 1 As shown, this embodiment provides a photovoltaic power prediction method, which specifically includes: Step 1: Collect one year of historical data of photovoltaic power stations and preprocess all historical data. The preprocessing process specifically includes data cleaning and data normalization of the historical data of photovoltaic power stations to remove noise and outliers in the data, unify the dimension and distribution range of the data, improve data quality, and lay a solid foundation for subsequent model training and prediction.

[0053] Among them, the historical data of photovoltaic power stations include photovoltaic power generation power, photoelectric conversion efficiency of photovoltaic cells, total area of ​​photovoltaic cell modules, and factors affecting the effective solar radiation intensity.

[0054] Step 2: Using the solar effective radiation intensity relationship, the historical effective solar radiation intensity is calculated based on the photovoltaic power generation power, the photovoltaic cell photoelectric conversion efficiency and the total area of ​​the photovoltaic cell module; at the same time, the main factors affecting the photovoltaic power are analyzed through Pearson correlation; here, the main factors include: wind direction, wind speed, temperature, air pressure, relative humidity and precipitation.

[0055] Among them, the specific relationship between the effective solar radiation intensity and photovoltaic power is as follows:

[0056] Where P is the photovoltaic power generation; A2 is the total area of ​​the photovoltaic cell assembly; η is the photoelectric conversion efficiency of the photovoltaic cell; I is the effective solar radiation intensity.

[0057] Among them, the formula for Pearson correlation analysis affecting the effective solar radiation intensity is:

[0058] Where R represents the Pearson correlation coefficient; x represents the factor to be analyzed; l represents the photovoltaic power time series.

[0059] Step 3: Build and train a solar effective radiation intensity prediction model, where the solar effective radiation intensity prediction model adopts an LSTM network structure.

[0060] like Figure 2 As shown in FIG. 1 , since the RNN neural network will experience the “gradient vanishing” phenomenon when processing a very long time series, in this embodiment, the following Figure 2 The LSTM network structure shown in the figure can not only effectively alleviate the problem caused by back propagation during RNN training, but also effectively handle the problem of long-term sequence dependencies. The LSTM network mainly includes input gate, output gate and forget gate. The forget gate determines whether to retain useful information, the input gate determines what information is stored in the memory unit, and the output gate determines the next hidden state. The calculation process is:

[0061] Where: it represents the input gate; W represents the weight matrix; xt represents the input data of the current time step t; ct and mt represent the activation vectors of the cell state and memory block; b represents the bias term; ft represents the forget gate.

[0062] like Figure 3 As shown, in this embodiment, the training process of the solar effective radiation intensity prediction model is as follows: S101, normalizing the data set of historical solar effective radiation intensity and main factors, and dividing it into a training set and a test set; S102. Use the training set to build an initial model for predicting the effective solar radiation intensity and train the model. Use the test set to test the model accuracy. The specific training set results are as follows: Figure 4 The test set test results are shown in Figure 5 As shown; S103, adjusting the parameters of the initial model for predicting the solar effective radiation intensity until the accuracy meets the requirements; S104, outputting the trained solar effective radiation intensity prediction model.

[0063] Step 4: Use the solar effective radiation intensity prediction model to predict the effective solar radiation value; here, collect relevant data to be predicted, including data values ​​corresponding to the main factors, and input the above data into the solar effective radiation intensity prediction model to obtain the effective solar radiation value, that is, the solar effective radiation intensity.

[0064] Finally, according to the specific relationship between the effective solar radiation intensity and the photovoltaic power, the photovoltaic power prediction value is calculated using the effective solar radiation value, and the photovoltaic power prediction result is output.

[0065] The results of the forecast data analysis are shown in Table 1: Table 1 shows the coefficient of determination, mean absolute error and mean deviation error of the training set and the test set.

[0066] Combination Figures 3 to 5 As shown in Table 1, the photovoltaic power prediction method provided in this embodiment has high prediction accuracy and reliability.

[0067] The present invention provides a photovoltaic power prediction method, which has the following advantages over the existing photovoltaic power prediction methods: This method not only estimates the historical solar effective radiation intensity based on photovoltaic power generation, photoelectric conversion efficiency and total area of ​​photovoltaic cell modules, but also further considers various main factors affecting the solar effective radiation intensity (such as wind direction, wind speed, temperature, air pressure, relative humidity and precipitation) by introducing a pre-trained solar effective radiation intensity prediction model. These factors are scientifically screened by Pearson correlation analysis to ensure the effectiveness and pertinence of the model input. More importantly, this method considers various factors that may affect the photovoltaic module receiving solar radiation in actual situations during the prediction process, such as cloud shadows, dust accumulation or other obstacles, which are often ignored in existing solar radiation estimation methods; the prediction model constructed by the LSTM network structure can capture the long-term dependence and nonlinear relationship in time series data, so as to more accurately predict the solar effective radiation intensity and obtain the prediction results of photovoltaic power based on this. In addition, the preprocessing steps of historical data (including data cleaning and data normalization) ensure the quality and consistency of the input data and reduce the impact of noise and outliers on the prediction results. The use of this prediction method not only improves the accuracy of photovoltaic power prediction, but also provides more reliable data support for the operation and maintenance of photovoltaic power stations, which helps to optimize energy distribution, improve power generation efficiency and reduce operation and maintenance costs.

[0068] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.

Claims

1. A photovoltaic power prediction method, characterized in that: include: Obtain historical solar effective radiation intensity based on photovoltaic power generation, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell modules; The pre-trained solar effective radiation intensity prediction model is used to predict the solar effective radiation intensity prediction data, and the photovoltaic power prediction result is obtained according to the solar effective radiation intensity prediction data; The solar effective radiation intensity prediction model is trained based on the historical solar effective radiation intensity and the main factors affecting the solar effective radiation intensity.

2. The photovoltaic power prediction method according to claim 1, characterized in that: The specific relationship between the effective solar radiation intensity and photovoltaic power is as follows: Among them, P is the photovoltaic power generation power; A2 is the total area of ​​the photovoltaic cell assembly; η is the photoelectric conversion efficiency of the photovoltaic cell; I is the effective solar radiation intensity.

3. The photovoltaic power prediction method according to claim 1, characterized in that: The solar effective radiation intensity prediction model adopts an LSTM network structure.

4. The photovoltaic power prediction method according to claim 1, characterized in that: The training process of the solar effective radiation intensity prediction model is as follows: The historical solar effective radiation intensity and the main factors screened are used as a data set, and the data set is divided into a training set and a test set; The training set is used to build and train the initial model for predicting the effective solar radiation intensity, and the test set is used to test the model accuracy; The initial solar effective radiation intensity prediction model is iterated cyclically until the model accuracy meets the preset accuracy requirements, and the solar effective radiation intensity prediction model is output.

5. The photovoltaic power prediction method according to claim 1, characterized in that: The main factors affecting the effective solar radiation intensity are screened using the Pearson correlation analysis method. The specific formula is as follows: Where R represents the Pearson correlation coefficient; x represents the factor to be analyzed; l represents the photovoltaic power time series.

6. The photovoltaic power prediction method according to claim 1, characterized in that: The main factors include: wind direction, wind speed, temperature, air pressure, relative humidity, and precipitation.

7. The photovoltaic power prediction method according to claim 1, characterized in that: Before obtaining the historical effective solar radiation intensity, the historical data of the photovoltaic power station is preprocessed; The preprocessing process specifically includes: data cleaning and data normalization of historical data of photovoltaic power stations; the historical data of photovoltaic power stations include photovoltaic power generation power, photoelectric conversion efficiency of photovoltaic cells and total area of ​​photovoltaic cell modules.

8. A photovoltaic power prediction system, used to implement the steps of the photovoltaic power prediction method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module is used to obtain historical solar effective radiation intensity based on photovoltaic power generation, photovoltaic cell photoelectric conversion efficiency and total area of ​​photovoltaic cell modules; A power prediction module is used to predict the solar effective radiation intensity prediction data using a pre-trained solar effective radiation intensity prediction model, and obtain a photovoltaic power prediction result based on the solar effective radiation intensity prediction data; The solar effective radiation intensity prediction model is trained based on historical data of photovoltaic power stations, historical solar effective radiation intensity and main factors affecting solar effective radiation intensity.

9. A device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the photovoltaic power prediction method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the photovoltaic power prediction method according to any one of claims 1 to 7.

Images