Photovoltaic power generation short-term prediction method and device based on satellite inversion data

Abstract

The invention discloses a photovoltaic power generation short-time prediction method and device based on satellite inversion data, and belongs to the technical field of photovoltaic power generation prediction, and the method comprises the steps: S1, building a photovoltaic power generation prediction data set; s2, preprocessing the photovoltaic power generation prediction data set; s3, constructing a photovoltaic power generation short-term prediction hybrid model; s4, training the photovoltaic power generation short-time prediction hybrid model based on the photovoltaic power generation prediction data set, and carrying out iteration until a loss function is converged; s5, acquiring satellite inversion data, meteorological data and photovoltaic power generation historical data of the target photovoltaic power station; and S6, obtaining a photovoltaic power generation short-term prediction result of the target photovoltaic power station according to the trained photovoltaic power generation short-term prediction hybrid model. According to the method, related information of satellite inversion data can be fully utilized, the over-fitting defect of a single model is avoided, and the accuracy of a photovoltaic power generation prediction result is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power generation prediction, and in particular relates to a method and device for short-term photovoltaic power generation prediction based on satellite inversion data. Background Art

[0002] With the widespread application of renewable energy, photovoltaic power generation, as an important component of it, has brought challenges to the stable operation of the power grid due to its intermittent and unpredictable nature. Photovoltaic power generation is significantly affected by weather, season and time, resulting in power supply fluctuations, which requires grid operators to accurately predict and manage to ensure the stability of the grid and the continuity of power supply. Current research mainly focuses on short-term and short-term photovoltaic power generation prediction, which helps grid operators quickly respond to the fluctuation of photovoltaic power generation through predictions on different time scales.

[0003] Photovoltaic power prediction methods can be divided into physical methods and machine learning methods. Among them, physical methods are based on the physical properties of photovoltaic cells and the principle of photoelectric conversion. They predict photovoltaic power by simulating the physical process of solar irradiance transmission and photovoltaic module power generation. Such methods usually require accurate meteorological data, detailed geographic information and photovoltaic system parameters, such as temperature, solar radiation and solar radiation distribution. The accuracy of this prediction method depends largely on the accuracy of meteorological data, which leads to difficulties in actual deployment and application. Machine learning methods are mainly based on artificial intelligence algorithms. Currently, there are methods that use long short-term memory networks (LSTM, Long Short-Term Memory) methods, convolutional neural networks (CNN) methods, and methods that combine CNN and LSTM to predict photovoltaic power. However, more and more literature shows that different meteorological data, model algorithms and hyperparameter settings may lead to different prediction results, that is, the accuracy of model prediction needs to be improved. For example, the LSTM method alone cannot make full use of spatial features, such as geographical location, terrain, etc., which have a significant impact on photovoltaic power generation efficiency. The CNN method alone usually does not have the ability to process time series data, and photovoltaic power generation data has obvious time dependence, and using the CNN method alone may not be effective. Although the method combining CNN and LSTM can take into account both spatial features and time series features, actual data show that the predicted values ​​output by this method fluctuate less and have larger errors compared with the actual values. This is because the input data only considers terrain effects, atmospheric effects and sky conditions, but does not fully utilize the relevant parameters of photovoltaic power generation.

[0004] Therefore, there is an urgent need for a new short-term prediction method and device for photovoltaic power generation. This method and device not only considers both spatial characteristics and time series characteristics at the same time, avoiding the overfitting of a single model, but also fully mines the data characteristics of photovoltaic power generation, thereby realizing accurate prediction of photovoltaic power generation power. Summary of the invention

[0005] In view of the defects existing in the above-mentioned prior art, the present invention provides a photovoltaic power generation short-term prediction method based on satellite inversion data, characterized in that the method comprises:

[0006] Step S1, establishing a photovoltaic power generation prediction data set, wherein the photovoltaic power generation prediction data set includes satellite inversion data, meteorological data, and photovoltaic power generation historical data;

[0007] Step S2, preprocessing the photovoltaic power generation prediction data set;

[0008] Step S3, constructing a photovoltaic power generation short-term prediction hybrid model, wherein the photovoltaic power generation short-term prediction hybrid model includes a convolutional neural network-long short-term memory network (CNN-LSTM) module, an LSTM module and a feature fusion module, and the output end of the CNN-LSTM module and the output end of the LSTM module are both connected to the feature fusion module;

[0009] The CNN-LSTM module is used to process the preprocessed satellite inversion data, and obtain a first result of photovoltaic power generation short-term prediction based on the satellite inversion data; the LSTM module is used to process the preprocessed meteorological data and the photovoltaic power generation historical data, and obtain a second result of photovoltaic power generation short-term prediction based on the meteorological data and the photovoltaic power generation historical data; the feature fusion module is used to fuse the first result of photovoltaic power generation short-term prediction and the second result of photovoltaic power generation short-term prediction to obtain a final prediction result;

[0010] Step S4, training the photovoltaic power generation short-term prediction hybrid model based on the photovoltaic power generation prediction data set, and iterating until the loss function converges;

[0011] Step S5, obtaining satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station;

[0012] Step S6: pre-process the satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station, and input them into the trained photovoltaic power generation short-term prediction hybrid model to obtain the photovoltaic power generation short-term prediction result of the target photovoltaic power station.

[0013] Furthermore, the preprocessing includes missing value processing, data normalization, enhanced Dickey-Fuller test and autocorrelation test.

[0014] Furthermore, the enhanced Dickey-Fuller test is specifically:

[0015] S21: Check the photovoltaic power generation prediction data set y according to the following formula t The unit root of :

[0016]

[0017] Among them, Δy t is the photovoltaic power generation prediction data set y t α is the parameter to be tested. If α=0, it means that the photovoltaic power generation prediction data set y t There is a unit root and it is non-stationary; is the lag term coefficient, p is the number of lag terms, βt is the time trend term, ∈ t is the error term;

[0018] S22: Use the least squares method to estimate α, β and

[0019] S23: Calculate the statistic of the enhanced Dickey-Fuller test, as shown in the following formula:

[0020]

[0021] in, is an estimate of α, yes The standard error of

[0022] S24: The calculated statistics ADF Sta Compared with the critical value, if ADF Sta is less than 0.05, then the photovoltaic power generation prediction data set y t For stationary data; if ADF Sta is greater than 0.05, then the photovoltaic power generation prediction data set y t is non-stationary data;

[0023] S25: For non-stationary data, the difference method is used to eliminate the photovoltaic power generation prediction data set y t The time series dependence of , then proceed to step S21 to continue calculating ADF Sta , until it is confirmed that the above data set is stable data.

[0024] Furthermore, after performing the enhanced Dickey-Fuller test, an autocorrelation test is continued, specifically:

[0025] S26: Perform autocorrelation test according to the following formula:

[0026]

[0027] Among them, h is the lag time, cov is the covariance calculation, var(y t ) is the photovoltaic power generation prediction data set y t Variance in time series;

[0028] S27: Determine whether the above calculation result ACF(h) exceeds the preset confidence interval threshold. If it exceeds the preset confidence interval threshold, it is considered that the photovoltaic power generation prediction data set y t Satisfy the correlation requirements of time series.

[0029] Furthermore, the CNN-LSTM module first extracts the spatial features of the satellite inversion data through the CNN layer, and maps the obtained feature map to the LSTM layer. The LSTM layer is used to extract time information for time series prediction. Finally, the first result of the short-term prediction of photovoltaic power generation is obtained through the fully connected layer.

[0030] Furthermore, the specific process of the CNN-LSTM module obtaining the first result of the photovoltaic power generation short-term prediction is as follows:

[0031] Step S311: The CNN layer processes the satellite inversion data through convolution and activation functions:

[0032] O CNN =ReLU(W CNN *X+b CNN );

[0033] Wherein, X is the satellite inversion data, W CNN and b CNN They are the weights and biases of the CNN layer, * represents the convolution operation, and ReLU represents the rectified linear activation unit function;

[0034] Step S312: Use pooling operation to reduce feature dimension, that is:

[0035] P = Pooling (O CNN );

[0036] Among them, Pooling represents the pooling operation;

[0037] Step S313: The output P of the CNN layer is used as the input of the LSTM layer. The dimension of the output P is D×T, where D is the feature depth and T is the time series step. The LSTM layer processes and outputs the hidden state h of each time step. t :

[0038] h t =LSTM(P, ht-1 );

[0039] Among them, h t is the hidden state at time step t, LSTM represents the operation of the LSTM unit, and h t-1 is the hidden state at the previous time step;

[0040] Step S314: The hidden state h of the last time step of the LSTM layer T After processing through the fully connected layer, the final output is obtained Specifically:

[0041] Z=W FC H+b FC ;

[0042]

[0043] Among them, W FC and b FC They are the bias and weight of the fully connected layer respectively, and softmax is the activation function.

[0044] Furthermore, after receiving the meteorological data and the photovoltaic power generation historical data, the LSTM module first combines the meteorological data and the photovoltaic power generation historical data using timestamp information, and then performs short-term photovoltaic power generation prediction on the combined data to obtain the second result of the short-term photovoltaic power generation prediction; wherein the meteorological data includes radiation, DHI, DNI, GHI, relative humidity, wind direction, dew point, pressure, temperature and solar zenith angle, and the photovoltaic power generation historical data includes photovoltaic historical power generation power.

[0045] The present invention also provides a photovoltaic power generation short-term prediction device based on satellite inversion data, characterized in that the device comprises:

[0046] A data set construction module, wherein the data set construction module is used to establish a photovoltaic power generation prediction data set, wherein the photovoltaic power generation prediction data set includes satellite inversion data, meteorological data, and photovoltaic power generation historical data;

[0047] A preprocessing module, the preprocessing module is used to preprocess the photovoltaic power generation prediction data set;

[0048] A model building module, wherein the model building module is used to build a hybrid model for short-term prediction of photovoltaic power generation, wherein the hybrid model for short-term prediction of photovoltaic power generation includes a CNN-LSTM module, an LSTM module and a feature fusion module, wherein the output end of the CNN-LSTM module and the output end of the LSTM module are both connected to the feature fusion module;

[0049] The CNN-LSTM module is used to process the preprocessed satellite inversion data, and obtain a first result of photovoltaic power generation short-term prediction based on the satellite inversion data; the LSTM module is used to process the preprocessed meteorological data and the photovoltaic power generation historical data, and obtain a second result of photovoltaic power generation short-term prediction based on the meteorological data and the photovoltaic power generation historical data; the feature fusion module is used to fuse the first result of photovoltaic power generation short-term prediction and the second result of photovoltaic power generation short-term prediction to obtain a final prediction result;

[0050] A training module, the training module is used to train the photovoltaic power generation short-term prediction hybrid model based on the photovoltaic power generation prediction data set, and iterate until the loss function converges;

[0051] An acquisition module, the acquisition module is used to acquire satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station;

[0052] A prediction module is used to pre-process the satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station, and input them into the trained photovoltaic power generation short-term prediction hybrid model to obtain the photovoltaic power generation short-term prediction result of the target photovoltaic power station.

[0053] The present invention also provides a computer-readable storage medium, which stores a computer program, characterized in that when the computer program is executed by a processor, the short-term prediction method for photovoltaic power generation based on satellite inversion data is implemented.

[0054] The present invention also provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is used to execute the computer program to implement the short-term prediction method for photovoltaic power generation based on satellite inversion data.

[0055] The present invention firstly processes the satellite inversion data using the CNN-LSTM module, fully utilizing the spatial characteristics and time series characteristics of the satellite inversion data, and obtains the first result of the short-term prediction of photovoltaic power generation; at the same time, the LSTM module is used to process the meteorological data and the historical data of photovoltaic power generation, and fully utilizing the time series characteristics of the meteorological data and the historical data of photovoltaic power generation, and obtains the second result of the short-term prediction of photovoltaic power generation; then the first result and the second result are fused through the fully connected layer, so as to obtain the final short-term prediction result of photovoltaic power generation. In the above manner, the relevant information of the satellite inversion data is fully utilized, while avoiding the overfitting defect of a single model, and strengthening the weight of the time series characteristics of the meteorological data and the historical data of photovoltaic power generation in the photovoltaic power generation prediction process, thereby improving the accuracy of the photovoltaic power generation prediction result. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present disclosure will become readily understood. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0057] Figure 1 The flowchart shows a method for short-term photovoltaic power generation prediction based on satellite inversion data according to an embodiment of the present invention.

[0058] Figure 2 The figure shows a hybrid model architecture for short-term prediction of photovoltaic power generation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0061] It should be understood that although the terms first, second, third, etc. may be used to describe ... in the embodiments of the present invention, these ... should not be limited to these terms. These terms are only used to distinguish .... For example, without departing from the scope of the embodiments of the present invention, the first ... may also be referred to as the second ..., and similarly, the second ... may also be referred to as the first ....

[0062] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0063] As used herein, the words “if” and “if” may be interpreted as “at the time of” or “when” or “in response to determining” or “in response to detecting,” depending on the context. Similarly, the phrases “if it is determined” or “if (stated condition or event) is detected” may be interpreted as “when it is determined” or “in response to determining” or “when detecting (stated condition or event)” or “in response to detecting (stated condition or event),” depending on the context.

[0064] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a product or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such product or device. In the absence of more restrictions, the elements defined by the sentence "includes a ..." do not exclude the existence of other identical elements in the product or device including the elements.

[0065] To realize short-term power prediction of photovoltaic power generation, Figure 1 As shown, an embodiment of the present invention discloses a photovoltaic power generation short-term prediction method based on satellite inversion data, the method comprising:

[0066] Step S1, establishing a photovoltaic power generation prediction data set, wherein the photovoltaic power generation prediction data set includes satellite inversion data, meteorological data and photovoltaic power generation historical data.

[0067] Satellite inversion data usually refers to data on various physical parameters of the earth's surface or atmosphere obtained through satellite remote sensing technology. These data are obtained by analyzing the electromagnetic wave information received by satellite sensors to invert the characteristics of the earth's surface or atmosphere, such as temperature, humidity, vegetation index, atmospheric component concentration, etc. When predicting photovoltaic power generation, satellite inversion data can provide solar radiation data with high temporal and spatial resolution, which helps to simulate the output of photovoltaic systems more accurately.

[0068] Meteorological data include solar radiation (W / ㎡), temperature (℃), dew point (℃), relative humidity (%), pressure (mBar), solar zenith angle (degrees), etc., and are all related to photovoltaic power generation. In the embodiment of the present invention, the Pearson correlation coefficient is used to determine whether the above-mentioned various meteorological data are significantly correlated with photovoltaic power generation power. According to the research results of the inventors, factors with significant correlation include: radiation, GHI, DHI, DNI, relative humidity, wind direction, dew point, pressure, temperature and solar zenith angle. In addition, although the solar zenith angle is negatively correlated, this factor still has a relatively good linear correlation. Among them, DHI (Direct Horizontal Irradiance) represents the direct solar radiation on the horizontal plane, that is, the radiation directly from the sun that is not scattered by the atmosphere. DNI (Direct Normal Irradiance) represents the direct solar radiation on the vertical plane, which is usually used for concentrating photovoltaic systems that track the sun. GHI (Global Horizontal Irradiance) represents the total solar radiation on the horizontal plane, including direct radiation and scattered radiation, but does not consider reflected radiation.

[0069] Photovoltaic power generation historical data includes historical power generation data, operating status and maintenance records of photovoltaic panels, etc. The above historical data can reflect the power generation trend of the performance of the photovoltaic system.

[0070] Step S2: preprocessing the photovoltaic power generation prediction data set.

[0071] As described above, in the embodiments of the present invention, satellite inversion data, meteorological data and photovoltaic power generation historical data from different sources are used for short-term photovoltaic power generation prediction. A slight change in the input data may lead to a significant change in the prediction output, affecting the accuracy of short-term photovoltaic power generation prediction. Therefore, the photovoltaic power generation prediction data set must be preprocessed to improve the accuracy of short-term photovoltaic power generation prediction.

[0072] In an embodiment of the present invention, the preprocessing includes missing value processing, data normalization, enhanced Dickey-Fuller test and autocorrelation test. Among them, the satellite inversion data, meteorological data and photovoltaic power generation historical data may all have data missing. In order to avoid the poor learning effect of the subsequent photovoltaic power generation short-term prediction hybrid model, the missing values ​​are filled using an interpolation method known in the art. At the same time, since the input data has different scales, it is necessary to convert the original data through a specific ratio. In an embodiment of the present invention, the Min-Max normalization method can be used for the above conversion. The above method belongs to a common method for data normalization in this field and will not be repeated here.

[0073] In the photovoltaic power generation short-term prediction hybrid model of the embodiment of the present invention, the CNN-LSTM module and the LSTM module both involve the application of time series models. Therefore, when preprocessing the data, it is necessary to first confirm whether the data is stationary data. The so-called stationary data means that the mean and variance of the exponential series do not differ significantly over time. If the data is not stationary, then the results estimated by this prediction model will be inaccurate. In the embodiment of the present invention, the enhanced Dickey-Fuller test is used to confirm that the photovoltaic power generation prediction data set y t Is it stationary data? If it is non-stationary data, the difference method is used to eliminate the dependence of the time series and convert it into stationary data.

[0074] In an embodiment of the present invention, the enhanced Dickey-Fuller test is specifically:

[0075] S21: Check the photovoltaic power generation prediction data set y according to the following formula t The unit root of :

[0076]

[0077] Among them, Δy t is the photovoltaic power generation prediction data set y t α is the parameter to be tested. If α=0, it means that the photovoltaic power generation prediction data set y t There is a unit root and it is non-stationary; is the lag term coefficient, p is the number of lag terms, βt is the time trend term, ∈ t is the error term;

[0078] S22: Use the least squares method to estimate α, β and

[0079] S23: Calculate the statistic of the enhanced Dickey-Fuller test, as shown in the following formula:

[0080]

[0081] in, is an estimate of α, yes The standard error of

[0082] S24: The calculated statistics ADF Sta Compared with the critical value, if ADF Sta is less than 0.05, then the photovoltaic power generation prediction data set y t For stationary data; if ADF Sta is greater than 0.05, then the photovoltaic power generation prediction data set y t is non-stationary data;

[0083] S25: For non-stationary data, the difference method is used to eliminate the photovoltaic power generation prediction data set y t The time series dependence of , then proceed to step S21 to continue calculating ADF Sta , until it is confirmed that the above data set is stable data.

[0084] After confirming the photovoltaic power generation prediction data set y t After the data is stabilized, the prediction factors of the hybrid model for short-term prediction of photovoltaic power generation need to be determined. That is, after performing the enhanced Dickey-Fuller test, the autocorrelation test is continued, specifically:

[0085] S26: Perform autocorrelation test according to the following formula:

[0086]

[0087] Among them, h is the lag time, cov is the covariance calculation, var(y t ) is the photovoltaic power generation prediction data set y t Variance in time series;

[0088] S27: Determine whether the above calculation result ACF(h) exceeds the preset confidence interval threshold. If it exceeds the preset confidence interval threshold, it is considered that the photovoltaic power generation prediction data set y t Satisfy the correlation requirements of time series.

[0089] In the embodiment of the present invention, y is obtained by the above formula: t and t-1 The correlation is 0.9, which is very close to 1, so we choose y t-1 As y t predictors of .

[0090] Step S3, constructing a photovoltaic power generation short-term prediction hybrid model, the photovoltaic power generation short-term prediction hybrid model includes a CNN-LSTM module, an LSTM module and a feature fusion module, and the output end of the CNN-LSTM module and the output end of the LSTM module are both connected to the feature fusion module.

[0091] Among them, the CNN-LSTM module is used to process the preprocessed satellite inversion data, and obtain a first result of short-term prediction of photovoltaic power generation based on the satellite inversion data; the LSTM module is used to process the preprocessed meteorological data and the photovoltaic power generation historical data, and obtain a second result of short-term prediction of photovoltaic power generation based on the meteorological data and the photovoltaic power generation historical data; the feature fusion module is used to fuse the first result of short-term prediction of photovoltaic power generation and the second result of short-term prediction of photovoltaic power generation to obtain a final prediction result.

[0092] In an embodiment of the present invention, the photovoltaic power generation short-term prediction hybrid model is divided into two branches to process different data types, and finally, the output of each branch is fused to obtain the final photovoltaic power generation prediction. The main reason is that this method can avoid the defects such as overfitting that may exist when using a single model. At the same time, the separate design of the LSTM module can increase the influence weight of meteorological data and photovoltaic power generation historical data in the model prediction process, thereby improving the accuracy and stability of the model prediction.

[0093] For machine learning models, the setting of hyperparameters is crucial because they affect the performance of the training algorithm and prevent the model from overfitting. Therefore, in this embodiment, grid search is used to optimize both the CNN-LSTM module and the LSTM module to find the best hyperparameters. Although the grid search process is time-consuming, after determining the ideal hyperparameters, the time for model training and testing is significantly reduced. In this embodiment, the final hyperparameters of the CNN-LSTM module and the LSTM module are as follows:

[0094] CNN-LSTM:

[0095]

[0096] LSTM:

[0097]

[0098] In the hybrid model for short-term prediction of photovoltaic power generation in an embodiment of the present invention, the CNN-LSTM module first extracts the spatial features of the satellite inversion data through the CNN layer, and maps the obtained feature map to the LSTM layer. The LSTM layer is used to extract time information for time series prediction. Finally, the first result of the short-term prediction of photovoltaic power generation is obtained through the fully connected layer.

[0099] Specifically, the specific process of the CNN-LSTM module obtaining the first result of the photovoltaic power generation short-term prediction is as follows:

[0100] Step S311: The CNN layer processes the satellite inversion data through convolution and activation functions:

[0101] O CNN =ReLU(W CNN *X+b CNN );

[0102] Wherein, X is the satellite inversion data, W CNN and b CNN They are the weights and biases of the CNN layer, * represents the convolution operation, and ReLU represents the rectified linear activation unit function;

[0103] Step S312: Use pooling operation to reduce feature dimension, that is:

[0104] P = Pooling (O CNN );

[0105] Among them, Pooling represents the pooling operation;

[0106] Step S313: The output P of the CNN layer is used as the input of the LSTM layer. The dimension of the output P is D×T, where D is the feature depth and T is the time series step. The LSTM layer processes and outputs the hidden state h of each time step. t :

[0107] h t =LSTM(P, h t-1 );

[0108] Among them, h t is the hidden state at time step t, LSTM represents the operation of the LSTM unit, and h t-1 is the hidden state at the previous time step;

[0109] Step S314: The hidden state h of the last time step of the LSTM layer T After processing through the fully connected layer, the final output is obtained Specifically:

[0110] Z=W FC H+b FC ;

[0111]

[0112] Among them, W FC and b FC They are the bias and weight of the fully connected layer respectively, and softmax is the activation function.

[0113] In an embodiment of the present invention, after receiving the meteorological data and the photovoltaic power generation history data, the LSTM module first combines the meteorological data and the photovoltaic power generation history data using timestamp information, and then performs a short-term photovoltaic power generation forecast on the combined data to obtain the second result of the short-term photovoltaic power generation forecast; wherein the meteorological data includes radiation, DHI, DNI, GHI, relative humidity, wind direction, dew point, pressure, temperature and solar zenith angle, and the photovoltaic power generation history data includes photovoltaic historical power generation power.

[0114] In a possible implementation, after the correspondence between the meteorological data and the photovoltaic power generation history data is determined by using the timestamp, as a simple method, the photovoltaic power generation history data can be directly attached to the end of the meteorological data to achieve a combination of the two.

[0115] After the CNN-LSTM module is used to obtain the first result of the photovoltaic power generation short-term prediction, and the LSTM module is used to obtain the second result of the photovoltaic power generation short-term prediction, the first result and the second result are input into the feature fusion module of the photovoltaic power generation short-term prediction hybrid model to obtain the final prediction result.

[0116] In the feature fusion module, feature fusion is first performed:

[0117]

[0118] in, is the output of the CNN-LSTM module, It is the output result of the LSTM module, and Concat represents the feature fusion operation;

[0119] Then, the fully connected layer of the feature fusion module maps the fused features to the final prediction result:

[0120]

[0121] in, is the predicted output, W f and b f are the weight and bias of the fully connected layer respectively.

[0122] Step S4, training the photovoltaic power generation short-term prediction hybrid model based on the photovoltaic power generation prediction data set, and iterating until the loss function converges;

[0123] Step S5, obtaining satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station;

[0124] Step S6: pre-process the satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station, and input them into the trained photovoltaic power generation short-term prediction hybrid model to obtain the photovoltaic power generation short-term prediction result of the target photovoltaic power station.

[0125] After obtaining accurate short-term forecast results for photovoltaic power generation, power operators can use them to conduct power dispatch and resource optimization to ensure the stable operation of the power grid and further improve the grid's ability to absorb photovoltaic energy.

[0126] like Figure 2 As shown, according to another embodiment of the present invention, the present invention also provides a photovoltaic power generation short-term prediction device based on satellite inversion data, the device comprising:

[0127] A data set construction module, wherein the data set construction module is used to establish a photovoltaic power generation prediction data set, wherein the photovoltaic power generation prediction data set includes satellite inversion data, meteorological data, and photovoltaic power generation historical data;

[0128] A preprocessing module, the preprocessing module is used to preprocess the photovoltaic power generation prediction data set;

[0129] A model building module, wherein the model building module is used to build a hybrid model for short-term prediction of photovoltaic power generation, wherein the hybrid model for short-term prediction of photovoltaic power generation includes a CNN-LSTM module, an LSTM module and a feature fusion module, wherein the output end of the CNN-LSTM module and the output end of the LSTM module are both connected to the feature fusion module;

[0130] The CNN-LSTM module is used to process the preprocessed satellite inversion data, and obtain a first result of photovoltaic power generation short-term prediction based on the satellite inversion data; the LSTM module is used to process the preprocessed meteorological data and the photovoltaic power generation historical data, and obtain a second result of photovoltaic power generation short-term prediction based on the meteorological data and the photovoltaic power generation historical data; the feature fusion module is used to fuse the first result of photovoltaic power generation short-term prediction and the second result of photovoltaic power generation short-term prediction to obtain a final prediction result;

[0131] A training module, the training module is used to train the photovoltaic power generation short-term prediction hybrid model based on the photovoltaic power generation prediction data set, and iterate until the loss function converges;

[0132] An acquisition module, the acquisition module is used to acquire satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station;

[0133] A prediction module is used to pre-process the satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station, and input them into the trained photovoltaic power generation short-term prediction hybrid model to obtain the photovoltaic power generation short-term prediction result of the target photovoltaic power station.

[0134] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer 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. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0135] The computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0136] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0137] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of boxes in the block diagram and / or the flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0138] The units involved in the embodiments of the present disclosure may be implemented by software or hardware, wherein the name of a unit does not limit the unit itself in some cases.

[0139] The above introduces the preferred embodiments of the present invention, which is intended to make the spirit of the present invention clearer and easier to understand, but is not intended to limit the present invention. All modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection outlined by the claims attached to the present invention.

Claims

1. A photovoltaic power generation short-term prediction method based on satellite inversion data, characterized in that: The method comprises: Step S1, establishing a photovoltaic power generation prediction data set, wherein the photovoltaic power generation prediction data set includes satellite inversion data, meteorological data, and photovoltaic power generation historical data; Step S2, preprocessing the photovoltaic power generation prediction data set; Step S3, constructing a photovoltaic power generation short-term prediction hybrid model, wherein the photovoltaic power generation short-term prediction hybrid model includes a CNN-LSTM module, an LSTM module and a feature fusion module, and the output end of the CNN-LSTM module and the output end of the LSTM module are both connected to the feature fusion module; The CNN-LSTM module is used to process the preprocessed satellite inversion data, and obtain a first result of photovoltaic power generation short-term prediction based on the satellite inversion data; the LSTM module is used to process the preprocessed meteorological data and the photovoltaic power generation historical data, and obtain a second result of photovoltaic power generation short-term prediction based on the meteorological data and the photovoltaic power generation historical data; the feature fusion module is used to fuse the first result of photovoltaic power generation short-term prediction and the second result of photovoltaic power generation short-term prediction to obtain a final prediction result; Step S4, training the photovoltaic power generation short-term prediction hybrid model based on the photovoltaic power generation prediction data set, and iterating until the loss function converges; Step S5, obtaining satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station; Step S6: After preprocessing the satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station, input them into the trained photovoltaic power generation short-term prediction hybrid model to obtain the photovoltaic power generation short-term prediction result of the target photovoltaic power station.

2. The photovoltaic power generation short-term prediction method based on satellite inversion data according to claim 1 is characterized in that: The preprocessing includes missing value processing, data normalization, enhanced Dickey-Fuller test and autocorrelation test.

3. The photovoltaic power generation short-term prediction method based on satellite inversion data according to claim 2 is characterized in that: The enhanced Dickey-Fuller test is specifically: S21: Check the photovoltaic power generation prediction data set y according to the following formula t The unit root of : Among them, Δy t is the photovoltaic power generation prediction data set y t α is the parameter to be tested. If α=0, it means that the photovoltaic power generation prediction data set y t There is a unit root and it is non-stationary; is the lag term coefficient, p is the number of lag terms, βt is the time trend term, ∈ t is the error term; S22: Use the least squares method to estimate α, β and S23: Calculate the statistic of the enhanced Dickey-Fuller test, as shown in the following formula: in, is an estimate of α, yes The standard error of S24: The calculated statistics ADF Sta Compared with the critical value, if ADF Sta is less than 0.05, then the photovoltaic power generation prediction data set y t For stationary data; if ADF Sta is greater than 0.05, then the photovoltaic power generation prediction data set y t is non-stationary data; S25: For non-stationary data, the difference method is used to eliminate the photovoltaic power generation prediction data set y t The time series dependence of , then proceed to step S21 to continue calculating ADF Sta , until it is confirmed that the above data set is stable data.

4. The photovoltaic power generation short-term prediction method based on satellite inversion data according to claim 3 is characterized in that: After performing the enhanced Dickey-Fuller test, the autocorrelation test is continued, specifically: S26: Perform autocorrelation test according to the following formula: Among them, h is the lag time, cov is the covariance calculation, var(y t ) is the photovoltaic power generation prediction data set y t Variance in time series; S27: Determine whether the above calculation result ACF(h) exceeds the preset confidence interval threshold. If it exceeds the preset confidence interval threshold, it is considered that the photovoltaic power generation prediction data set y t Satisfy the correlation requirements of time series.

5. The photovoltaic power generation short-term prediction method based on satellite inversion data according to claim 1 is characterized in that: The CNN-LSTM module first extracts the spatial features of the satellite inversion data through the CNN layer, and maps the obtained feature map to the LSTM layer. The LSTM layer is used to extract time information for time series prediction. Finally, the first result of the short-term prediction of photovoltaic power generation is obtained through the fully connected layer.

6. The photovoltaic power generation short-term prediction method based on satellite inversion data according to claim 5 is characterized in that: The specific process of the CNN-LSTM module obtaining the first result of the photovoltaic power generation short-term prediction is as follows: Step S311: The CNN layer processes the satellite inversion data through convolution and activation functions: O CNN =ReLU(W CNN *X+b CNN ); Wherein, X is the satellite inversion data, W CNN and b CNN They are the weights and biases of the CNN layer, * represents the convolution operation, and ReLU represents the rectified linear activation unit function; Step S312: Use pooling operation to reduce feature dimension, that is: P=Pooling(O CNN ); Among them, Pooling represents the pooling operation; Step S313: The output P of the CNN layer is used as the input of the LSTM layer. The dimension of the output P is D×T, where D is the feature depth and T is the time series step. The LSTM layer processes and outputs the hidden state h of each time step t. t : h t =LSTM(P,h t-1 ); Among them, h t is the hidden state at time step t, LSTM represents the operation of the LSTM unit, and h t-1 is the hidden state at the previous time step; Step S314: The hidden state h of the last time step of the LSTM layer T After processing through the fully connected layer, the final output is obtained Specifically: Z=W FC H+b FC ; Among them, W FC and b FC They are the bias and weight of the fully connected layer respectively, and softmax is the activation function.

7. The photovoltaic power generation short-term prediction method based on satellite inversion data according to claim 1 is characterized in that: After receiving the meteorological data and the photovoltaic power generation history data, the LSTM module first combines the meteorological data and the photovoltaic power generation history data using the timestamp information, and then performs a short-term photovoltaic power generation forecast on the combined data to obtain the second result of the short-term photovoltaic power generation forecast; wherein the meteorological data includes radiation, DHI, DNI, GHI, relative humidity, wind direction, dew point, pressure, temperature and solar zenith angle, and the photovoltaic power generation history data includes photovoltaic historical power generation power.

8. A photovoltaic power generation short-term prediction device based on satellite inversion data, characterized in that: The device comprises: A data set construction module, wherein the data set construction module is used to establish a photovoltaic power generation prediction data set, wherein the photovoltaic power generation prediction data set includes satellite inversion data, meteorological data, and photovoltaic power generation historical data; A preprocessing module, the preprocessing module is used to preprocess the photovoltaic power generation prediction data set; A model building module, wherein the model building module is used to build a hybrid model for short-term prediction of photovoltaic power generation, wherein the hybrid model for short-term prediction of photovoltaic power generation includes a CNN-LSTM module, an LSTM module and a feature fusion module, wherein the output end of the CNN-LSTM module and the output end of the LSTM module are both connected to the feature fusion module; The CNN-LSTM module is used to process the preprocessed satellite inversion data, and obtain a first result of photovoltaic power generation short-term prediction based on the satellite inversion data; the LSTM module is used to process the preprocessed meteorological data and the photovoltaic power generation historical data, and obtain a second result of photovoltaic power generation short-term prediction based on the meteorological data and the photovoltaic power generation historical data; the feature fusion module is used to fuse the first result of photovoltaic power generation short-term prediction and the second result of photovoltaic power generation short-term prediction to obtain a final prediction result; A training module, the training module is used to train the photovoltaic power generation short-term prediction hybrid model based on the photovoltaic power generation prediction data set, and iterate until the loss function converges; An acquisition module, the acquisition module is used to acquire satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station; A prediction module is used to pre-process the satellite inversion data, meteorological data and photovoltaic power generation history data of the target photovoltaic power station, and input them into the trained photovoltaic power generation short-term prediction hybrid model to obtain the photovoltaic power generation short-term prediction result of the target photovoltaic power station.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for short-term photovoltaic power generation prediction based on satellite inversion data as described in any one of claims 1 to 7 is implemented.

10. An electronic device, comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is used to execute the computer program to implement the photovoltaic power generation short-term prediction method based on satellite inversion data as described in any one of claims 1 to 7.

Images