CN118399378A - Method, apparatus, device, storage medium and program product for predicting generated power - Google Patents

Abstract

The present application relates to a generated power prediction method, apparatus, device, storage medium, and program product. The method comprises the following steps: acquiring target photovoltaic power generation data of a target area in a target period; inputting the target photovoltaic power generation data into a power prediction model to obtain a power generation value of a target region in a future period; the power prediction model is obtained by training a circulating neural network comprising an attention mechanism by adopting sample photovoltaic power generation data; the sample photovoltaic power generation data are obtained by performing correlation analysis and clustering processing on the historical photovoltaic power generation data. By adopting the method, the accuracy of the prediction of the generated power can be improved.

Description

Power generation prediction method, device, equipment, storage medium and program product

Technical Field

The present application relates to the field of photovoltaic power generation technology, and in particular to a method, device, equipment, storage medium and program product for predicting power generation.

Background technique

In recent years, photovoltaic power generation has developed rapidly, and photovoltaic installed capacity has increased year by year. Vigorously developing new energy sources such as photovoltaics has become an important measure to promote energy transformation. However, photovoltaic output is greatly affected by the weather, and has obvious nonlinearity, volatility and uncertainty. After being connected to the grid, it will affect the stability, safety and economy of the power grid. Therefore, accurate prediction of photovoltaic power generation can help plan the grid dispatching plan, reduce grid failures, and reduce losses. It is of great significance to the optimization of grid dispatching and the economic operation of photovoltaic power stations.

The current existing technology predicts photovoltaic power generation by directly predicting the output power using mathematical models. Common prediction methods include mathematical statistical prediction methods, artificial intelligence prediction methods, and hybrid prediction methods. However, the overall complexity of the current direct prediction model is still relatively simple, the stability is relatively weak, and the historical data of photovoltaic power generation is irregular, which makes the prediction accuracy of the current direct prediction model not high.

Summary of the invention

Based on this, it is necessary to provide a power generation prediction method, device, equipment, storage medium and program product to accurately predict the power generation in response to the above technical problems.

In a first aspect, the present application provides a method for predicting power generation, comprising:

Obtain target photovoltaic power generation data for a target area within a target period;

Input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period;

Among them, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In one embodiment, correlation analysis and clustering processing are performed on historical photovoltaic power generation data, including:

The correlation analysis of each historical photovoltaic power generation data is carried out to obtain the main meteorological indicators among each meteorological indicator; according to the main meteorological indicators, the historical photovoltaic power generation data are clustered to obtain sample photovoltaic power generation data.

In one embodiment, correlation analysis is performed on each historical photovoltaic power generation data to obtain the main meteorological indicators among the meteorological indicators, including:

According to the meteorological index data and the power generation value in each historical photovoltaic power generation data, the correlation coefficient between each meteorological index and the power generation is determined; according to the correlation coefficient between each meteorological index and the power generation, the main meteorological index is selected from each meteorological index.

In one embodiment, each historical photovoltaic power generation data is clustered according to the main meteorological indicators to obtain sample photovoltaic power generation data, including:

For each major meteorological indicator, cluster processing is performed on the meteorological indicator data corresponding to the major meteorological indicator in each historical photovoltaic power generation data to obtain the clustering processing results corresponding to the major meteorological indicator; according to the clustering processing results corresponding to each major meteorological indicator, the abnormal data in each historical photovoltaic power generation data is eliminated to obtain the sample photovoltaic power generation data.

In one embodiment, before performing correlation analysis on each historical photovoltaic power generation data to obtain the main meteorological indicators among the meteorological indicators, it includes:

Perform data cleaning on each historical photovoltaic power generation data.

In one of the embodiments, the recurrent neural network includes a convolutional gated recurrent unit (ConvGRU) network; wherein the ConvGRU network includes a ConvGRU layer, a normalization layer, an attention layer, and a fully connected layer.

In a second aspect, the present application also provides a power generation prediction device, comprising:

An acquisition module is used to acquire target photovoltaic power generation data of a target area within a target period of time;

The prediction module is used to input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period; wherein, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

Obtain target photovoltaic power generation data for a target area within a target period;

Input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period;

Among them, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

Obtain target photovoltaic power generation data for a target area within a target period;

Input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period;

Among them, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

Obtain target photovoltaic power generation data for a target area within a target period;

Input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period;

Among them, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

The above-mentioned power generation prediction method, device, equipment, storage medium and program product obtain high-quality data with a strong correlation with power generation by performing correlation analysis and clustering processing on historical photovoltaic power generation data; then, the processed data is used to train the power prediction model, so that the power prediction model can well learn the characteristics of data related to power generation, so that the power prediction model can accurately predict the power generation value based on standard photovoltaic power generation data, thereby improving the accuracy of power generation prediction.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

FIG1 is an application environment diagram of a power generation prediction method provided in an embodiment of the present application;

FIG2 is a schematic diagram of a flow chart of a power generation prediction method provided in an embodiment of the present application;

FIG3 is a schematic diagram of the structure of an attention mechanism provided in an embodiment of the present application;

FIG4 is a schematic diagram of the structure of a power prediction model provided in an embodiment of the present application;

FIG5 is a schematic diagram of a process for performing correlation analysis and clustering processing provided in an embodiment of the present application;

FIG6 is a structural block diagram of a power generation prediction device provided in an embodiment of the present application;

FIG7 is a structural block diagram of another power generation prediction device provided in an embodiment of the present application;

FIG8 is a diagram of the internal structure of a computer device provided in an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

The power generation prediction method provided in the embodiment of the present application can be applied in the application environment shown in Figure 1. Among them, the terminal 102 communicates with the server 104 through the network. The server 104 obtains various meteorological indicator data from the terminal 102 (sensor); further, the server 104 extracts the characteristics of each meteorological indicator data through the power prediction model, and predicts the power generation value based on the extracted characteristics. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices, and Internet of Things devices may include smart car-mounted devices, etc. The server 104 can be implemented as an independent server or a server cluster consisting of multiple servers.

In an exemplary embodiment, as shown in FIG. 2 , a method for predicting power generation is provided, which is described by taking the method applied to the server 104 in FIG. 1 as an example, and includes:

S201, obtaining target photovoltaic power generation data of a target area within a target period of time.

The photovoltaic power generation data includes a plurality of meteorological index data. Optionally, the meteorological index data can be acquired through sensor collection; different meteorological index data can be collected through different sensors.

Optionally, the meteorological indicator data includes, but is not limited to, air temperature data collected by a temperature sensor, wind speed data collected by a wind speed sensor, humidity data collected by a humidity sensor, and radiation intensity data collected by an irradiance sensor.

Exemplarily, the target area is any area where photovoltaic power generation equipment is deployed; the target period is a reference period for predicting future power generation, for example, it can be a period of time before the current time, such as 24 hours.

For example, the photovoltaic power generation power of area A in the next 24h-48h is predicted, and the meteorological index data of area A within 24h before the current time is obtained, such as temperature, wind speed, humidity, horizontal irradiance, diffuse horizontal irradiance and other meteorological index data.

S202, inputting the target photovoltaic power generation data into a power prediction model to obtain the power generation value of the target area in the future period.

The power prediction model is obtained by training a recurrent neural network with an attention mechanism using sample photovoltaic power generation data; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering on historical photovoltaic power generation data. The historical photovoltaic power generation data includes historical photovoltaic power generation data corresponding to multiple historical periods; each historical photovoltaic power generation data includes power generation value and multiple meteorological indicator data.

Optionally, the recurrent neural network in the power prediction model includes a convolutional gated recurrent unit (ConvGRU) network; wherein the ConvGRU network includes a ConvGRU layer, a normalization layer, an attention layer, and a fully connected layer. In the embodiment of the present application, the Adam optimization algorithm is selected, the learning rate is 0.01, and the epoch is 100.

It should be noted that the prediction accuracy of the ConvGRU network is not high when facing long time series data, and it is easy to cause the loss of important feature information. Therefore, the attention mechanism is introduced into the ConvGRU network to emphasize key feature information, avoid information loss and improve prediction accuracy.

Exemplarily, the target photovoltaic power generation data is input into a power prediction model, which extracts the temporal and spatial characteristics of the target photovoltaic power generation data, and further combines the historical photovoltaic power generation data to obtain the predicted power generation value of the target area in the future period.

The above-mentioned power generation prediction method obtains high-quality data with a strong correlation with power generation by performing correlation analysis and clustering processing on historical photovoltaic power generation data; then, the processed data is used to train the power prediction model, so that the power prediction model can well learn the characteristics of data related to power generation, so that the power prediction model can accurately predict the power generation value based on photovoltaic power generation data, thereby improving the accuracy of power generation prediction.

Based on the above embodiment, optionally, as shown in FIG3 , the attention mechanism is a mechanism that imitates cognitive attention, which can be regarded as consisting of a query (Q), a key (K) and a value (V), and the weighted sum of the values is obtained by calculation. By adding the attention mechanism to the recurrent neural network, the attention can be given different weights to the extracted feature information, thereby highlighting important information. The calculation process is divided into three stages:

1) Input Q and obtain the attention vector through dot product operation. The detailed process is as follows:

(1)

Among them, Q is the attention layer input vector (query vector), and Ki _is the key vector of certain features in the historical photovoltaic power generation data. Among them, imagine Q as a series of <K, V> data pairs.

2) In the fully connected layer, the Softmax function is used to normalize the weight coefficients obtained in the previous step to obtain the correlation between the query vector Q and a set of key-value pairs (K _i , V _i ). The detailed process is as follows:

(2)

3) Obtain the corresponding attention weight value through weighted summation. The detailed process is as follows:

(3)

Furthermore, as shown in Figure 4, the gated recurrent unit is a type of recurrent neural network (RNN), which controls input, memory and other information through a gating mechanism to make predictions at the current time step. The ConvGRU network is a variant of the RNN with a gating mechanism, which uses reset gates and update gates to control the flow of information and can effectively extract the spatial and temporal features of the input data.

Optionally, the detailed process of the ConvGRU network is as follows:

(4)

(5)

(6)

(7)

Among them, σ is the Sigmoid activation function; * represents the convolution operation; ⊙ represents element-wise multiplication; R _t is the reset gate; Z _t is the update gate; X _t is the input of the network layer at time t; H _t-1 is the hidden state at time t-1; is the candidate set; _Wxr , _Wxz , _Wxh , _Whr , _Whz , _Whh are weights; _br , _bz , _bh are biases.

Based on the above embodiment, the present application embodiment explains the above embodiment S202 in detail. Specifically, the present application embodiment involves a process of performing correlation analysis and clustering processing on historical photovoltaic power generation data, as shown in FIG5 , which specifically includes the following steps:

S501, performing correlation analysis on each historical photovoltaic power generation data to obtain the main meteorological indicators among the meteorological indicators.

Among them, the historical photovoltaic power generation data includes meteorological index data collected by various sensors during the historical period.

It should be noted that photovoltaic equipment converts solar energy, so the power generation of photovoltaic equipment is affected by weather conditions. For example, photovoltaic power generation has a certain relationship with meteorological indicators such as temperature (degrees Celsius), humidity (%), wind speed (m/s), diffuse irradiance (watts/square meter), and horizontal irradiance (watts/square meter), but different meteorological indicators have different degrees of influence on power generation. Therefore, it is necessary to analyze the correlation between various meteorological indicators and power generation.

Optionally, one possible implementation method of the embodiment of the present application is: performing correlation analysis on various historical photovoltaic power generation data through a correlation analysis algorithm, and finally determining the main meteorological indicators based on the results of the correlation analysis.

Another possible implementation method of the embodiment of the present application is to determine the correlation coefficient between each meteorological indicator and the power generation power based on the meteorological indicator data and the power generation power value in each historical photovoltaic power generation data; and select the main meteorological indicator from each meteorological indicator based on the correlation coefficient between each meteorological indicator and the power generation power.

For example, historical photovoltaic power generation data includes meteorological index data such as horizontal irradiance, diffuse horizontal irradiance, wind speed, wind direction, humidity, rainfall, and temperature. The Pearson correlation coefficient method is used for correlation analysis to calculate the correlation coefficient between each meteorological index and the power generation; according to the correlation coefficient, the meteorological index with a higher correlation coefficient is selected, for example, the meteorological index corresponding to the correlation coefficient higher than the coefficient threshold is selected; or the meteorological index is sorted in order from high to low in terms of the correlation coefficient, and the top 10% of the meteorological indexes are selected. For example, the Pearson correlation coefficient r(X _i , Y) (i=1, 2, ..., m) is used as a correlation analysis index to analyze the correlation between m meteorological indexes _Xi (i=1, 2, ..., m) and the power generation Y. The Pearson correlation coefficient calculation expression is:

(8)

Among them, _Xi is the data of the i-th meteorological index in the historical photovoltaic power generation data (i.e., meteorological index data); Y is the power generation value in the historical photovoltaic power generation data; is the average value of the data of the i-th meteorological index; is the average value of the power generation value in the historical photovoltaic power generation data; N is the number of single meteorological indicators, that is, the number of groups of historical photovoltaic power generation data; among them, the historical photovoltaic power generation data corresponding to a historical period is regarded as a group of data.

It should be noted that if r>0, it means that X and Y have a positive correlation; if r<0, it means that X and Y have a negative correlation. =1, X and Y are completely correlated, and the larger the absolute value of r, the stronger the correlation between X and Y. Optionally, further divisions can be made, for example, 0.8<r<1, representing extremely strong correlation; 0.6<r<0.8, representing strong correlation; 0.4<r<0.6, representing moderate correlation; 0.2<r<0.4, representing weak correlation; 0<r<0.2, representing extremely weak or no correlation.

S502, clustering the historical photovoltaic power generation data according to the main meteorological indicators to obtain sample photovoltaic power generation data.

Among them, clustering is to divide a data set into different classes or clusters according to a specific standard (such as distance), so that the similarity of data objects in the same cluster is as large as possible, and the difference of data objects in different clusters is as large as possible.

One possible implementation method of the embodiment of the present application is: for the main meteorological indicators, the similarity between the main meteorological indicator data can be measured, and the indicator data exceeding the distance threshold can be eliminated to obtain sample photovoltaic power generation data.

Another possible implementation method of the embodiment of the present application is: for each main meteorological indicator, clustering processing is performed on the meteorological indicator data corresponding to the main meteorological indicator in each historical photovoltaic power generation data to obtain the clustering processing result corresponding to the main meteorological indicator; according to the clustering processing results corresponding to each main meteorological indicator, the abnormal data in each historical photovoltaic power generation data is eliminated to obtain sample photovoltaic power generation data.

Exemplarily, based on the main meteorological indicators, density-based spatial clustering analysis of applications with noise (DBSCAN) is performed on each historical photovoltaic power generation data, and the parameter values of the domain radius and the minimum number of samples in the DBSCAN cluster analysis are determined. According to the distance measurement formula and the core points, the historical photovoltaic power generation data of the main meteorological indicators are traversed, and abnormal points such as isolated points and discrete points in the data are identified and eliminated, so as to finally obtain sample photovoltaic power generation data.

In an embodiment of the present application, the correlation between each meteorological indicator and the power generation is analyzed, each meteorological indicator is screened, and the meteorological indicators with a greater impact on the power generation are selected; further, a cluster analysis is performed on each meteorological indicator obtained after the screening, and outliers in the data are eliminated, thereby ensuring the validity and high quality of the sample photovoltaic power generation data.

On the basis of the above-mentioned embodiments, before the correlation analysis of each historical photovoltaic power generation data is performed to obtain the main meteorological indicators in each meteorological indicator, the embodiment of the present application specifically further includes: data cleaning of each historical photovoltaic power generation data. Among them, data cleaning includes processing missing data, outliers and noise, and performing normalization or standardization operations.

It should be noted that due to special circumstances such as equipment aging, system failure or human factors, the data recorded by the sensor may contain abnormal values and missing values.

Specifically, you can choose to use the mean, median, and interpolation methods to fill missing values, and promptly remove or correct duplicate values and erroneous values; or, you can delete blank values and set non-numeric values to 0; or you can use algorithms to clean the data, such as the Isolation Forest (iForest) algorithm; and then standardize and normalize the data to ensure the quality and accuracy of the data.

In the embodiment of the present application, by pre-cleaning the historical photovoltaic power generation data, the quality and accuracy of the data are improved to a certain extent.

It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the present application also provides a power generation prediction device for implementing the power generation prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more power generation prediction device embodiments provided below can refer to the limitations of the power generation prediction method above, and will not be repeated here.

In an exemplary embodiment, as shown in FIG6 , a power generation prediction device 1 is provided, comprising: an acquisition module 10 and a prediction module 20, wherein:

The acquisition module 10 is used to acquire target photovoltaic power generation data of a target area within a target period of time.

The prediction module 20 is used to input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period; wherein the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In one embodiment, as shown in FIG7 , the prediction module 20 specifically further includes:

The correlation analysis unit 21 is used to perform correlation analysis on each historical photovoltaic power generation data to obtain the main meteorological indicators among the meteorological indicators.

The clustering processing unit 22 is used to perform clustering processing on each historical photovoltaic power generation data according to the main meteorological indicators to obtain sample photovoltaic power generation data.

In one embodiment, the correlation analysis unit 21 is further configured to:

According to the meteorological index data and the power generation value in each historical photovoltaic power generation data, the correlation coefficient between each meteorological index and the power generation is determined; according to the correlation coefficient between each meteorological index and the power generation, the main meteorological index is selected from each meteorological index.

In one embodiment, the clustering processing unit 22 is specifically used for:

For each major meteorological indicator, cluster processing is performed on the meteorological indicator data corresponding to the major meteorological indicator in each historical photovoltaic power generation data to obtain the clustering processing results corresponding to the major meteorological indicator; according to the clustering processing results corresponding to each major meteorological indicator, the abnormal data in each historical photovoltaic power generation data is eliminated to obtain the sample photovoltaic power generation data.

In one embodiment, the power generation prediction device 1 specifically further includes:

The cleaning unit is used to clean the historical photovoltaic power generation data.

In one embodiment, the recurrent neural network includes a convolutional gated recurrent unit (ConvGRU) network, wherein the ConvGRU network includes a ConvGRU layer, a batch normalization layer, an attention layer, and a fully connected layer.

Each module in the above-mentioned power generation prediction device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG8. The computer device includes a processor, a memory, an input/output interface (Input/Output, referred to as I/O) and a communication interface. The processor, the memory and the input/output interface are connected via a system bus, and the communication interface is connected to the system bus via the input/output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store power generation prediction data. The input/output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a power generation prediction method is implemented.

Those skilled in the art will understand that the structure shown in FIG. 8 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

Obtain target photovoltaic power generation data for a target area within a target period;

Input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period;

Among them, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

The correlation analysis of each historical photovoltaic power generation data is carried out to obtain the main meteorological indicators among each meteorological indicator; according to the main meteorological indicators, the historical photovoltaic power generation data are clustered to obtain sample photovoltaic power generation data.

In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

According to the meteorological index data and the power generation value in each historical photovoltaic power generation data, the correlation coefficient between each meteorological index and the power generation is determined; according to the correlation coefficient between each meteorological index and the power generation, the main meteorological index is selected from each meteorological index.

In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

For each major meteorological indicator, cluster processing is performed on the meteorological indicator data corresponding to the major meteorological indicator in each historical photovoltaic power generation data to obtain the clustering processing results corresponding to the major meteorological indicator; according to the clustering processing results corresponding to each major meteorological indicator, the abnormal data in each historical photovoltaic power generation data is eliminated to obtain the sample photovoltaic power generation data.

In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

Perform data cleaning on each historical photovoltaic power generation data.

In one embodiment, the recurrent neural network includes a convolutional gated recurrent unit (ConvGRU) network, wherein the ConvGRU network includes a ConvGRU layer, a batch normalization layer, an attention layer, and a fully connected layer.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

Obtain target photovoltaic power generation data for a target area within a target period;

Input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period;

Among them, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

The correlation analysis of each historical photovoltaic power generation data is carried out to obtain the main meteorological indicators among each meteorological indicator; according to the main meteorological indicators, the historical photovoltaic power generation data are clustered to obtain sample photovoltaic power generation data.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

According to the meteorological index data and the power generation value in each historical photovoltaic power generation data, the correlation coefficient between each meteorological index and the power generation is determined; according to the correlation coefficient between each meteorological index and the power generation, the main meteorological index is selected from each meteorological index.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

For each major meteorological indicator, cluster processing is performed on the meteorological indicator data corresponding to the major meteorological indicator in each historical photovoltaic power generation data to obtain the clustering processing results corresponding to the major meteorological indicator; according to the clustering processing results corresponding to each major meteorological indicator, the abnormal data in each historical photovoltaic power generation data is eliminated to obtain the sample photovoltaic power generation data.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

Perform data cleaning on each historical photovoltaic power generation data.

In one embodiment, the recurrent neural network includes a convolutional gated recurrent unit (ConvGRU) network, wherein the ConvGRU network includes a ConvGRU layer, a batch normalization layer, an attention layer, and a fully connected layer.

In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

Obtain target photovoltaic power generation data for a target area within a target period;

Input the target photovoltaic power generation data into the power prediction model to obtain the power generation value of the target area in the future period;

Among them, the power prediction model is obtained by using sample photovoltaic power generation data to train a recurrent neural network including an attention mechanism; the sample photovoltaic power generation data is obtained by performing correlation analysis and clustering processing on historical photovoltaic power generation data.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

The correlation analysis of each historical photovoltaic power generation data is carried out to obtain the main meteorological indicators among each meteorological indicator; according to the main meteorological indicators, the historical photovoltaic power generation data are clustered to obtain sample photovoltaic power generation data.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

According to the meteorological index data and the power generation value in each historical photovoltaic power generation data, the correlation coefficient between each meteorological index and the power generation is determined; according to the correlation coefficient between each meteorological index and the power generation, the main meteorological index is selected from each meteorological index.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

For each major meteorological indicator, cluster processing is performed on the meteorological indicator data corresponding to the major meteorological indicator in each historical photovoltaic power generation data to obtain the clustering processing results corresponding to the major meteorological indicator; according to the clustering processing results corresponding to each major meteorological indicator, the abnormal data in each historical photovoltaic power generation data is eliminated to obtain the sample photovoltaic power generation data.

In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

Perform data cleaning on each historical photovoltaic power generation data.

In one embodiment, the recurrent neural network includes a convolutional gated recurrent unit (ConvGRU) network, wherein the ConvGRU network includes a ConvGRU layer, a batch normalization layer, an attention layer, and a fully connected layer.

It should be noted that the regional information (including but not limited to equipment information within the region, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the region or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims (10)

1. A method of generating power prediction, the method comprising:

acquiring target photovoltaic power generation data of a target area in a target period;

inputting the target photovoltaic power generation data into a power prediction model to obtain a power generation value of the target region in a future period;

The power prediction model is obtained by training a circulating neural network comprising an attention mechanism by adopting sample photovoltaic power generation data; the sample photovoltaic power generation data are obtained by performing correlation analysis and clustering treatment on historical photovoltaic power generation data.

2. The method of claim 1, wherein the historical photovoltaic power generation data comprises historical photovoltaic power generation data corresponding to a plurality of historical time periods; each historical photovoltaic power generation data comprises a power generation value and a plurality of meteorological index data;

correspondingly, the correlation analysis and clustering processing of the historical photovoltaic power generation data comprises the following steps:

performing correlation analysis on each historical photovoltaic power generation data to obtain main meteorological indexes in each meteorological index;

and clustering each historical photovoltaic power generation data according to the main meteorological indexes to obtain sample photovoltaic power generation data.

3. The method according to claim 2, wherein the performing correlation analysis on each historical photovoltaic power generation data to obtain a main weather indicator of each weather indicator comprises:

According to meteorological index data and power generation values in the historical photovoltaic power generation data, determining a correlation coefficient between each meteorological index and power generation power;

and selecting a main weather index from the weather indexes according to the correlation coefficient between each weather index and the generated power.

4. The method according to claim 2, wherein the clustering the historical photovoltaic power generation data according to the main meteorological index to obtain sample photovoltaic power generation data comprises:

clustering weather index data corresponding to the main weather index in each historical photovoltaic power generation data aiming at each main weather index to obtain a clustering result corresponding to the main weather index;

And removing abnormal data in each historical photovoltaic power generation data according to clustering processing results corresponding to each main meteorological index to obtain sample photovoltaic power generation data.

5. The method of claim 2, wherein prior to said correlating each historical photovoltaic power generation data to obtain a primary one of each weather indicator, the method further comprises:

and cleaning the data of each historical photovoltaic power generation data.

6. The method of claim 1, wherein the recurrent neural network comprises a convolutional gated recurrent unit ConvGRU network; wherein the ConvGRU network includes ConvGRU layers, a normalization layer, an attention layer, and a full connection layer.

7. A generated power prediction apparatus, characterized in that the apparatus comprises:

The acquisition module is used for acquiring target photovoltaic power generation data of the target area in a target period;

The prediction module is used for inputting the target photovoltaic power generation data into a power prediction model to obtain a power generation value of the target region in a future period; the power prediction model is obtained by training a circulating neural network comprising an attention mechanism by adopting sample photovoltaic power generation data; the sample photovoltaic power generation data are obtained by performing correlation analysis and clustering treatment on historical photovoltaic power generation data.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.