Method, device and equipment for screening seasonal scale influence factors of wind and light resources and medium

Abstract

The invention discloses a wind and light resource seasonal scale influence factor screening method, device and equipment and a medium, which are used for solving the problems that a correlation analysis method is generally adopted for screening influence factors influencing a certain climate event in the existing climatic science, and the screened factors have correlation and do not have causality, so that the screening efficiency is low. And a prediction model established according to the method is low in precision. The method comprises the following steps: acquiring a wind and light resource seasonal scale influence factor, and preprocessing the wind and light resource seasonal scale influence factor to obtain a preprocessed influence factor; calculating a causal effect variable of the preprocessing influence factor and a prediction variable; screening an influence factor set from the preprocessed influence factors according to the causal effect variables; constructing a Bayesian network model by adopting the influence factor set; and generating an optimal influence factor set according to the Bayesian network model.

Description

Technical Field

[0001] The present invention relates to the technical field of wind and solar resources, and in particular to a method, device, equipment and medium for screening seasonal scale influencing factors of wind and solar resources. Background Art

[0002] With the increasing utilization of clean and renewable energy, the need for medium- and long-term forecasts of renewable energy on seasonal and annual scales has become increasingly important. This is to meet the demands for refined management of grid operations, ensure power system energy balance, develop medium- and long-term power generation plans, and improve renewable energy absorption capacity at a macro level. Currently, forecasts for electricity consumption on a monthly or longer scale primarily consider the strong correlation between electricity consumption and local climate characteristics. This involves statistical analysis of resources or electricity consumption, and then combining forecasts with data from different years and long-term climate forecasts. Applications are now placing higher demands on medium- and long-term renewable energy forecasts on scales beyond the monthly scale, even seasonal and annually. These forecasts require not only qualitative predictions of the occurrence of abnormal wind and solar resource events, but also quantitative predictions of the degree of abnormality. This relies on the seasonal wind and solar resource forecasting techniques established in the field of short-term climate forecasting.

[0003] Short-term climate prediction remains the most challenging problem in the field of climate forecasting. Currently, dynamical climate models are primarily used to simulate climate events and the climate system and provide forecasts. Correlation-based empirical statistical methods are also being applied to predict climate anomalies. Understanding climate variability helps construct correlation-based empirical models. In recent years, various artificial intelligence techniques have been used to improve the predictive performance of short-term climate forecasts, using correlation-based features as input. However, due to factors such as inaccurate physical parameterization schemes in climate models and errors in initial and boundary conditions, climate models are unable to accurately reproduce certain physical processes that lead to climate anomalies. These factors limit the model's predictive skill. To improve the predictive skill of climate models, it is necessary to develop correction models that incorporate physically well-defined influencing factors to calibrate the model's predictions. Furthermore, both empirical statistical and artificial intelligence models require relevant predictive (influencing) factors as input.

[0004] Current empirical models for predicting climate anomalies rely on correlations or non-causal dependency indicators. The inappropriate selection of seasonal factors affecting wind and solar resources leads to low accuracy in the prediction models built on these factors. Summary of the Invention

[0005] The present invention provides a method, device, equipment and medium for screening seasonal scale influencing factors of wind and solar resources, which is used to solve the technical problem that in existing climatology, the influencing factors that affect a certain climate event are generally screened by correlation analysis. The factors screened out often have only correlation but no causality, and the prediction model established based on this method has low accuracy.

[0006] The present invention provides a method for screening seasonal impact factors of wind and solar resources, comprising:

[0007] Obtain seasonal impact factors of wind and solar resources, and preprocess them to obtain preprocessed impact factors;

[0008] Calculating the causal effect variables of the pretreatment influencing factors and the predictor variables;

[0009] screening an influencing factor set from the pre-processing influencing factors according to the causal effect variable;

[0010] Constructing a Bayesian network model using the impact factor set;

[0011] An optimal impact factor set is generated according to the Bayesian network model.

[0012] Optionally, the seasonal scale influencing factors of wind and solar resources include sea temperature factors, atmospheric circulation factors and land surface factors;

[0013] The sea temperature factors include the El Nino NINO3.4 index, the El Nino NINO central region index, the sea temperature in the Indian Ocean key area, and the sea temperature in the South China Sea;

[0014] The atmospheric circulation factors include the Pacific Decadal Oscillation Index, the North Atlantic Oscillation Index, the Southern Oscillation Index, the Western Pacific Subtropical High Area and Intensity Index, the East Asian Trough Position Index, and the Polar Vortex Area and Intensity Index;

[0015] The land surface factors include snow cover in key areas of Eurasia and sea ice in key areas of the Arctic.

[0016] Optionally, the step of obtaining the seasonal impact factor of wind and solar resources and preprocessing the same to obtain the preprocessed impact factor includes:

[0017] Obtain seasonal impact factors of wind and solar resources;

[0018] Normalizing the seasonal impact factor of the wind and solar resources to obtain a normalized impact factor;

[0019] Performing empirical orthogonal function decomposition on the normalized impact factor to obtain a preprocessing impact factor.

[0020] Optionally, the step of calculating the causal effect variable of the pretreatment influencing factor and the predictor variable includes:

[0021] Obtain the degree of influence of pretreatment influencing factors on predictor variables;

[0022] Calculate the enhancement causal effect variable and the inhibition causal effect variable according to the degree of the effect;

[0023] The difference between the enhancement causal effect variable and the inhibition causal effect variable is calculated to obtain the causal effect variable of the pretreatment influencing factor.

[0024] Optionally, the Bayesian network model is:

[0025]

[0026] in, is the event generated by the Bayesian network structure S for the data sample D, is the posterior probability of the wind and solar resource impact factor sample under the Bayesian network, is the influencing factor sample, Y is the prior knowledge, is the conditional probability of each Bayesian network structure under several selected network structures, For data variables The joint probability contribution value.

[0027] Optionally, after the step of generating an optimal impact factor set according to the Bayesian network model, the method further includes:

[0028] The optimal influencing factor set is used to perform climate prediction.

[0029] The present invention also provides a device for screening seasonal impact factors of wind and solar resources, comprising:

[0030] The preprocessing module is used to obtain the seasonal impact factors of wind and solar resources and preprocess them to obtain preprocessed impact factors;

[0031] A causal effect variable calculation module, used to calculate the causal effect variables of the pretreatment influencing factors and the predictor variables;

[0032] An impact factor set screening module, configured to screen an impact factor set from the pre-processing impact factors according to the causal effect variable;

[0033] A Bayesian network model construction module, used to construct a Bayesian network model using the impact factor set;

[0034] The optimal impact factor set generation module is used to generate the optimal impact factor set according to the Bayesian network model.

[0035] Optionally, the preprocessing module includes:

[0036] The submodule for obtaining seasonal impact factors of wind and solar resources is used to obtain seasonal impact factors of wind and solar resources;

[0037] A normalization submodule, configured to normalize the seasonal impact factor of the wind and solar resources to obtain a normalized impact factor;

[0038] The preprocessing impact factor generation submodule is used to perform empirical orthogonal function decomposition on the normalized impact factor to obtain the preprocessing impact factor.

[0039] The present invention further provides an electronic device, comprising a processor and a memory:

[0040] The memory is used to store program code and transmit the program code to the processor;

[0041] The processor is used to execute the method for screening seasonal scale impact factors of wind and solar resources as described in any one of the above items according to the instructions in the program code.

[0042] The present invention also provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the method for screening seasonal scale impact factors of wind and solar resources as described in any one of the above items.

[0043] The above technical solution demonstrates the following advantages: The present invention obtains seasonal-scale influencing factors of wind and solar resources and preprocesses them to obtain preprocessed influencing factors; calculates causal effect variables for the preprocessed influencing factors; selects a set of influencing factors with causal relationships from the preprocessed influencing factors based on the causal effect variables; constructs a Bayesian network model using the influencing factor set; and generates an optimal set of influencing factors based on the Bayesian network model. This method thereby selects a set of influencing factors causally related to seasonal-scale wind and solar resource anomalies. Furthermore, the Bayesian network model establishes a dependency relationship between seasonal-scale wind and solar resources and the influencing factors to further extract the optimal set of influencing factors, thereby improving the accuracy of screening seasonal-scale influencing factors of wind and solar resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 A flowchart of the steps of a method for screening seasonal impact factors of wind and solar resources provided by an embodiment of the present invention;

[0046] Figure 2 A flowchart of a method for screening seasonal impact factors of wind and solar resources provided by another embodiment of the present invention;

[0047] Figure 3A flow chart of a method for screening seasonal impact factors of wind and solar resources provided by an embodiment of the present invention;

[0048] Figure 4 This is a structural block diagram of a device for screening seasonal impact factors of wind and solar resources provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The embodiments of the present invention provide a method, device, equipment and medium for screening seasonal scale influencing factors of wind and solar resources, which are used to solve the technical problem that in existing climatology, the influencing factors that affect a certain climate event are generally screened by correlation analysis. The factors screened out often have only correlation but no causality, and the prediction model established based on this has low accuracy.

[0050] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] See also Figure 1 , Figure 1 A flowchart of the steps of a method for screening seasonal impact factors of wind and solar resources provided by an embodiment of the present invention.

[0052] The present invention provides a method for screening seasonal impact factors of wind and solar resources, which may specifically include the following steps:

[0053] Step 101: Obtain seasonal impact factors of wind and solar resources, and preprocess them to obtain preprocessed impact factors;

[0054] In an embodiment of the present invention, a set of possible influencing factors of wind and solar resources on a seasonal scale may be selected according to the climatic characteristics of different zones, and preprocessed so that they have the same data format, facilitating subsequent identical processing.

[0055] The preprocessing method is as follows: normalize the seasonal scale influencing factors of wind and solar resources to obtain normalized influencing factors; perform empirical orthogonal function (EOF) decomposition on the normalized influencing factors, extract the main modes of these factors (variance contribution > 5%), and obtain preprocessed influencing factors.

[0056] Step 102, calculating the causal effect variables of the pre-processing influencing factors;

[0057] First, we obtain the degree of influence of the pre-processing influencing factors on the predicted variables, and then calculate the enhanced causal effect variables and the suppressed causal effect variables based on the degree of influence: climate events are divided into three states: positive anomaly (strong), normal (normal), and negative anomaly (weak), for example, the photovoltaic resource amount is abnormally large, the photovoltaic resource amount is normal, and the photovoltaic resource amount is abnormally small. The enhanced causal effect variables and the suppressed causal effect variables are:

[0058]

[0059]

[0060] in, Indicates impact factor It has an enhancing effect on wind and solar resources and is an enhanced causal effect variable; Indicates impact factor It has an inhibitory and weakening effect on wind and solar resources and is an inhibitory causal effect variable.

[0061] Then find the difference between the enhanced causal effect variable and the suppressed causal effect variable:

[0062]

[0063] Get the causal effect variable of the pre-treatment influencing factor. If , then the impact factor It has a strengthening and promoting effect on wind and solar resources; if , then the impact factor It has a weakening and inhibitory effect on wind and solar resources; if , it means that there is no influence relationship between the two. The larger the absolute value of The higher the impact on wind and solar resources.

[0064] Step 103, screening an impact factor set from pre-processed impact factors according to the causal effect variable;

[0065] The causal effect variable, CE, is used to determine whether a factor affects wind and solar resources and calculate the extent of that impact. CE is a combination of the conditional probabilities of two random events (X, Y), where Y represents the predicted variable (in this case, seasonal wind and solar resources) and X represents the influencing factor. For a given predicted variable, the causal effect variable, CE, can screen for the influencing factor with the greatest conditional probability dependency, thereby characterizing the causal effect.

[0066] Step 104, constructing a Bayesian network model using the impact factor set;

[0067] Step 105: Generate an optimal impact factor set based on the Bayesian network model.

[0068] Using the above-mentioned set of influencing factors that have a causal relationship with wind and solar resources , constructing a Bayesian network for learning, and finally obtaining the optimal set of seasonal influencing factors for wind and solar resources. Bayesian network is a graphical model used to represent the continuous probability distribution of a set of variables. It provides a natural way to represent causal information and discover the potential relationship between data. The description of Bayesian network consists of the following two parts: First, the network structure , It is a directed graph where each node represents a data variable , for Middle Node The set of parent nodes; the second is The local probability distribution of , Each element in is a data variable The conditional probability density of ,Depend on A Bayesian network is determined. Learning a Bayesian network is to find a Bayesian network model that can most realistically reflect the dependency relationship between data variables. The Bayesian network uses a graphical method to describe the relationship between data, with causal and probabilistic semantics. The factors selected have strong physical significance and are helpful for making predictions using the causal relationship between data. Combined with the causal effect The calculated Bayesian network can obtain the causal relationship between each influencing factor and wind and solar resources, and the causal relationship between each influencing factor, thereby screening out the optimal set of influencing factors.

[0069] The present invention obtains seasonal-scale influencing factors of wind and solar resources and preprocesses them to obtain preprocessed influencing factors; calculates causal effect variables for the preprocessed influencing factors; selects a set of influencing factors with causal relationships from the preprocessed influencing factors based on the causal effect variables; constructs a Bayesian network model using the influencing factor set; and generates an optimal set of influencing factors based on the Bayesian network model. This method thus selects a set of influencing factors that are causally related to seasonal-scale wind and solar resource anomalies. The Bayesian network model then establishes a dependency relationship between seasonal-scale wind and solar resources and the influencing factors to further extract the optimal set of influencing factors, thereby improving the accuracy of screening seasonal-scale influencing factors of wind and solar resources.

[0070] See also Figure 2 , Figure 2 This is a flowchart of a method for screening seasonal impact factors of wind and solar resources provided by another embodiment of the present invention. Specifically, the following steps may be included:

[0071] Step 201, obtaining seasonal impact factors of wind and solar resources;

[0072] Step 202: normalize the seasonal impact factor of wind and solar resources to obtain a normalized impact factor;

[0073] Step 203, performing empirical orthogonal function decomposition on the normalized impact factor to obtain a preprocessed impact factor;

[0074] In the embodiment of the present invention, the seasonal scale influencing factors of wind and solar resources include sea temperature factors, atmospheric circulation factors and land surface factors;

[0075] Sea temperature factors include the El Nino NINO3.4 index, the El Nino NINO central region index, the sea temperature in the key areas of the Indian Ocean, and the sea temperature in the South China Sea;

[0076] Atmospheric circulation factors include the Pacific Decadal Oscillation Index, the North Atlantic Oscillation Index, the Southern Oscillation Index, the Western Pacific Subtropical High Area and Intensity Index, the East Asian Trough Position Index, and the Polar Vortex Area and Intensity Index;

[0077] Land surface factors include snow cover in key areas of Eurasia and sea ice in key areas of the Arctic.

[0078] All factors were normalized before being input into the model for statistical inference. Because factors such as sea temperature in the Indian Ocean key region, sea temperature in the South China Sea, snow cover in the Eurasian key region, and sea ice in the Arctic key region are unevenly distributed, and area averages cannot represent their complex spatial variability, empirical orthogonal function (EOF) decomposition was used to extract the main modes of these factors (with variance contributions >5%) as model inputs.

[0079] Step 204, calculating the causal effect variables of the pre-treatment influencing factors and the predictor variables;

[0080] In one example, step 204 may include the following sub-steps:

[0081] S41, obtain the influence degree of pre-treatment influencing factors on the predicted variables;

[0082] S42, calculates the enhancement causal effect variable and the inhibition causal effect variable according to the degree of influence;

[0083] S43, calculate the difference between the enhanced causal effect variable and the suppressed causal effect variable to obtain the causal effect variable of the pretreatment influencing factor.

[0084] In the specific implementation, the expression of the causal effect variable CE is as follows:

[0085]

[0086] From the above formula, we can see that there are two different influencing conditions for the causal effect variable: , positive indicates the impact factor is the state of a positive event, and negative represents the impact factor The status of a negative event. There are two different consequences If the predicted amount Not affected by impact factor The relationship between the influencing conditions and the influencing results changes with the change of If the left side of the equation is greater than the right side, it means that the influencing condition It will enhance the results When the left side term is less than the right side term, it means the influence condition Results It has a weakening and inhibitory effect.

[0087] In determining a specific impact factor When determining whether there is an impact on wind and solar resources and the extent of the impact, the climatological method is used to divide climate events into three states: positive anomaly (strong), normal (normal), and negative anomaly (weak). For example, the amount of photovoltaic resources is abnormally large, the amount of photovoltaic resources is normal, and the amount of photovoltaic resources is abnormally small. Specifically, it can be expressed as:

[0088]

[0089] To quantify the impact factor Between wind and solar resources The strength of , further defines three indices according to the above formula:

[0090]

[0091]

[0092]

[0093] in, Indicates impact factor It has an enhancing effect on wind and solar resources. Indicates impact factor It has a suppressive and weakening effect on wind and solar resources. , then the impact factor It has a strengthening and promoting effect on wind and solar resources; if , then the impact factor It has a weakening and inhibitory effect on wind and solar resources; if , it means that there is no influence relationship between the two. The larger the absolute value of The higher the impact on wind and solar resources, the higher the impact on wind and solar resources. The measurement of causal effects can improve the ability of Bayesian networks to screen effective influencing factors.

[0094] Step 205 , screening a set of influencing factors with causal relationships from the pre-processed influencing factors according to the causal effect variables;

[0095] Step 206, constructing a Bayesian network model using the impact factor set;

[0096] Step 207, generating an optimal impact factor set according to the Bayesian network model;

[0097] In the embodiment of the present invention, a set of influencing factors having a greater impact on wind and solar resources is obtained according to the above steps. , construct a Bayesian network for learning, obtain a Bayesian network model that represents the dependency between influencing factors and wind and solar resources, and finally obtain the optimal set of influencing factors for wind and solar resources at the seasonal scale. The specific algorithm is as follows:

[0098] The learning process of Bayesian network is based on data samples and prior knowledge , find the posterior probability The largest Bayesian network structure process. From the Bayesian probability formula, we can know that:

[0099]

[0100] in, is the event generated by the Bayesian network structure S for the data sample D; D represents the set of influencing factors Y represents the possible physical mechanism affecting wind and solar resources; is the posterior probability of the wind and solar resource impact factor sample under the Bayesian network, is the joint probability of the wind and solar resource impact factor sample and the Bayesian network, is the prior probability of sample D.

[0101] First, several possible network structures are selected based on the possible physical mechanism (prior knowledge) of the impact factors on wind and solar resources; assuming that the parameter variable of the prior probability is Dirichlet distribution, the joint probability Just the index coefficient of the Dirichlet distribution To decide, then:

[0102]

[0103] in is the conditional probability of each Bayesian network structure under several selected possible network structures, For data variables The joint probability The contribution value of each The calculation is independent, and each network structure is calculated and selected. The largest structure is the optimal Bayesian network structure that matches the dependency between influencing factors and wind and solar resources. Therefore, the causal relationship between each influencing factor and wind and solar resources, as well as the causal relationship between each influencing factor, can be derived from the network diagram.

[0104] Step 208: Use the optimal influencing factor set to perform climate prediction.

[0105] After the optimal set of influencing factors is collected, it can be used for subsequent prediction model building to improve the prediction skills of seasonal wind and solar resources. The specific prediction model building method can adopt the methods commonly used by those skilled in the art, and the present invention does not impose specific limitations on this.

[0106] The present invention obtains seasonal-scale influencing factors of wind and solar resources and preprocesses them to obtain preprocessed influencing factors; calculates causal effect variables for the preprocessed influencing factors; selects a set of influencing factors with causal relationships from the preprocessed influencing factors based on the causal effect variables; constructs a Bayesian network model using the influencing factor set; and generates an optimal set of influencing factors based on the Bayesian network model. This method thus selects a set of influencing factors that are causally related to seasonal-scale wind and solar resource anomalies. The Bayesian network model then establishes a dependency relationship between seasonal-scale wind and solar resources and the influencing factors to further extract the optimal set of influencing factors, thereby improving the accuracy of screening seasonal-scale influencing factors of wind and solar resources.

[0107] See also Figure 3 For ease of understanding, the embodiments of the present invention are described below through specific examples:

[0108] Focusing on medium- and long-term (seasonal scale) new energy forecasts, we first determine the prediction variables as seasonal-scale wind and solar resource anomalies; consider and quantify the impact of climate systems that play an important role in seasonal-scale wind and solar resources, pre-select a set of influencing factors based on prior physical knowledge, normalize the data, and use empirical orthogonal function (EOF) decomposition to extract key modes; introduce advanced causal effect analysis in statistics, calculate the causal effect quantity CE, and preliminarily screen the set of influencing factors with causal relationships; combine the Bayesian network machine learning model to develop a causal strategy-hybrid machine learning model, and finally obtain the optimal set of influencing factors, which can improve the screening and classification accuracy of seasonal-scale influencing factors of wind and solar resources.

[0109] See also Figure 4 , Figure 4This is a structural block diagram of a device for screening seasonal impact factors of wind and solar resources provided by an embodiment of the present invention.

[0110] An embodiment of the present invention provides a device for screening seasonal impact factors of wind and solar resources, comprising:

[0111] The preprocessing module 401 is used to obtain the seasonal impact factor of wind and solar resources and preprocess it to obtain the preprocessed impact factor;

[0112] A causal effect variable calculation module 402 is used to calculate the causal effect variables of the pre-processing influencing factors and the predictor variables;

[0113] An impact factor set screening module 403 is used to screen an impact factor set having a causal relationship from pre-processed impact factors according to the causal effect variable;

[0114] A Bayesian network model construction module 404 is used to construct a Bayesian network model using an impact factor set;

[0115] The optimal impact factor set generation module 405 is used to generate the optimal impact factor set according to the Bayesian network model.

[0116] In this embodiment of the present invention, the preprocessing module 401 includes:

[0117] The submodule for obtaining seasonal impact factors of wind and solar resources is used to obtain seasonal impact factors of wind and solar resources;

[0118] The normalization submodule is used to normalize the seasonal impact factors of wind and solar resources to obtain normalized impact factors;

[0119] The preprocessing impact factor generation submodule is used to perform empirical orthogonal function decomposition on the normalized impact factor to obtain the preprocessing impact factor.

[0120] In this embodiment of the present invention, the causal effect variable calculation module 402 includes:

[0121] The influence degree acquisition submodule is used to obtain the influence degree of the pre-processing influencing factors on the predicted variables;

[0122] A submodule for calculating enhanced causal effect variables and suppressed causal effect variables, used for calculating enhanced causal effect variables and suppressed causal effect variables according to the degree of influence;

[0123] The causal effect variable calculation submodule is used to obtain the difference between the enhanced causal effect variable and the suppressed causal effect variable to obtain the causal effect variable of the pretreatment influencing factor.

[0124] The Bayesian network model is:

[0125]

[0126] in, is the event generated by the Bayesian network structure S for the data sample D, is the posterior probability of the wind and solar resource impact factor sample under the Bayesian network, is the influencing factor sample, Y is the prior knowledge, is the conditional probability of each Bayesian network structure under several selected network structures, For data variables The joint probability contribution value.

[0127] In an embodiment of the present invention, the following further comprises:

[0128] The prediction module is used to make climate predictions using the optimal set of influencing factors.

[0129] An embodiment of the present invention further provides an electronic device, the device including a processor and a memory:

[0130] The memory is used to store program codes and transmit the program codes to the processor;

[0131] The processor is used to execute the method for screening seasonal scale impact factors of wind and solar resources according to the instructions in the program code.

[0132] An embodiment of the present invention further provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the method for screening seasonal scale impact factors of wind and solar resources according to an embodiment of the present invention.

[0133] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0134] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0135] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, embodiments of the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0137] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0138] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0139] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0140] It should be noted that the user information (including but not limited to user device information, user personal information, 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 user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0141] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0142] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for screening seasonal impact factors of wind and solar resources, characterized in that: include: Obtain seasonal impact factors of wind and solar resources, and preprocess them to obtain preprocessed impact factors; Calculating the causal effect variables of the pretreatment influencing factors and the predictor variables; screening an influencing factor set from the pre-processing influencing factors according to the causal effect variable; Constructing a Bayesian network model using the impact factor set; An optimal impact factor set is generated according to the Bayesian network model.

2. The method according to claim 1, characterized in that The seasonal influencing factors of wind and solar resources include sea temperature factors, atmospheric circulation factors and land surface factors; The sea temperature factors include the El Nino NINO3.4 index, the El Nino NINO central region index, the sea temperature in the Indian Ocean key area, and the sea temperature in the South China Sea; The atmospheric circulation factors include the Pacific Decadal Oscillation Index, the North Atlantic Oscillation Index, the Southern Oscillation Index, the Western Pacific Subtropical High Area and Intensity Index, the East Asian Trough Position Index, and the Polar Vortex Area and Intensity Index; The land surface factors include snow cover in key areas of Eurasia and sea ice in key areas of the Arctic.

3. The method according to claim 1, characterized in that The step of obtaining the seasonal impact factor of wind and solar resources and preprocessing the same to obtain the preprocessed impact factor includes: Obtain seasonal impact factors of wind and solar resources; Normalizing the seasonal impact factor of the wind and solar resources to obtain a normalized impact factor; Performing empirical orthogonal function decomposition on the normalized impact factor to obtain a preprocessing impact factor.

4. The method according to claim 1, wherein The step of calculating the causal effect variable of the pretreatment influencing factor and the predictor variable includes: Obtain the degree of influence of pretreatment influencing factors on predictor variables; Calculate the enhancement causal effect variable and the inhibition causal effect variable according to the degree of the effect; The difference between the enhancement causal effect variable and the inhibition causal effect variable is calculated to obtain the causal effect variable of the pretreatment influencing factor.

5. The method according to claim 1, wherein The Bayesian network model is: in, is the event generated by the Bayesian network structure S for the data sample D, is the posterior probability of the wind and solar resource impact factor sample under the Bayesian network, is the influencing factor sample, Y is the prior knowledge, is the conditional probability of each Bayesian network structure under several selected network structures, For data variables The joint probability contribution value.

6. The method according to claim 1, characterized in that After the step of generating the optimal impact factor set according to the Bayesian network model, the method further includes: The optimal influencing factor set is used to perform climate prediction.

7. A device for screening seasonal impact factors of wind and solar resources, characterized in that: include: The preprocessing module is used to obtain the seasonal impact factors of wind and solar resources and preprocess them to obtain preprocessed impact factors; A causal effect variable calculation module, used to calculate the causal effect variables of the pretreatment influencing factors and the predictor variables; An impact factor set screening module, configured to screen an impact factor set from the pre-processing impact factors according to the causal effect variable; A Bayesian network model construction module, used to construct a Bayesian network model using the impact factor set; The optimal impact factor set generation module is used to generate the optimal impact factor set according to the Bayesian network model.

8. The device according to claim 7, characterized in that The preprocessing module includes: The submodule for obtaining seasonal impact factors of wind and solar resources is used to obtain seasonal impact factors of wind and solar resources; A normalization submodule, configured to normalize the seasonal impact factor of the wind and solar resources to obtain a normalized impact factor; The preprocessing impact factor generation submodule is used to perform empirical orthogonal function decomposition on the normalized impact factor to obtain the preprocessing impact factor.

9. An electronic device, characterized in that: The device includes a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the method for screening seasonal scale impact factors of wind and solar resources according to any one of claims 1 to 6 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store program code, and the program code is used to execute the method for screening seasonal scale impact factors of wind and solar resources according to any one of claims 1 to 6.

Images