Energy power pattern prediction method and device and electronic equipment

Abstract

The invention discloses an energy power pattern prediction method and device and electronic equipment. The method comprises the following steps: acquiring current energy power data (including energy power trade data, energy power technical data and energy power strategy data) corresponding to a plurality of entities in a region; based on the current energy power data corresponding to the plurality of entities, determining energy power relation data corresponding to the plurality of dimensions; constructing a plurality of energy power relation networks based on the energy power relation data corresponding to the plurality of dimensions; according to the plurality of energy power relation networks, determining importance evaluation results corresponding to the plurality of dimensions; and determining an energy power pattern prediction result of the region in the prediction time period according to the importance evaluation results corresponding to the multiple dimensions. According to the method and the device, the technical problem of inaccurate energy and power pattern evaluation among entities caused by incomplete consideration of factors during energy and power relationship network analysis in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of energy power, in particular to an energy power pattern prediction method and device and electronic equipment. BACKGROUND

[0002] Analyzing and understanding the energy power pattern is of great significance for formulating energy strategies, ensuring energy security and guiding industrial investment. The energy power pattern can be regarded as a complex and multi-dimensional network system. The energy power relationship network is constructed by using the concept of graph theory, which can better depict the relationship between nodes in the network and provide the simplest and most persuasive research method for the cooperation relationship between energy power network entities. Network analysis is divided into overall measurement and local role measurement. The former emphasizes the overall analysis of network structure, taking all actors in the network as the research object. The latter focuses on the position and role analysis of the network, focusing on the influence of network relationship on individual behavior and the influence of individual behavior on network structure. In the related art, only a single dimension, such as geopolitical strategy and oil and gas trade, is used to analyze the energy power relationship network, which does not comprehensively consider the factors, resulting in inaccurate evaluation of the energy power pattern between entities and being not conducive to the evaluation of the development status of energy power and the formulation of energy power strategies.

[0003] At present, no effective solution has been proposed to solve the above problems. SUMMARY

[0004] The embodiments of the present application provide an energy power pattern prediction method and device and electronic equipment, at least to solve the technical problem that the factors are not comprehensive when the energy power relationship network is analyzed in the related art, resulting in inaccurate evaluation of the energy power pattern between entities.

[0005] According to an aspect of some embodiments of the present application, there is provided a method for predicting an energy power pattern, comprising: obtaining current energy power data corresponding to each of a plurality of entities in a region, wherein the current energy power data comprises energy power trade data, energy power technology data, and energy power strategy data of the corresponding entity in a current time period; determining energy power relationship data corresponding to each of a plurality of dimensions based on the current energy power data corresponding to each of the plurality of entities, wherein the energy power relationship data comprises a correlation degree of each of the plurality of entities in the corresponding dimension, and the plurality of dimensions comprises a correlation degree of an energy power trade dimension, an energy power technology dimension, and an energy power strategy dimension; constructing a plurality of energy power relationship networks based on the energy power relationship data corresponding to each of the plurality of dimensions, wherein the plurality of energy power relationship networks correspond to the plurality of dimensions one by one; determining an importance evaluation result corresponding to each of the plurality of dimensions based on the plurality of energy power relationship networks, wherein the importance evaluation result comprises an importance of each of the plurality of entities in the corresponding dimension; and determining an energy power pattern prediction result of the region in a prediction time period based on the importance evaluation result corresponding to each of the plurality of dimensions, wherein the energy power pattern prediction result is used to indicate an energy power development trend of the region as a whole.

[0006] Optionally, the constructing the plurality of energy power relationship networks based on the energy power relationship data corresponding to each of the plurality of dimensions comprises: constructing the plurality of energy power relationship networks based on the energy power relationship data corresponding to each of the plurality of dimensions, taking the plurality of entities as nodes, and taking the correlation degree of each of the plurality of entities in the corresponding dimension as an edge.

[0007] Optionally, the determining the importance evaluation result corresponding to each of the plurality of dimensions based on the plurality of energy power relationship networks comprises: converting any energy power relationship network in the plurality of energy power relationship networks into a weight matrix; solving a plurality of eigenvalues of the weight matrix and a feature vector corresponding to each of the plurality of eigenvalues; obtaining the importance evaluation result of any dimension based on the feature vector corresponding to the maximum eigenvalue in the plurality of eigenvalues; and obtaining the importance evaluation result corresponding to each of the plurality of dimensions in the same manner.

[0008] Optionally, the determining the energy power pattern prediction result of the region in the prediction time period based on the importance evaluation result corresponding to each of the plurality of dimensions comprises: obtaining a classification result corresponding to each of the plurality of dimensions based on the importance evaluation result corresponding to each of the plurality of dimensions and a plurality of support vector machine classifiers, wherein the classification result comprises a classification category of each of the plurality of entities in the corresponding dimension, and the plurality of dimensions and the plurality of support vector machine classifiers correspond to each other one by one; determining a target weight value corresponding to each of the plurality of dimensions; and obtaining the energy power pattern prediction result based on the classification result corresponding to each of the plurality of dimensions and the target weight value corresponding to each of the plurality of dimensions.

[0009] Optionally, the determining of the target weight value corresponding to each of the plurality of dimensions comprises: obtaining environmental change data in a preset historical time window, wherein the environmental change data at least comprises energy trade data of the region, a technology innovation rate, and an energy power policy stability index; determining a weight adjustment coefficient corresponding to each of the plurality of dimensions according to the environmental change data in the preset historical time window, wherein the weight adjustment coefficient of the energy power trade dimension is proportional to the energy trade data, the weight adjustment coefficient of the energy power technology dimension is proportional to the technology innovation rate, and the weight adjustment coefficient of the energy power policy dimension is proportional to the energy power policy stability index; and correcting a current weight coefficient corresponding to each of the plurality of dimensions based on the weight adjustment coefficient corresponding to each of the plurality of dimensions to obtain a target weight coefficient corresponding to each of the plurality of dimensions.

[0010] Optionally, the determining of the energy power relationship data corresponding to each of the plurality of dimensions based on the current energy power data corresponding to each of the plurality of entities comprises: in a case where the energy power trade data comprises energy power resource trade volume and power transmission capacity, obtaining the energy power relationship data corresponding to the energy power trade dimension by: performing normalization processing on the resource trade volume and the power transmission capacity between any two entities of the plurality of entities to obtain normalized resource trade volume and normalized power transmission capacity; and performing weighted summation operation on the normalized resource trade volume and the normalized power transmission capacity to obtain a correlation degree of the two entities in the energy power trade dimension; and obtaining the correlation degree of each two entities in the energy power trade dimension in the same manner.

[0011] Optionally, the determining of the energy power relationship data corresponding to each of the plurality of dimensions based on the current energy power data corresponding to each of the plurality of entities comprises: obtaining the energy power relationship data corresponding to the energy power technology dimension by: in a case where the energy power technology data comprises cross-entity literature citation number, key equipment trade amount, and R&D cooperation project fund related to energy power, performing normalization processing on the cross-entity literature citation number, the key equipment trade amount, and the R&D cooperation project fund between any two entities of the plurality of entities to obtain normalized cross-entity literature citation number, normalized key equipment trade amount, and normalized R&D cooperation project fund; and performing weighted summation operation on the normalized cross-entity literature citation number, the normalized key equipment trade amount, and the normalized R&D cooperation project fund to obtain a correlation degree of the two entities in the energy power technology dimension; and obtaining the correlation degree of each two entities in the energy power technology dimension in the same manner.

[0012] Optionally, based on the current energy power data corresponding to each of the plurality of entities, the energy power relationship data corresponding to each of the plurality of dimensions is determined, including: obtaining the energy power relationship data corresponding to the energy power strategy dimension by the following manner: in the case that the energy power strategy data includes the common bidding record and the strategy fit degree, the common bidding record and the strategy fit degree between any two entities of the plurality of entities are normalized respectively to obtain the normalized common bidding record and the normalized strategy fit degree, wherein the strategy fit degree is obtained based on the text similarity of the energy power strategy of any two entities; the normalized common bidding record and the normalized strategy fit degree are subjected to weighted summation operation to obtain the correlation degree of any two entities in the energy power strategy dimension; and the correlation degree of each two entities in the energy power strategy dimension is obtained by the manner of obtaining the correlation degree of any two entities in the energy power strategy dimension.

[0013] According to another aspect of the embodiment of the present application, an energy power pattern prediction device is further provided, including: an energy power data acquisition module, configured to acquire current energy power data corresponding to each of a plurality of entities in a region, wherein the current energy power data includes energy power trade data, energy power technology data and energy power strategy data of the corresponding entity in a current period; a relationship data acquisition module, configured to determine energy power relationship data corresponding to each of a plurality of dimensions based on the current energy power data corresponding to each of the plurality of entities, wherein the energy power relationship data includes the correlation degree of each two entities in the corresponding dimension, and the plurality of dimensions include the correlation degree of the energy power trade dimension, the energy power technology dimension and the energy power strategy dimension; a network construction module, configured to construct a plurality of energy power relationship networks based on the energy power relationship data corresponding to each of the plurality of dimensions, wherein the plurality of energy power relationship networks correspond to the plurality of dimensions one by one; an importance evaluation module, configured to determine importance evaluation results corresponding to each of the plurality of dimensions according to the plurality of energy power relationship networks, wherein the importance evaluation results include the importance corresponding to each of the plurality of entities in the corresponding dimension; and a prediction module, configured to determine energy power pattern prediction results of the plurality of entities in a prediction period according to the importance evaluation results corresponding to each of the plurality of dimensions, wherein the energy power pattern prediction results are used to indicate the mutual dependence degree of energy power between entities.

[0014] Optionally, the network construction module is further configured to construct the plurality of energy power relationship networks based on the energy power relationship data corresponding to each of the plurality of dimensions, taking the plurality of entities as nodes and the correlation degree of each two entities in the corresponding dimension as edges.

[0015] Optionally, the importance evaluation module comprises: a conversion submodule, configured to convert any of the plurality of energy power relationship networks into a weight matrix; a solving submodule, configured to solve a plurality of eigenvalues of the weight matrix and a plurality of eigenvectors corresponding to the plurality of eigenvalues respectively; an evaluation submodule, configured to obtain an importance evaluation result of any dimension based on an eigenvector corresponding to a maximum eigenvalue in the plurality of eigenvalues; and a first repeated obtaining submodule, configured to obtain importance evaluation results corresponding to a plurality of dimensions respectively in a manner of obtaining the importance evaluation result of any dimension.

[0016] Optionally, the prediction module comprises: a classification submodule, configured to obtain classification results corresponding to the plurality of dimensions respectively based on the importance evaluation results corresponding to the plurality of dimensions respectively and a plurality of support vector machine classifiers, wherein the classification results comprise classification categories of the plurality of entities in the corresponding dimensions, and the plurality of dimensions and the plurality of support vector machine classifiers correspond to each other in one-to-one manner; a weight determination submodule, configured to determine target weight values corresponding to the plurality of dimensions respectively; and a prediction submodule, configured to obtain an energy power pattern prediction result based on the classification results corresponding to the plurality of dimensions respectively and the target weight values corresponding to the plurality of dimensions respectively.

[0017] Optionally, the weight determination submodule comprises: a data obtaining unit, configured to obtain environmental change data in a preset historical time window, wherein the environmental change data at least comprises energy trade data of a region, a technology innovation rate and an energy power policy stability index; a coefficient adjustment unit, configured to determine weight adjustment coefficients corresponding to the plurality of dimensions respectively according to the environmental change data in the preset historical time window, wherein the weight adjustment coefficient of an energy power trade dimension is proportional to the energy trade data, the weight adjustment coefficient of an energy power technology dimension is proportional to the technology innovation rate, and the weight adjustment coefficient of an energy power policy dimension is proportional to the energy power policy stability index; and a weight correction unit, configured to correct current weight coefficients corresponding to the plurality of dimensions respectively based on the weight adjustment coefficients corresponding to the plurality of dimensions respectively to obtain target weight coefficients corresponding to the plurality of dimensions respectively.

[0018] Optionally, the relationship data obtaining module comprises: a first normalization submodule, configured to perform normalization processing on resource trade volume and power transmission capacity between any two of the plurality of entities respectively to obtain normalized resource trade volume and normalized power transmission capacity; a first weighting submodule, configured to perform weighted summation operation on the normalized resource trade volume and the normalized power transmission capacity to obtain a correlation degree of any two entities in the energy power trade dimension; and a second repeated obtaining submodule, configured to obtain the correlation degree of each two entities in the energy power trade dimension in a manner of obtaining the correlation degree of any two entities in the energy power trade dimension.

[0019] Optionally, the relationship data obtaining module comprises: a second normalization submodule, configured to, in the case that the energy power technology data comprises cross-entity literature citation number, key equipment trade volume, and R&D cooperation project fund related to energy power, normalize the cross-entity literature citation number, the key equipment trade volume, and the R&D cooperation project fund between any two entities in the plurality of entities respectively to obtain normalized cross-entity literature citation number, normalized key equipment trade volume, and normalized R&D cooperation project fund; a second weighting submodule, configured to perform weighted summation operation on the normalized cross-entity literature citation number, the normalized key equipment trade volume, and the normalized R&D cooperation project fund to obtain the correlation degree of any two entities in the energy power technology dimension; and a third repeated obtaining submodule, configured to obtain the correlation degree of each two entities in the energy power technology dimension in the manner of obtaining the correlation degree of any two entities in the energy power technology dimension.

[0020] Optionally, the relationship data obtaining module comprises: a third normalization submodule, configured to, in the case that the energy power strategy data comprises common bidding records and strategy fit degrees, normalize the common bidding records and the strategy fit degrees between any two entities in the plurality of entities respectively to obtain normalized common bidding records and normalized strategy fit degrees, wherein the strategy fit degree is obtained based on the text similarity of the energy power strategies of any two entities; a third weighting submodule, configured to perform weighted summation operation on the normalized common bidding records and the normalized strategy fit degrees to obtain the correlation degree of any two entities in the energy power strategy dimension; and a fourth repeated obtaining submodule, configured to obtain the correlation degree of each two entities in the energy power strategy dimension in the manner of obtaining the correlation degree of any two entities in the energy power strategy dimension.

[0021] According to another aspect of the embodiments of the present application, a nonvolatile storage medium is also provided, which stores a plurality of instructions adapted to be loaded and executed by a processor to implement the energy power pattern prediction method of any one of the embodiments.

[0022] According to another aspect of the embodiments of the present application, an electronic device is also provided, which comprises one or more processors and a memory, the memory being configured to store one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the energy power pattern prediction method of any one of the embodiments.

[0023] According to another aspect of the embodiments of the present application, a computer program product is also provided, which comprises a computer program, the computer program being executed by a processor to implement the steps of the energy power pattern prediction method of any one of the embodiments.

[0024] In the embodiment of the present application, the current energy power data corresponding to each of the plurality of entities in the region is obtained, wherein the current energy power data includes energy power trade data, energy power technology data and energy power strategy data of the corresponding entity in the current period; based on the current energy power data corresponding to each of the plurality of entities, energy power relationship data corresponding to each of the plurality of dimensions is determined, wherein the energy power relationship data includes the correlation degree of each two entities in the plurality of entities in the corresponding dimension, and the plurality of dimensions include the correlation degree of the energy power trade dimension, the energy power technology dimension and the energy power strategy dimension; based on the energy power relationship data corresponding to each of the plurality of dimensions, a plurality of energy power relationship networks are constructed, wherein the plurality of energy power relationship networks correspond one by one to the plurality of dimensions; according to the plurality of energy power relationship networks, an importance evaluation result corresponding to each of the plurality of dimensions is determined, wherein the importance evaluation result includes the importance of each of the plurality of entities in the corresponding dimension; according to the importance evaluation result corresponding to each of the plurality of dimensions, an energy power pattern prediction result of the region in the prediction period is determined, wherein the energy power pattern prediction result is used to indicate the energy power development trend of the region as a whole, so as to achieve the purpose of multi-dimensional quantitative analysis of the correlation degree of each entity (such as city, country, etc.) in the region in the energy power trade, technology and strategy, construction of the corresponding energy power relationship network, evaluation of the importance of each dimension, and finally comprehensive analysis and prediction of the energy power pattern between entities in a specific future period, so as to realize the technical effect of accurate prediction and evaluation of the evolution trend of the energy power pattern, and further solve the technical problem that the energy power relationship network analysis in the related art does not consider all factors, resulting in inaccurate evaluation of the energy power pattern between entities. BRIEF DESCRIPTION OF DRAWINGS

[0025] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation of the present application. In the drawings:

[0026] Figure 1 is a flowchart of an energy power pattern prediction method according to an embodiment of the present application;

[0027] Figure 2 is a schematic diagram of an optional relationship network converted into a weight matrix according to an embodiment of the present application;

[0028] Figure 3 is a flowchart of an optional energy power pattern prediction method according to an embodiment of the present application;

[0029] Figure 4 is a schematic diagram of an energy power pattern prediction device according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application in order to make the technical personnel in the technical field better understand the technical solutions. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by the person of ordinary skill in the art without creative work should belong to the protection scope of the present application.

[0031] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] According to the embodiments of the present application, a method embodiment of energy power pattern prediction is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0033] Figure 1 is a flowchart of the energy power pattern prediction method according to the embodiments of the present application, as shown in Figure 1 The method comprises the following steps:

[0034] Step S102, acquiring current energy power data corresponding to each of a plurality of entities in a region, wherein the current energy power data includes energy power trade data, energy power technology data and energy power strategy data of the corresponding entity in the current period;

[0035] Optionally, the current energy power data obtained in this step includes, but is not limited to, data reflecting trade volume, technical cooperation degree, and policy consistency of energy power (i.e., energy power trade data, energy power technology data, and energy power policy data). The plurality of entities can be, but are not limited to, a plurality of cities in a region and a plurality of countries. The original data can be obtained from various channels (such as public reports, industry databases, and enterprise reports), and then the original data is sorted and standardized to ensure the quality and applicability of the data. The current energy power data collected in this step can reflect comprehensive information of the current energy power activities of each entity, and provide basic data support for subsequent analysis.

[0036] In step S104, based on the current energy power data corresponding to each of the plurality of entities, energy power relationship data corresponding to each of a plurality of dimensions is determined, wherein the energy power relationship data includes an association degree of each two entities in the plurality of entities in the corresponding dimension, and the plurality of dimensions include an association degree of energy power trade dimension, energy power technology dimension, and energy power policy dimension.

[0037] Optionally, in this step, the association degree of each two entities in the trade, technology, and policy dimensions is calculated to quantify the energy power interaction between the entities in different dimensions and prepare data for subsequent network construction. The energy power relationship data between N entities in an entity can be obtained and preprocessed. The factors affecting the energy power pattern are a complex and interrelated system, and this embodiment analyzes and describes the energy power relationship between different entities from the energy power trade, technology, and policy dimensions.

[0038] In an optional embodiment, based on the current energy power data corresponding to each of the plurality of entities, the energy power relationship data corresponding to each of a plurality of dimensions is determined, including: in the case that the energy power trade data includes energy power resource trade volume and power transmission capacity, the energy power relationship data corresponding to the energy power trade dimension is obtained by: normalizing the resource trade volume and the power transmission capacity between any two entities in the plurality of entities respectively to obtain normalized resource trade volume and normalized power transmission capacity; performing weighted summation operation on the normalized resource trade volume and the normalized power transmission capacity to obtain an association degree of any two entities in the energy power trade dimension; and obtaining the association degree of each two entities in the energy power trade dimension in the manner of obtaining the association degree of any two entities in the energy power trade dimension.

[0039] Optionally, first, for the resource trade volume and transmission capacity in the energy power trade data, normalization processing is carried out. Normalization is a data processing method in statistics, and the purpose is to convert the original data to a unified scale, eliminate the influence of dimension or order of magnitude difference on subsequent analysis. Specifically, by calculating the relative size of the resource trade volume and transmission capacity of each entity relative to other entities, and then converting it to a standardized interval of [0, 1]. Next, the weighted sum operation is performed on the normalized resource trade volume and transmission capacity. This process aims to comprehensively consider the contribution of resource trade and transmission in energy power trade, and by specifying appropriate weight factors, the influence of one dimension or the balance of both can be emphasized. The result of the weighted sum gives the overall correlation degree of any two entities in the energy power trade dimension, that is, their closeness in the trade network. Through the above steps, the correlation degree of any two entities in the energy power trade dimension can be extended to calculate the correlation degree between all entity pairs in this dimension, forming a complete correlation matrix to describe the interaction strength of each entity in the entire regional trade network.

[0040] In the construction of energy power relationship network, especially in the process of trade network construction, the importance of data preprocessing, and how to quantitatively reflect the energy power trade correlation degree between entities through normalization and weighted sum operation. This way can ensure that even in the case of different data magnitudes or units, the closeness of the trade relationship between entities can be accurately reflected, providing a solid foundation for subsequent analysis and prediction work.

[0041] It should be noted that in the weighted sum operation process, the weight value is used to reflect the importance of the normalized resource trade volume and the normalized transmission capacity in the correlation degree calculation. The weight values corresponding to the normalized resource trade volume and the normalized transmission capacity can be the same or different, for example, the weights can be set to 1, or they can be set according to specific needs. For example, in the case of multiple entities being countries, considering that the cross-border transmission capacity is still relatively small at present, the weights of resource trade volume and transmission capacity can be set as 6:4.

[0042] Optionally, for any two entities, entity v i and entity v j energy power trade relationship, reflecting the flow of resources and electricity between entities, the trade relationship (i.e. the correlation degree of the energy power trade dimension) in a period of time is obtained as follows:

[0043] Cw ij =∑Rs ij +El ij

[0044] where Rsij El is the normalized trade volume of a single energy variety (e.g. coal, oil, gas). ij El is the normalized transmission capacity.

[0045] In an alternative embodiment, based on the current energy power data corresponding to each of the plurality of entities, the energy power relationship data corresponding to each of the plurality of dimensions is determined, including: obtaining the energy power relationship data corresponding to the energy power technology dimension by: in the case that the energy power technology data includes the number of cross-entity literature citations, the trade volume of key equipment, and the R&D cooperation project funds related to energy power, respectively normalizing the number of cross-entity literature citations, the trade volume of key equipment, and the R&D cooperation project funds between any two entities of the plurality of entities to obtain normalized cross-entity literature citations, normalized trade volume of key equipment, and normalized R&D cooperation project funds; and performing weighted summation operation on the normalized cross-entity literature citations, the normalized trade volume of key equipment, and the normalized R&D cooperation project funds to obtain the correlation degree of the two entities in the energy power technology dimension; and obtaining the correlation degree of each two entities in the energy power technology dimension by the above method.

[0046] Optionally, first, the data related to energy power technology, including the number of cross-entity literature citations, the trade volume of key equipment, and the R&D cooperation project funds, is normalized. By calculating the relative values of the technical data between entities and then converting them to a standard range (e.g. [0, 1]), the strength of the energy power technology relationship between entities can be effectively compared and analyzed. Then, the normalized cross-entity literature citations, the trade volume of key equipment, and the R&D cooperation project funds are subjected to weighted summation operation. The purpose of this step is to comprehensively consider the importance of the above three technical indicators to the technical relationship between entities. By assigning different weights to each indicator, the influence of a certain indicator on the technical correlation degree of entities can be emphasized, or balanced consideration between indicators can be achieved. The result of the weighted summation is the correlation degree of any two entities in the energy power technology dimension, reflecting their mutual dependence and cooperation in technology. By using the above method to calculate the correlation degree of any two entities in the energy power technology dimension, the technical correlation degree between all pairs of entities can be further calculated, and a comprehensive technical relationship data matrix can be constructed. This matrix describes the interaction of each entity in the entire technical network, providing detailed and quantitative technical information for subsequent energy power pattern prediction.

[0047] It should be noted that in the weighted summation operation, the weight value is used to reflect the importance of the normalized cross-entity literature citation number, the normalized key equipment trade volume, and the normalized R&D cooperation project fund in the correlation calculation. The weight values corresponding to the normalized cross-entity literature citation number, the normalized key equipment trade volume, and the normalized R&D cooperation project fund can be the same or different. For example, the weights can be set to 1, or they can be set according to specific needs.

[0048] Optionally, for any two entities, i.e., entities v i and v j , the energy technology relationship between them reflects the dependence relationship and cooperation dynamics of the entities in the energy and power industry chain. The technology relationship in a period of time (i.e., the correlation degree of the energy and power technology dimension) is obtained in the following manner:

[0049] Tw ij =∑Po ij +∑Eq ij +∑Ob ij

[0050] wherein Po ij represents the normalized cross-border literature (such as patents) citation number in a specific field (such as photovoltaic technology, power grid technology, and digital technology), Eq ij represents the normalized single key equipment trade volume (such as photovoltaic modules, wind power equipment, and gas turbines), and Ob ij represents the normalized single R&D cooperation project fund.

[0051] In an optional embodiment, based on the current energy and power data corresponding to each of the plurality of entities, the energy and power relationship data corresponding to each of the plurality of dimensions is determined, including: obtaining the energy and power relationship data corresponding to the energy and power strategy dimension in the following manner: in the case that the energy and power strategy data includes common bidding records and strategy fit degrees, the common bidding records and the strategy fit degrees between any two entities of the plurality of entities are normalized respectively to obtain normalized common bidding records and normalized strategy fit degrees, wherein the strategy fit degree is obtained based on the text similarity of the energy and power strategies of the any two entities; and performing a weighted summation operation on the normalized common bidding records and the normalized strategy fit degrees to obtain the correlation degree of the any two entities in the energy and power strategy dimension; and obtaining the correlation degree of each two entities in the energy and power strategy dimension in the manner of obtaining the correlation degree of the any two entities in the energy and power strategy dimension.

[0052] Optionally, in the energy power policy dimension, two core data considered are the common bidding records and the policy fit degree. The common bidding records reflect the cooperation history and frequency between entities in energy projects, while the policy fit degree is calculated based on the similarity of energy power policy texts between entities, embodying the consistency of policy orientation. Normalization is performed on these two data, aiming to convert them to the same numerical range, facilitating comparison and subsequent weighted summation operation. The normalized data can more fairly and intuitively reflect the closeness of policy relationship between entities. By weighted summation, the normalized common bidding records and the normalized policy fit degree are combined to obtain the correlation degree of any two entities in the energy power policy dimension. The weighted summation considers the respective weights of the common bidding records and the policy fit degree, which can be obtained based on historical data analysis or set according to expert evaluation, to reflect the different contributions of these two indicators to the policy dimension correlation degree. The result of weighted summation provides a quantitative indicator, clearly showing the cooperation level and consistency level between entities in the policy layer. The above method is used to calculate the policy dimension correlation degree between all entity pairs. By repeatedly executing the above calculation process, a matrix containing all entity policy relationships can be generated to describe the overall policy network. Through normalization and weighted summation operation, the relationship strength between entities in the energy power policy dimension within the region can be effectively quantified, providing data support for constructing the policy network.

[0053] It should be noted that in the weighted summation operation process, the weight value is used to reflect the importance of the normalized common bidding records and the normalized policy fit degree in the correlation degree calculation. The respective weight values of the normalized common bidding records and the normalized policy fit degree can be the same or different, for example, the weights can be set to 1, or they can be set according to specific scene needs.

[0054] Optionally, for the energy policy relationship between any two entities, i.e., entity v i and entity v j , reflecting the energy policy influence and consistency, the policy relationship in a period of time (i.e., the correlation degree of the energy power policy dimension) is obtained as follows:

[0055] Pw ij =∑Og ij +∑Sm ij

[0056] where Og ij is the common bidding record in a project, and Sm ij is the fit degree of a rule, represented by the text similarity of policy files between entities.

[0057] In step S106, a plurality of energy power relationship networks are constructed based on the energy power relationship data corresponding to the plurality of dimensions respectively, wherein the plurality of energy power relationship networks correspond to the plurality of dimensions one by one.

[0058] Optionally, in this step, three independent energy power relationship networks are created, corresponding to the energy power trade, technology and strategy dimensions respectively. The nodes in the network can represent entities (such as cities, countries, etc.), the edges represent the energy power relationship between entities, and the weight of the edge reflects the strength or closeness of the relationship, so as to realize the visualization of the abstract energy power relationship.

[0059] In an optional embodiment, constructing a plurality of energy power relationship networks based on energy power relationship data corresponding to a plurality of dimensions respectively comprises: constructing a plurality of energy power relationship networks based on a plurality of energy power relationship data corresponding to a plurality of dimensions respectively, taking a plurality of entities as nodes, and taking the correlation degree of each two entities in the corresponding dimension as an edge.

[0060] Optionally, first, the energy power relationship data in the trade, technology and strategy dimensions is used to quantify the interaction degree between different entities in each dimension. When constructing the network, each entity participating in the analysis is defined as a node in the network. The node represents a basic constituent unit in the energy power pattern, i.e. each geographical or economic entity. The correlation degree of each two entities in the corresponding dimension is taken as an edge, which is used to represent the connection between the two entities, and the weight of the edge is determined according to the correlation degree of the two entities in a certain dimension (one of trade, technology and strategy). Specifically, if two entities have a large amount of interaction in energy power trade, the edge connecting the two nodes in the trade network will have a high weight, and vice versa. Based on the relationship data in different dimensions, three independent networks, i.e. trade network, technology network and strategy network, are constructed. Each network independently reflects the interaction and correlation degree between entities in the dimension. For example, the energy power relationship network between entities within an entity is constructed in layers. The nodes v i in the energy power relationship network represent entities, and the edges e ij represent the connection relationship between entities. According to the attributes of the edges, the energy power relationship network is divided into three layers, i.e. trade, technology and strategy, and three types of undirected relationship networks (i.e. energy power relationship networks) G c (V, E), G t (V, E) and G p (V, E) are constructed, wherein G c (V, E) represents the trade network, G t (V, E) represents the technology network, and G p (V, E) represents the strategy network.

[0061] In the above manner, by abstracting entities such as cities and countries into nodes, quantifying the interactions between entities and representing them through the weights of edges, the method of the present embodiment constructs multiple energy and power relationship networks, each of which focuses on revealing the internal connections of the energy and power pattern in different dimensions. This network structure not only provides a way to visualize the energy and power relationship, but also lays a solid foundation for subsequent analysis steps (such as importance evaluation). The construction of the network enables the prediction model to better understand the nature of the energy and power pattern between entities, thereby making more accurate predictions and providing strong support for decision-makers.

[0062] In step S108, according to the multiple energy and power relationship networks, the importance evaluation results corresponding to the multiple dimensions are determined respectively, wherein the importance evaluation results include the importance of each entity in the corresponding dimension.

[0063] Optionally, in this step, by further analyzing each energy and power relationship network, the importance of each entity in each dimension is determined, which is used for subsequent evaluation of the importance of each dimension in determining the energy and power pattern.

[0064] In an optional embodiment, according to the multiple energy and power relationship networks, determining the importance evaluation results corresponding to the multiple dimensions respectively includes: converting any energy and power relationship network in the multiple energy and power relationship networks into a weight matrix; solving multiple eigenvalues of the weight matrix and the eigenvectors corresponding to the multiple eigenvalues respectively; obtaining the importance evaluation result of any dimension based on the eigenvector corresponding to the maximum eigenvalue in the multiple eigenvalues; and obtaining the importance evaluation results corresponding to the multiple dimensions respectively in the manner of obtaining the importance evaluation result of any dimension.

[0065] Optionally, first, each energy power relationship network constructed based on different dimensions is converted into a weight matrix. The weight matrix is a matrix form representation, where each element corresponds to the weight of the edge between two nodes in the network, and if there is no direct connection between the two nodes, the element in the matrix is 0. The purpose of converting into a weight matrix is to further process mathematically and algorithmically, facilitating the subsequent calculation of eigenvalues and eigenvectors. Perform eigenvalue decomposition on the generated weight matrix. Eigenvalue decomposition can reveal the intrinsic properties of the matrix. By solving the eigenvalues and eigenvectors of the weight matrix, the most important direction and information in the matrix can be found. The eigenvalue represents the scaling ratio of each eigenvector under matrix transformation, and the large eigenvalue corresponds to the direction with large information amount or important structure in the weight matrix. The eigenvector represents the stable direction of the weight matrix transformation, and for the eigenvector corresponding to the maximum eigenvalue, it can highlight the distribution and role of important nodes in the network. The importance evaluation result is obtained based on the eigenvector corresponding to the maximum eigenvalue. This means that this eigenvector is used to quantify and evaluate the relative importance of each entity in a certain dimension. The eigenvector corresponding to the maximum eigenvalue reveals the nodes (i.e. entities) with the most influence or centrality in the network, which play a key role in connecting and influencing other nodes. By analyzing this eigenvector, the relative importance score of each entity in a specific dimension (such as trade, technology, strategy) can be obtained. This score can be used in subsequent prediction models as a key input parameter for predicting changes in the energy power pattern. The importance evaluation result in any dimension is obtained by using the above method, and then it is extended to multiple dimensions to obtain a comprehensive importance evaluation atlas. This means that by analyzing the trade network, technology network and strategy network respectively, the importance in each dimension can be evaluated independently, and then the evaluation results are integrated to provide a multi-angle perspective for the prediction of the energy power pattern, enhancing the accuracy and reliability of the prediction.

[0066] Optionally, Figure 2 is an optional schematic diagram of converting a relationship network into a weight matrix according to an embodiment of the present application. The energy power relationship network is converted into a weight matrix, and the eigenvalues and eigenvectors are solved. The weight reflects the importance of the corresponding edge in the entire relationship network, and based on the energy power relationship data, the weight is assigned as follows: if there is an edge connection between entity nodes v i and v j in the relationship network V and there is energy trade, then e ij =Cw ij ; if there is no edge connection between entity nodes v i and v j , then e ij =0, so the N×N trade weight matrix CW t in the t time period can be obtained. Similarly, the technology weight matrix TWt , a policy weight matrix PW t . The maximum eigenvalue λ max and the corresponding eigenvector are solved for the above matrix respectively, the eigenvector reveals the importance of each node (i.e. entity) in different networks, and the eigenvector of the maximum eigenvalue obtained (i.e. the importance evaluation result) can be used as the input sample of the subsequent classifier.

[0067] In step S110, the energy power pattern prediction result of the region in the prediction period is determined according to the importance evaluation result corresponding to each of the plurality of dimensions, wherein the energy power pattern prediction result is used to indicate the energy power development trend of the region as a whole.

[0068] Optionally, in this step, a prediction model can be established by combining a machine learning method to predict the energy power pattern prediction result, so as to predict the energy power development trend of the region as a whole in the future period of time. The development trend includes an upward trend or a downward trend. The upward trend can represent that the development and cooperation of the region as a whole in the energy power aspect are in a positive trend, and the development and cooperation are closer. The downward trend can represent that the development and cooperation of the region as a whole in the energy power aspect are in a negative trend, and the closeness of the development and cooperation is relatively reduced. The above method provides an accurate and comprehensive energy power pattern prediction method, which not only considers three key dimensions of trade, technology and strategy, but also dynamically adjusts the importance of each dimension of the entire region, so that the prediction result is more in line with the actual situation, and provides strong support for energy strategy making, investment decision and cooperation planning between entities.

[0069] In an optional embodiment, determining the energy power pattern prediction result of the region in the prediction period according to the importance evaluation result corresponding to each of the plurality of dimensions comprises: based on the importance evaluation result corresponding to each of the plurality of dimensions, using a plurality of support vector machine classifiers to obtain a plurality of classification results corresponding to each of the plurality of dimensions, wherein the classification result includes the classification category of the plurality of entities in the corresponding dimension, and the plurality of dimensions and the plurality of support vector machine classifiers correspond one by one; determining a target weight value corresponding to each of the plurality of dimensions; and based on the classification result corresponding to each of the plurality of dimensions and the target weight value corresponding to each of the plurality of dimensions, obtaining the energy power pattern prediction result.

[0070] Optionally, for each dimension of the energy power relationship (energy power trade, technology, strategy), a support vector machine classifier is created and trained respectively. Support vector machine is a supervised learning model used for classification and regression analysis, which is particularly suitable for classification problems in high-dimensional space. Since the importance evaluation result of each dimension reveals the relative importance of entities in that dimension, by training the support vector machine (SVM) classifier of a specific dimension, the classification can be based on the importance of all entities in the corresponding dimension within the region, and the influence of the dimension on the overall future energy power pattern of the region can be predicted. Each SVM classifier will output a classification result according to the trained model, which is used to indicate the overall regional tendency (positive or negative) for that dimension. For example, for the energy power trade dimension, a positive tendency can represent a positive attitude towards cooperation and investment in energy power trade in the region as a whole in the future period, and a negative tendency can represent a negative attitude towards cooperation and investment in energy power trade by all entities in the future period; for the energy power technology dimension, a positive tendency can represent a positive attitude towards cooperation and investment in energy power technology in the region as a whole in the future period, and a negative tendency can represent a negative attitude towards cooperation and investment in energy power technology in the region as a whole in the future period; for the energy power strategy dimension, a positive tendency can represent a positive attitude towards cooperation and investment in energy power strategy in the region as a whole in the future period, and a negative tendency can represent a negative attitude towards cooperation and investment in energy power strategy in the region as a whole in the future period. While classifying and predicting each dimension, the target weight value of each dimension also needs to be determined. These weight values reflect the relative importance of different dimensions in predicting the energy power pattern. The determination of the weight can be pre-set, or it can be dynamically changed to ensure the accuracy and reliability of the prediction result. Finally, the classification results of each dimension are combined with the corresponding dimension weight values, and the weighted average method is used to obtain the energy power pattern prediction result of the region as a whole. This result not only considers the classification categories of the region as a whole in the trade, technology and strategy dimensions, but also combines the importance of these dimensions, so as to comprehensively and accurately predict the development trend of the energy power pattern of the region as a whole in the future period.

[0071] By the above mode, the energy and power relationship data in different dimensions are classified and predicted by using the support vector machine classifier, and the target weight value is determined by combining the importance evaluation result of each dimension, which can comprehensively consider and evaluate the interaction and influence of entities in trade, technology, strategy and other aspects, so that the predicted energy and power pattern is more in line with the actual situation, has stronger prediction ability and decision support value. This multi-dimensional and multi-classifier prediction method can not only improve the comprehensiveness of prediction, but also ensure the flexibility and accuracy of prediction results through weight value adjustment, which can effectively respond to the dynamic changes of energy market and policy environment.

[0072] Optionally, the plurality of support vector machine classifiers are obtained by training the corresponding initial classifiers based on the historical importance evaluation results and the corresponding historical classification results of the plurality of dimensions. The support vector machine performs binary classification on data in a supervised learning manner, and solves the maximum interval hyperplane of the learning sample. The energy and power trade, technology and strategy three types of classification models (i.e. support vector machine classifiers) are trained. For the feature sample of the prediction period, the support vector machine classification decision functions fc(w ij ), ft(w ij ), and fp(w ij ) are calculated, and the classification results of each dimension in the prediction period are further obtained. Based on the classification results, the final energy and power pattern prediction result can be determined by a weighted voting mechanism in the following form: f t+k = 0.4fc(w) + 0.3ft(w) + 0.3fp(w).

[0073] In an optional embodiment, determining the target weight value corresponding to each of the plurality of dimensions comprises: obtaining environmental change data in a preset historical time window, wherein the environmental change data at least includes: regional energy trade data, technology innovation rate, and energy and power policy stability index; determining a weight adjustment coefficient corresponding to each of the plurality of dimensions according to the environmental change data in the preset historical time window, wherein the weight adjustment coefficient of the energy and power trade dimension is proportional to the energy trade data, the weight adjustment coefficient of the energy and power technology dimension is proportional to the technology innovation rate, and the weight adjustment coefficient of the energy and power policy dimension is proportional to the energy and power policy stability index; and correcting a current weight coefficient corresponding to each of the plurality of dimensions based on the weight adjustment coefficient corresponding to each of the plurality of dimensions to obtain a target weight coefficient corresponding to each of the plurality of dimensions.

[0074] Optionally, first, it is necessary to collect environmental change-related information within a preset historical time window. These data at least cover regional energy trade data, technical innovation rate, and energy power policy stability index. Regional energy trade data reflects changes in the economic situation of each entity, technical innovation rate reflects the speed of technological progress, and energy power policy stability index measures policy consistency and stability. Then, according to the collected historical environmental change data, the weight adjustment coefficient of each dimension is determined. These coefficients are used to adjust the current dimension weight to reflect the influence of environmental changes on the energy power pattern prediction. Specifically: the weight adjustment coefficient of the energy power trade dimension is proportional to the regional energy trade data. This means that when the economic fluctuation is large, the weight of energy power trade will be increased, because economic fluctuation directly affects trade activities. The weight adjustment coefficient of the energy power technology dimension is proportional to the technical innovation rate. The acceleration of technical innovation rate indicates that the role of technology in the energy power pattern is enhanced, so the weight of this dimension will also increase accordingly. The weight adjustment coefficient of the energy power policy dimension is proportional to the energy power policy stability index. The region with high policy stability will have a higher policy dimension weight in the prediction model, because a stable policy environment is conducive to the long-term planning and development of energy power. Based on the above-determined weight adjustment coefficients, the current weight coefficients are corrected to obtain target weight coefficients that are closer to the actual situation. The correction process can be realized through mathematical operations, such as multiplication, that is, multiplying the current weight coefficient with the corresponding weight adjustment coefficient to update the weight coefficient, so that it reflects the impact of the latest environmental changes. This dynamic adjustment of weight coefficients ensures that the classifier can adapt to changes in different environments, improving the accuracy and practicality of the prediction.

[0075] In the above manner, by introducing environmental change data to dynamically adjust the weights of different dimensions, the flexibility and adaptability of the prediction model are embodied. Through comparative analysis with historical environmental data, the actual importance of different dimensions in a specific period can be more accurately captured, thereby guiding and supporting the decision-making of the support vector machine classifier, so that the final energy power pattern prediction result is closer to reality, with stronger prediction power and decision-making reference value.

[0076] Optionally, in the process of self-adaptive adjustment of the weight coefficients corresponding to each dimension based on environmental change data, a weight constraint mechanism can be set to ensure that the weight value adjustment of any single dimension does not cause unreasonable deviation of the total weight distribution, maintaining the relative balance and reasonableness of the weight values.

[0077] Through the steps S102 to S110, the correlation between entities (such as cities, countries, etc.) in the region in the energy and power trade, technology and strategy can be quantitatively analyzed in multiple dimensions, and the corresponding energy and power relationship network can be constructed, the importance of each dimension can be evaluated, and the energy and power pattern between entities in a specific future period can be finally analyzed and predicted, so that the technical effects of accurate prediction and evaluation of the evolution trend of the energy and power pattern are achieved, and the technical problems that the factors are not comprehensive when the energy and power relationship network is analyzed in the related art, and the energy and power pattern between entities is not accurately evaluated.

[0078] Based on the above embodiments and optional embodiments, the present application proposes an optional implementation of an energy and power pattern prediction method, Figure 3 which is an optional energy and power pattern prediction method according to an embodiment of the present application, as shown in Figure 3 , which can be understood as an energy and power pattern prediction method based on a relationship network, and is suitable for evaluating the overall energy and power development trend of a region. This embodiment mainly consists of energy and power data preprocessing between entities, relationship network generation, feature matrix construction, classifier prediction, etc. The specific implementation process of the energy and power pattern prediction proposed in this embodiment is as follows:

[0079] S1, obtaining and preprocessing the energy and power relationship data between N entities (such as countries) in the entity. The factors affecting the energy and power pattern are a complex and interrelated system, and this embodiment analyzes and describes the energy and power relationship between different entities from the dimensions of energy and power trade, technology and strategy.

[0080] ①The energy and power trade relationship between entity v i and entity v j reflects the flow of resources and electricity between entities, and the trade relationship (i.e., the correlation degree of the energy and power trade dimension) in a period of time is obtained by the following method:

[0081] Cw ij =∑Rs ij +El ij

[0082] wherein Rs ij is the normalized trade volume of a single energy product (such as coal, oil, gas), and El ij is the normalized transmission capacity.

[0083] ②The energy and power technology relationship between entity v i and entity v j reflects the dependence relationship and cooperation dynamics of entities in the energy and power industry chain, and the technology relationship (i.e., the correlation degree of the energy and power technology dimension) in a period of time is obtained by the following method:

[0084] Twij =∑Po ij +∑Eq ij +∑Ob ij

[0085] where Po ij represents normalized cross-border literature (e.g. patents) citation number in specific field (e.g. photovoltaic technology, grid technology, digitization technology), Eq ij represents normalized trade value of single key equipment (e.g. photovoltaic module, wind power equipment, gas turbine), Ob ij represents normalized funding of single R&D cooperation project.

[0086] ③Energy strategy relationship between entity v i and entity v j reflects the influence and consistency of energy strategy, and the strategy relationship in a period of time (i.e. the correlation degree of energy power strategy dimension) is obtained by the following way:

[0087] Pw ij =∑Og ij +∑Sm ij

[0088] where Og ij is the common bidding record in a project, Sm ij is the degree of compliance of a rule, which is represented by the text similarity of strategy files between entities.

[0089] S2, build the energy power relationship network between entities in the entity. The node v i in the energy power relationship network is an entity, and the edge e ij represents the connection relationship between entities. According to the attributes of the edge, the energy power relationship network is divided into trade, technology and strategy three layers, and three types of undirected relationship network (i.e. energy power relationship network) G c (V,E), G t (V,E), G p (V,E) are constructed, where G c (V,E) represents the trade network, G t (V,E) represents the technology network, and G p (V,E) represents the strategy network.

[0090] S3, convert the energy power relationship network into a weight matrix, and solve the eigenvalue and eigenvector. The weight reflects the importance of the corresponding edge in the whole relationship network. Based on the energy power relationship data in step S1, the weight is valued as follows:

[0091] If there is an edge connection between entity nodes v i and v j and there is energy trade, then eij =Cw ij If entity node v i and v j If there is no edge connection, then e ij =0, thus obtaining the N×N trade weight matrix CW for time period t. t Similarly, the technology weight matrix TW can be obtained. t Strategy weight matrix PW t .

[0092] Solve for the largest eigenvalue λ for each of the above matrices. max The corresponding feature vectors reveal the importance of each node (i.e., entity) in different networks, and thus provide the input samples for the classifier.

[0093] S4, Support Vector Machine (SVM) classification prediction. The SVM performs binary classification on the data using supervised learning, solving for the maximum margin hyperplane on the training samples. Based on the hierarchical network in step S3, and inputting the samples from step S4 at time tk (i.e., the historical time period), a three-class classification model (i.e., the SVM classifier) ​​for energy and electricity trade, technology, and strategy is trained. For the feature samples at time t+k (i.e., the prediction time period), the SVM classification decision function fc(w) is calculated. ij ), ft(w ij ), fp(w ij The system further obtains the classification results for each dimension during the prediction period, and based on these classification results, determines the final energy and electricity pattern prediction results through a weighted voting mechanism in the following form: f t+k =0.4fc(w)+0.3ft(w)+0.3fp(w).

[0094] It should be noted that, in the related art, the energy and power relationship network analysis based on the geopolitical strategy and oil and gas trade considers less technical and strategic factors, and the embodiment comprehensively considers the energy and power trade, technology and strategy, and measures the overall energy and power relationship network, and belongs to the field of energy and power pattern analysis. Specifically, in the cross-entity energy and power relationship network analysis, the trade flow is usually used to depict the geopolitical economic relationship, but the energy and power pattern evolution is affected by many factors such as geographical environment, technical equipment, economic trade and development strategy, and the embodiment comprehensively considers various factors to accurately and comprehensively reflect the dynamic change trend of the energy and power pattern evolution. In addition, the energy and power pattern analysis based on the cross-entity strategy relationship focuses on the analysis of the evolution of the influence of the node entity in the energy and power development and the analysis of the network characteristics, and lacks the judgment of the future development trend, and the embodiment measures the overall energy and power pattern evolution trend, which is of great significance for energy and power strategy formulation and guiding industrial investment. In terms of specific implementation means, the embodiment can more objectively and comprehensively describe the mutual dependence and mutual influence relationship between entities according to the method for obtaining data based on the main influencing factors of the energy and power relationship. At the same time, the energy and power relationship hierarchical network construction method is adopted to decompose the complex energy and power relationship network system, reduce the difficulty of network matrix construction, and improve the accuracy of energy and power relationship analysis and evaluation.

[0095] In the embodiment, an energy and power pattern prediction device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be described again. As used below, the term "module" "device" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware or a combination of software and hardware is also possible and contemplated.

[0096] According to the embodiment of the present application, a device embodiment for implementing the above-mentioned energy and power pattern prediction method is also provided, Figure 4 is a structural schematic diagram of an energy and power pattern prediction device according to the embodiment of the present application, as Figure 4 shown, the above-mentioned energy and power pattern prediction device comprises: an energy and power data acquisition module 200, a relationship data acquisition module 202, a network construction module 204, an importance evaluation module 206, and a prediction module 208, wherein:

[0097] The energy and power data acquisition module 200 is configured to acquire current energy and power data of a plurality of entities in a region, wherein the current energy and power data comprises energy and power trade data, energy and power technology data and energy and power strategy data of the corresponding entity in the current period.

[0098] The relationship data obtaining module 202 is connected to the energy power data obtaining module 200, and is configured to determine energy power relationship data corresponding to each of a plurality of dimensions based on current energy power data corresponding to each of a plurality of entities, wherein the energy power relationship data comprises a correlation degree of each of two entities in the plurality of entities in the corresponding dimension, and the plurality of dimensions comprises a correlation degree of an energy power trade dimension, an energy power technology dimension, and an energy power policy dimension;

[0099] The network constructing module 204 is connected to the relationship data obtaining module 202, and is configured to construct a plurality of energy power relationship networks based on the energy power relationship data corresponding to each of the plurality of dimensions, wherein the plurality of energy power relationship networks correspond to the plurality of dimensions one by one.

[0100] The importance evaluating module 206 is connected to the network constructing module 204, and is configured to determine an importance evaluation result corresponding to each of the plurality of dimensions according to the plurality of energy power relationship networks, wherein the importance evaluation result comprises an importance corresponding to each of the plurality of entities in the corresponding dimension.

[0101] The prediction module 208 is connected to the importance evaluating module 206, and is configured to determine an energy power pattern prediction result of a region in a prediction period based on the importance evaluation result corresponding to each of the plurality of dimensions, wherein the energy power pattern prediction result is used to indicate an energy power development trend of the region as a whole.

[0102] Optionally, the network constructing module is further configured to construct the plurality of energy power relationship networks based on the energy power relationship data corresponding to each of the plurality of dimensions, with the plurality of entities as nodes and the correlation degree of each of two entities in the corresponding dimension as an edge.

[0103] Optionally, the importance evaluating module comprises: a conversion submodule configured to convert any of the plurality of energy power relationship networks into a weight matrix; a solving submodule configured to solve a plurality of eigenvalues of the weight matrix and a feature vector corresponding to each of the plurality of eigenvalues; an evaluation submodule configured to obtain the importance evaluation result of any dimension based on the feature vector corresponding to a maximum eigenvalue in the plurality of eigenvalues; and a first repeated obtaining submodule configured to obtain the importance evaluation result corresponding to each of the plurality of dimensions in the same manner as the importance evaluation result of any dimension.

[0104] Optionally, the prediction module comprises: a classification submodule, configured to obtain a classification result corresponding to each of the plurality of dimensions based on the importance evaluation result corresponding to each of the plurality of dimensions by using a plurality of support vector machine classifiers, wherein the classification result comprises a classification category of the plurality of entities in the corresponding dimension, and the plurality of dimensions and the plurality of support vector machine classifiers correspond to each other in one-to-one manner; a weight determination submodule, configured to determine a target weight value corresponding to each of the plurality of dimensions; and a prediction submodule, configured to obtain an energy and power pattern prediction result based on the classification result corresponding to each of the plurality of dimensions and the target weight value corresponding to each of the plurality of dimensions.

[0105] Optionally, the weight determination submodule comprises: a data acquisition unit, configured to acquire environmental change data in a preset historical time window, wherein the environmental change data at least comprises energy trade data of a region, a technology innovation rate, and an energy and power policy stability index; a coefficient adjustment unit, configured to determine a weight adjustment coefficient corresponding to each of the plurality of dimensions according to the environmental change data in the preset historical time window, wherein the weight adjustment coefficient of the energy and power trade dimension is proportional to the energy trade data of the region, the weight adjustment coefficient of the energy and power technology dimension is proportional to the technology innovation rate, and the weight adjustment coefficient of the energy and power policy dimension is proportional to the energy and power policy stability index; and a weight correction unit, configured to correct a current weight coefficient corresponding to each of the plurality of dimensions based on the weight adjustment coefficient corresponding to each of the plurality of dimensions to obtain a target weight coefficient corresponding to each of the plurality of dimensions.

[0106] Optionally, the relationship data acquisition module comprises: a first normalization submodule, configured to perform normalization processing on the resource trade volume and the power transmission capacity between any two of the plurality of entities respectively to obtain normalized resource trade volume and normalized power transmission capacity; a first weighting submodule, configured to perform weighted summation operation on the normalized resource trade volume and the normalized power transmission capacity to obtain a correlation degree of any two entities in the energy and power trade dimension; and a second repeated acquisition submodule, configured to obtain the correlation degree of each two entities in the energy and power trade dimension in the same manner as that of obtaining the correlation degree of any two entities in the energy and power trade dimension.

[0107] Optionally, the relational data acquisition module includes: a second normalization submodule, used to normalize the cross-entity literature citations, key equipment trade volume, and R&D cooperation project funds between any two entities when the energy and power technology data includes cross-entity literature citations, key equipment trade volume, and R&D cooperation project funds related to energy and power, to obtain normalized cross-entity literature citations, normalized key equipment trade volume, and normalized R&D cooperation project funds; a second weighting submodule, used to perform a weighted summation operation on the normalized cross-entity literature citations, normalized key equipment trade volume, and normalized R&D cooperation project funds to obtain the correlation degree between any two entities in the energy and power technology dimension; and a third repeated acquisition submodule, used to obtain the correlation degree between every two entities in the energy and power technology dimension by adopting the same method as obtaining the correlation degree between any two entities in the energy and power technology dimension.

[0108] Optionally, the relationship data acquisition module includes: a third normalization submodule, used to normalize the joint bidding records and strategy fit between any two entities among multiple entities, given that the energy and power strategy data includes joint bidding records and strategy fit, to obtain normalized joint bidding records and normalized strategy fit, wherein the strategy fit is obtained based on the text similarity of the energy and power strategies of any two entities; a third weighting submodule, used to perform a weighted summation operation on the normalized joint bidding records and normalized strategy fit to obtain the correlation between any two entities in the energy and power strategy dimension; and a fourth repeated acquisition submodule, used to obtain the correlation between each pair of entities in the energy and power strategy dimension by adopting the same method as obtaining the correlation between any two entities in the energy and power strategy dimension.

[0109] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0110] It should be noted that the energy and power data acquisition module 200, relationship data acquisition module 202, network construction module 204, importance assessment module 206, and prediction module 208 mentioned above correspond to steps S102 to S110 in the embodiments. The instances and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run on a computer terminal.

[0111] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the embodiments, and will not be repeated here.

[0112] The aforementioned energy and power pattern prediction device may also include a processor and a memory. The aforementioned energy and power data acquisition module 200, relationship data acquisition module 202, network construction module 204, importance assessment module 206, prediction module 208, etc., are all stored in the memory as program modules, and the processor executes the aforementioned program modules stored in the memory to realize the corresponding functions.

[0113] The processor contains a core that retrieves the corresponding program modules from memory. One or more cores may be configured. Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory includes at least one memory chip.

[0114] According to an embodiment of this application, an embodiment of a non-volatile storage medium is also provided. Optionally, in this embodiment, the non-volatile storage medium includes a stored program, wherein, when the program is running, it controls the device containing the non-volatile storage medium to execute any of the aforementioned energy and power pattern prediction methods.

[0115] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals, and the non-volatile storage medium includes stored programs.

[0116] Optionally, a program may be used to control the device containing the non-volatile storage medium to execute any of the above-mentioned energy and power pattern prediction method steps during program execution.

[0117] According to an embodiment of this application, an embodiment of a processor is also provided. Optionally, in this embodiment, the processor is used to run a program, wherein the program executes any of the above-described energy and power pattern prediction methods.

[0118] According to an embodiment of this application, an embodiment of a computer program product is also provided, which, when executed on a data processing device, is adapted to execute a program that initializes the energy and power pattern prediction method steps described above.

[0119] This invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of any of the above-described energy and power pattern prediction methods.

[0120] The order of the above embodiments of the present invention is merely for description and does not represent the superiority or inferiority of the embodiments.

[0121] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0122] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of modules described above can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between modules, and may be electrical or other forms.

[0123] The modules described above as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0124] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0125] If the aforementioned integrated modules are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable non-volatile storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a non-volatile storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned non-volatile storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0126] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for predicting energy and electricity patterns, characterized in that, include: Obtain the current energy and electricity data corresponding to multiple entities within the region, wherein the current energy and electricity data includes the energy and electricity trade data, energy and electricity technology data, and energy and electricity strategy data of the corresponding entities in the current time period; Based on the current energy and electricity data corresponding to each of the multiple entities, energy and electricity relationship data corresponding to each of the multiple dimensions are determined. The energy and electricity relationship data includes the correlation between each pair of entities in the corresponding dimension. The multiple dimensions include the correlation between the energy and electricity trade dimension, the energy and electricity technology dimension, and the energy and electricity strategy dimension. Based on the energy and electricity relationship data corresponding to each of the multiple dimensions, multiple energy and electricity relationship networks are constructed, wherein each of the multiple energy and electricity relationship networks corresponds one-to-one with the multiple dimensions; Based on the multiple energy and power relationship networks, the importance assessment results corresponding to each of the multiple dimensions are determined, wherein the importance assessment results include the importance of each of the multiple entities under the corresponding dimension; Based on the importance assessment results corresponding to the various dimensions, the predicted energy and power pattern of the region during the forecast period is determined, wherein the predicted energy and power pattern is used to indicate the overall energy and power development trend of the region.

2. The method according to claim 1, characterized in that, The construction of multiple energy and electricity relationship networks based on the energy and electricity relationship data corresponding to each of the multiple dimensions includes: Based on the energy and electricity relationship data corresponding to each of the multiple dimensions, the multiple entities are used as nodes, and the correlation between each pair of entities in the corresponding dimension is used as an edge to construct the multiple energy and electricity relationship network.

3. The method according to claim 1, characterized in that, The step of determining the importance assessment results for each of the multiple dimensions based on the multiple energy and power relationship networks includes: Convert any one of the multiple energy and power relationship networks into a weight matrix; Solve for multiple eigenvalues ​​of the weight matrix, and the eigenvectors corresponding to each of the multiple eigenvalues; Based on the feature vector corresponding to the largest feature value among the multiple feature values, the importance evaluation result of any dimension is obtained; The importance assessment results for each of the multiple dimensions are obtained by using the method of obtaining the importance assessment results for any one of the dimensions.

4. The method according to claim 1, characterized in that, The step of determining the energy and power pattern forecast results for the region during the forecast period based on the importance assessment results corresponding to the multiple dimensions includes: Based on the importance evaluation results corresponding to each of the multiple dimensions, multiple support vector machine classifiers are used to obtain the classification results corresponding to each of the multiple dimensions. The classification results include the classification categories of the multiple entities under the corresponding dimensions, and the multiple dimensions and the multiple support vector machine classifiers correspond one-to-one. Determine the target weight values ​​corresponding to each of the multiple dimensions; Based on the classification results corresponding to each of the multiple dimensions, and the target weight values ​​corresponding to each of the multiple dimensions, the predicted results of the energy and power pattern are obtained.

5. The method according to claim 4, characterized in that, Determining the target weight values ​​corresponding to each of the multiple dimensions includes: Obtain environmental change data within a preset historical time window, wherein the environmental change data includes at least: energy trade data, technological innovation rate, and energy and power strategy stability index of the region; Based on the environmental change data within the preset historical time window, the weight adjustment coefficients corresponding to each of the multiple dimensions are determined. Among them, the weight adjustment coefficient of the energy and electricity trade dimension is proportional to the energy trade data, the weight adjustment coefficient of the energy and electricity technology dimension is proportional to the technology innovation rate, and the weight adjustment coefficient of the energy and electricity strategy dimension is proportional to the energy and electricity strategy stability index. Based on the weight adjustment coefficients corresponding to each of the multiple dimensions, the current weight coefficients corresponding to each of the multiple dimensions are corrected to obtain the target weight coefficients corresponding to each of the multiple dimensions.

6. The method according to any one of claims 1 to 5, characterized in that, The determination of energy and electricity relationship data corresponding to multiple dimensions based on the current energy and electricity data corresponding to each of the multiple entities includes: Given that the energy and electricity trade data includes energy and electricity resource trade volume and transmission capacity, the energy and electricity relationship data corresponding to the energy and electricity trade dimension is obtained as follows: The resource trade volume and power transmission capacity between any two entities among the plurality of entities are normalized to obtain normalized resource trade volume and normalized power transmission capacity. The normalized resource trade volume and normalized transmission capacity are weighted and summed to obtain the correlation between any two entities in the energy and electricity trade dimension. The correlation between any two entities in the energy and electricity trade dimension is obtained by using the method of obtaining the correlation between any two entities in the energy and electricity trade dimension.

7. The method according to any one of claims 1 to 5, characterized in that, The determination of energy and electricity relationship data corresponding to multiple dimensions based on the current energy and electricity data corresponding to each of the multiple entities includes: The energy and power relationship data corresponding to the energy and power technology dimension is obtained in the following manner: When the energy and power technology data includes cross-entity literature citations, key equipment trade volume, and R&D cooperation project funding related to energy and power, the cross-entity literature citations, key equipment trade volume, and R&D cooperation project funding between any two entities are normalized to obtain normalized cross-entity literature citations, normalized key equipment trade volume, and normalized R&D cooperation project funding. The normalized cross-entity literature citation count, normalized key equipment trade volume, and normalized R&D cooperation project funding are weighted and summed to obtain the correlation degree between any two entities in the energy and power technology dimension. The correlation between any two entities in the energy and power technology dimension is obtained by using the method of obtaining the correlation between any two entities in the energy and power technology dimension.

8. The method according to any one of claims 1 to 5, characterized in that, The determination of energy and electricity relationship data corresponding to multiple dimensions based on the current energy and electricity data corresponding to each of the multiple entities includes: The energy and power relationship data corresponding to the energy and power strategy dimension is obtained in the following manner: When the energy and power strategy data includes joint bidding records and strategy fit, the joint bidding records and strategy fit between any two entities among the multiple entities are normalized to obtain normalized joint bidding records and normalized strategy fit, wherein the strategy fit is obtained based on the text similarity of the energy and power strategies of the two entities. The normalized joint bidding records and the normalized strategy fit are weighted and summed to obtain the correlation between any two entities in the energy and power strategy dimension. The correlation between any two entities in the energy and power strategy dimension is obtained by using the method of obtaining the correlation between any two entities in the energy and power strategy dimension.

9. An energy and power pattern prediction device, characterized in that, include: The energy and electricity data acquisition module is used to acquire the current energy and electricity data corresponding to multiple entities within the region. The current energy and electricity data includes the energy and electricity trade data, energy and electricity technology data, and energy and electricity strategy data of the corresponding entities in the current time period. The relationship data acquisition module is used to determine the energy and electricity relationship data corresponding to multiple dimensions based on the current energy and electricity data corresponding to each of the multiple entities. The energy and electricity relationship data includes the correlation degree between each pair of entities in the corresponding dimension. The multiple dimensions include the correlation degree of energy and electricity trade dimension, energy and electricity technology dimension and energy and electricity strategy dimension. The network construction module is used to construct multiple energy and electricity relationship networks based on the energy and electricity relationship data corresponding to the multiple dimensions, wherein the multiple energy and electricity relationship networks correspond one-to-one with the multiple dimensions; The importance assessment module is used to determine the importance assessment results corresponding to each of the multiple dimensions based on the multiple energy and power relationship networks, wherein the importance assessment results include the importance of each of the multiple entities under the corresponding dimension; The prediction module is used to determine the predicted energy and power pattern of the region during the prediction period based on the importance assessment results of the multiple dimensions, wherein the predicted energy and power pattern is used to indicate the overall energy and power development trend of the region.

10. The apparatus according to claim 9, characterized in that, The network building module is also used for: Based on the energy and electricity relationship data corresponding to each of the multiple dimensions, the multiple entities are used as nodes, and the correlation between each pair of entities in the corresponding dimension is used as an edge to construct the multiple energy and electricity relationship network.

11. The apparatus according to claim 9, characterized in that, The importance assessment module includes: The transformation submodule is used to convert any one of the multiple energy and power relationship networks into a weight matrix; The solution submodule is used to solve for multiple eigenvalues ​​of the weight matrix and the eigenvectors corresponding to each of the multiple eigenvalues; The evaluation submodule is used to obtain the importance evaluation result of any dimension based on the feature vector corresponding to the largest feature value among the multiple feature values; The first repeated acquisition submodule is used to obtain the importance assessment results corresponding to each of the multiple dimensions by adopting the method of obtaining the importance assessment result of any one of the dimensions.

12. The apparatus according to claim 9, characterized in that, The prediction module includes: The classification submodule is used to obtain the classification results corresponding to each of the multiple dimensions by using multiple support vector machine classifiers based on the importance evaluation results corresponding to each of the multiple dimensions. The classification results include the classification categories of the multiple entities under the corresponding dimensions, and the multiple dimensions and the multiple support vector machine classifiers correspond one-to-one. The weight determination submodule is used to determine the target weight values ​​corresponding to each of the multiple dimensions; The prediction submodule is used to obtain the energy and power pattern prediction result based on the classification results corresponding to each of the multiple dimensions and the target weight values ​​corresponding to each of the multiple dimensions.

13. The apparatus according to claim 12, characterized in that, The weight determination submodule includes: The data acquisition unit is used to acquire environmental change data within a preset historical time window, wherein the environmental change data includes at least: energy trade data, technological innovation rate, and energy and power strategy stability index of the region. The coefficient adjustment unit is used to determine the weight adjustment coefficients corresponding to each of the multiple dimensions based on the environmental change data within the preset historical time window. The weight adjustment coefficient of the energy and electricity trade dimension is proportional to the energy trade data, the weight adjustment coefficient of the energy and electricity technology dimension is proportional to the technology innovation rate, and the weight adjustment coefficient of the energy and electricity strategy dimension is proportional to the energy and electricity strategy stability index. The weight correction unit is used to correct the current weight coefficients of the multiple dimensions based on the weight adjustment coefficients corresponding to each of the multiple dimensions, so as to obtain the target weight coefficients of the multiple dimensions.

14. The apparatus according to any one of claims 9 to 13, characterized in that, The relational data acquisition module includes: The first normalization submodule is used to normalize the resource trade volume and power transmission capacity between any two entities among the plurality of entities, respectively, to obtain normalized resource trade volume and normalized power transmission capacity. The first weighted submodule is used to perform a weighted summation operation on the normalized resource trade volume and the normalized transmission capacity to obtain the correlation degree between any two entities in the energy and electricity trade dimension. The second repeatable acquisition submodule is used to obtain the correlation degree of each pair of entities in the energy and electricity trade dimension by obtaining the correlation degree of any two entities in the energy and electricity trade dimension.

15. The apparatus according to any one of claims 9 to 13, characterized in that, The relational data acquisition module includes: The second normalization submodule is used to normalize the cross-entity literature citations, key equipment trade volume, and R&D cooperation project funds between any two entities in the energy and power technology data, when the energy and power technology data includes cross-entity literature citations, key equipment trade volume, and R&D cooperation project funds related to energy and power, respectively, to obtain normalized cross-entity literature citations, normalized key equipment trade volume, and normalized R&D cooperation project funds. The second weighted submodule is used to perform a weighted summation operation on the normalized cross-entity document citation count, normalized key equipment trade volume, and normalized R&D cooperation project funds to obtain the correlation degree between any two entities in the energy and power technology dimension. The third repeated acquisition submodule is used to obtain the correlation degree of each pair of entities in the energy and power technology dimension by obtaining the correlation degree of any two entities in the energy and power technology dimension.

16. The apparatus according to any one of claims 9 to 13, characterized in that, The relational data acquisition module includes: The third normalization submodule is used to normalize the joint bidding records and strategy fit between any two entities among the multiple entities when the energy and power strategy data includes joint bidding records and strategy fit, so as to obtain normalized joint bidding records and normalized strategy fit, wherein the strategy fit is obtained based on the text similarity of the energy and power strategies of the two entities. The third weighted submodule is used to perform a weighted summation operation on the normalized joint bidding records and the normalized strategy fit to obtain the correlation between any two entities in the energy and power strategy dimension. The fourth repeat acquisition submodule is used to obtain the correlation degree of each pair of entities in the energy and power strategy dimension by obtaining the correlation degree of any two entities in the energy and power strategy dimension.

17. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the energy and power pattern prediction method according to any one of claims 1 to 8.

Images