Processing method and system based on power transaction decision data

Abstract

The present invention relates to the field of data processing, and in particular to a processing method and system based on power trading decision data. The method comprises the following steps: acquiring real-time power trading data; performing time series analysis on real-time power trading data to generate power trading time series data; performing trading decision analysis on power trading time series data to generate trading time series decision data; performing multi-dimensional non-uniform pulse decomposition on trading time series decision data using spline decomposition technology to generate multi-dimensional decision control point data; performing spatial position analysis on multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; performing multi-dimensional coordinate mapping on multi-dimensional decision control point data based on multi-dimensional control point spatial coordinate data to construct a multi-dimensional decision space; performing control point time series change analysis on the multi-dimensional decision space to generate control point time series change data. The present invention realizes efficient and accurate power data processing.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method and system for processing power transaction decision data. Background Art

[0002] With the development of the power market and the increasing complexity of power trading, the processing of power trading decision data has become crucial. In power trading, a large amount of data is involved, including market prices, supply and demand, power load forecasts, power equipment status, etc. These data are high-dimensional, diverse and real-time, and need to be effectively analyzed and processed to support decision-making and risk management. Traditional power trading decision-making methods often rely on experience and manual intervention, and there are problems of low decision-making efficiency and poor accuracy. Therefore, in order to meet the needs of modern power trading, an intelligent processing method based on power trading decision data is needed. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes a processing method and system based on power trading decision data to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned object, the present invention provides a processing method based on power transaction decision data, comprising the following steps: Step S1: acquiring real-time power transaction data; performing time series analysis on the real-time power transaction data to generate power transaction time series data; performing transaction decision analysis on the power transaction time series data to generate transaction time series decision data; performing multi-dimensional non-uniform pulse decomposition on the transaction time series decision data using spline decomposition technology to generate multi-dimensional decision control point data;

[0005] Step S2: Perform spatial position analysis on the multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; perform multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point spatial coordinate data to construct a multi-dimensional decision space; perform control point temporal change analysis on the multi-dimensional decision space to generate control point temporal change data;

[0006] Step S3: performing evolution trajectory analysis on the control point time series change data to generate control point evolution trajectory data; performing spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law; performing evolution trend prediction on the multidimensional decision space through the decision control point evolution law to generate multidimensional decision trend prediction data;

[0007] Step S4: Optimize the decision structure of the multidimensional decision space using the multidimensional decision trend prediction data to generate a multidimensional decision optimization space; perform control point association mining on the multidimensional decision control point data to generate control point association data; perform influence domain coverage calculation on the multidimensional decision space using the control point association data to generate decision influence range data;

[0008] Step S5: performing decision transfer path analysis on the multi-dimensional decision control point data according to the decision impact range data to generate decision transfer path data; performing associated network topology analysis on the decision transfer path data to generate external topology structure data; performing dynamic decision network construction on the external topology structure data to generate an external decision topology network;

[0009] Step S6: Perform decision path space analysis on the multidimensional decision optimization space to generate internal decision path space data; perform decision path network analysis on the external decision topology network to generate external decision path network data; construct a knowledge graph for the internal decision path space data and the external decision path network data, and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

[0010] The present invention can obtain accurate power trading time series data by performing time series analysis on real-time power trading data. Such analysis can reveal the trading rules, trends and periodic changes in the power market, and provide basic data for subsequent trading decisions. Trading decision analysis using power trading time series data can generate trading time series decision data, which can be used to evaluate the effectiveness of trading strategies, risk management and profit prediction. By performing spatial position analysis and multidimensional coordinate mapping on multidimensional decision control point data, a multidimensional decision space can be constructed. Such space construction can convert decision problems into multidimensional coordinate spaces, which is convenient for decision analysis and optimization. Time series change analysis on multidimensional decision control point data can obtain control point Time series change data, these data help to understand the evolution trend and key moments of decision control points, and provide a time series basis for decision-making. By performing evolution trajectory analysis and spatial morphological structure drift analysis on the time series change data of control points, the evolution law of decision control points can be generated, which helps to understand the change trend and regularity of decision control points and provide a basis for decision prediction and optimization. Based on the evolution law of decision control points, the evolution trend of multidimensional decision space is predicted, and multidimensional decision trend prediction data can be generated. These data can be used to predict future decision changes and trends and guide decision making. The decision structure of multidimensional decision space is optimized using multidimensional decision trend prediction data, and a multidimensional decision optimization space can be generated. This space can provide more optimized decisions. By mining the association of multi-dimensional decision control points and calculating the coverage of the influence domain, decision influence range data can be generated. These data can reveal the mutual influence and transmission relationship between decisions, help identify important decision factors and potential risks, and conduct path analysis on the decision influence range data to reveal the transmission path between decision control points and the way of decision transmission, helping decision makers understand the transmission mechanism and path selection of decisions. By conducting association network topology analysis on decision transmission path data, external topology structure data can be constructed to show the association relationship and network structure between decision control points, providing decision makers with a global perspective and overall understanding. By conducting association network topology analysis, The external decision topology network organizes and displays the decision control points in the form of a network, providing decision makers with intuitive decision topology structures and related information. The decision path space analysis is performed on the multidimensional decision optimization space, which can reveal different decision paths in the space. This helps to understand the characteristics, advantages and disadvantages of different decision paths, and provide decision makers with diversity and selectivity of decision plans. The external decision path network analysis is based on the decision topology network. By analyzing the network structure and node relationships, the location and connection method of different decision paths in the overall network are identified. The knowledge graph is a structured knowledge representation form. By integrating and modeling the internal decision path space data and the external decision path network data, a transaction decision knowledge graph is constructed.The construction of knowledge graph helps to express and store the information and relevance of decision paths, forming a knowledge graph database, which provides a visual and queryable way for decision makers to more easily obtain and use knowledge and information related to transaction decisions.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: obtaining real-time power transaction data;

[0013] Step S12: performing time series analysis on the real-time power transaction data to generate power transaction time series data;

[0014] Step S13: identifying decision factors for the power transaction time series data and generating decision factor data;

[0015] Step S14: performing transaction decision analysis on the power transaction time series data according to the decision factor data to generate transaction time series decision data;

[0016] Step S15: Use spline decomposition technology to perform multi-dimensional non-uniform pulse decomposition on the transaction timing decision data to generate multi-dimensional decision control point data.

[0017] The present invention provides decision makers with the latest market intelligence and transaction information by acquiring real-time power transaction data, so that they can make timely decision responses. Real-time data reflects the current market conditions and transaction trends, which helps decision makers seize the best transaction opportunities. By performing time series analysis on real-time power transaction data, the data can be converted into a time series form, so that the trend and change law of the data are more clearly visible. This helps decision makers to intuitively understand the dynamic changes of transaction data and discover the patterns and trends therein. By analyzing the power transaction time series data, the trends and periodic changes in the data can be identified. Decision makers can use this trend information to make predictions in order to better grasp the market trend and make corresponding transaction decisions. By analyzing the power transaction time series data, factors that have an important impact on transaction decisions can be identified and extracted. These factors may include supply and demand relationships, price fluctuations, seasonal changes, etc., and decision makers can use these factor data as a basis for decision making. By performing transaction decision analysis on the power transaction time series data, decision makers can formulate more optimized transaction strategies based on the decision factor data. This helps to maximize market opportunities, reduce risks, and improve transaction benefits. Using spline decomposition technology to decompose transaction timing decision data can analyze and control the data in multiple dimensions. This helps decision makers gain a deeper understanding of all aspects of transaction data and conduct decision analysis from multiple perspectives. Non-uniform pulse decomposition is a method of decomposing time series data into pulse signals of different frequencies and amplitudes. Through this decomposition, decision makers can observe and control the volatility and changes of transaction data in a more granular manner, thereby formulating decision strategies more accurately.

[0018] Preferably, the specific steps of step S15 are:

[0019] Step S151: segmenting the transaction timing decision data into timing windows to generate timing window data;

[0020] Step S152: extracting the transaction timing decision data by using the timing window data to generate a transaction timing decision sequence;

[0021] Step S153: using the spline decomposition technology to perform multi-dimensional non-uniform pulse decomposition on the transaction timing decision sequence to generate a one-dimensional pulse component data set;

[0022] Step S154: reconstructing a multi-dimensional spatial pulse path for the one-dimensional pulse component data set to generate a multi-dimensional spatial pulse path;

[0023] Step S155: performing pulse direction mutation analysis on the multidimensional space pulse path and marking the multidimensional space pulse mutation point;

[0024] Step S156: extract decision control nodes from the transaction timing decision sequence according to the multi-dimensional space pulse mutation points to generate multi-dimensional decision control point data.

[0025] The present invention divides the continuous time series data into multiple windows by performing time series window segmentation on the transaction time series decision data, and each window represents the data in a relatively small time period. This segmentation is helpful for further analysis and processing of the data, making the characteristics of the data more clear and controllable. By performing sequence extraction on the time series window data, the data in each window can be arranged in chronological order to form a transaction time series decision sequence. This serialized data structure is more suitable for subsequent analysis and processing, and is convenient for applying various sequence models and algorithms. The transaction time series decision sequence is pulse decomposed by spline decomposition technology, and the sequence can be decomposed into different pulse components. Each pulse component represents a local fluctuation or change in the sequence, which helps to capture the important features and change patterns in the sequence. By reconstructing the multidimensional space pulse path, the one-dimensional pulse component data set can be converted into a multidimensional space data set. This visualization method helps to intuitively observe and understand the relationship and evolution trend between different pulse components. By performing pulse direction mutation analysis on the multi-dimensional space pulse path, the mutation points in the path can be marked. The mutation points represent the locations where the direction of the path changes significantly, which helps to identify important turning points and key moments in the trading decision data. According to the multi-dimensional space pulse mutation points, the decision control nodes of the trading timing decision sequence can be extracted. These nodes represent important time points and key decision stages in the decision-making process. Extracting these nodes helps to analyze and control the key links of the decision. The generation of multi-dimensional decision control point data enables decision analysis and control in various dimensions. By conducting a comprehensive analysis of decision control points in different dimensions, more comprehensive and accurate decision guidance can be obtained, thereby improving the effectiveness and accuracy of power trading decisions.

[0026] Preferably, the specific steps of step S2 are:

[0027] Step S21: Perform spatial position analysis on the multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data;

[0028] Step S22: performing multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point space coordinate data to construct a multi-dimensional decision space;

[0029] Step S23: performing cluster analysis on the multi-dimensional decision control point data to generate decision control point cluster data;

[0030] Step S24: dividing the multidimensional decision space into cluster regions according to the decision control point cluster data to generate a multidimensional decision space region;

[0031] Step S25: performing control point time series change analysis on the multi-dimensional decision space region to generate control point time series change data.

[0032] Through spatial position analysis, the present invention can assign specific spatial coordinates to each multidimensional decision control point. These coordinates represent the position of each control point in the multidimensional decision space, which helps to understand and visualize the spatial relationship between the control points. Through multidimensional coordinate mapping, the multidimensional decision control point data can be mapped to the multidimensional decision space, thus establishing a multidimensional space based on decision points for representing and analyzing the multidimensional features of decisions. The construction of the multidimensional decision space enables the analysis of decisions in multiple dimensions. By observing and comparing the distribution and change trends of decision points in different dimensions, more comprehensive and accurate decision information can be obtained, and the effect of power trading decisions can be further optimized. Through cluster analysis, The multidimensional decision control point data can be divided into different clusters. Each cluster represents a group of control points with similar decision characteristics. The generation of clusters helps to discover potential patterns and group behaviors in decision data. According to the decision control point clustering data, the multidimensional decision space can be divided into different regions. Each region contains control points with similar decision characteristics. This spatial division helps to subdivide the decision space into different subspaces, so as to better understand and analyze the characteristics and patterns of decisions. By performing time series change analysis on the control points in the multidimensional decision space region, the time evolution characteristics of the decision can be revealed, which helps to understand the changing trends and patterns of decisions in different time periods and provide a reference for predicting future decision changes.

[0033] Preferably, the specific steps of step S3 are:

[0034] Step S31: Identify edge control points in the multidimensional decision space area and mark edge decision control points;

[0035] Step S32: performing layout structure analysis on the multidimensional decision space based on the edge decision control points to generate decision space layout structure data;

[0036] Step S33: performing evolution trajectory analysis on the control point time series change data through the multi-dimensional decision space layout structure data to generate control point evolution trajectory data;

[0037] Step S34: performing spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law;

[0038] Step S35: Predict the evolution trend of the multidimensional decision space according to the evolution law of the decision control points to generate multidimensional decision trend prediction data.

[0039] The present invention can determine which decision points are located at the edge of the decision space by identifying edge control points in the multidimensional decision space area. Edge decision points often have special significance and influence in the decision-making process, so marking these points can help identify key decision points. The marking of edge decision control points helps to analyze the boundary characteristics of the decision space. By observing the distribution and changes of edge decision points, the boundary morphology of the decision space can be understood, and the structure and characteristics of the decision space can be further understood. By performing layout structure analysis based on edge decision control points, the overall layout and structural characteristics of the decision space can be revealed, which helps to understand the relative positions and relationships between different decision points in the decision space, and further explore the overall laws and patterns of decision-making. By performing evolution trajectory analysis on the time series change data of control points through multidimensional decision space layout structure data, the control points can be revealed. The changing trajectory of the points in the decision space helps to understand the evolution process of the decision points over time and identify the development trend and change pattern of the decision. By performing spatial morphological structure drift analysis on the control point evolution trajectory data, the morphological structure change law of the decision points in the decision space can be discovered, which helps to understand the relationship evolution between decision points and the dynamic changes of the decision space, revealing the inherent laws and mechanisms of decision-making. Based on the evolution law of the decision control points, the evolution trend of the multidimensional decision space can be predicted, which can help identify potential changes and trends in the decision space, predict possible decision patterns and development directions in advance, and support the generation of decision trend prediction data to provide support for decision making and decision evaluation. Decision makers can adjust decision strategies and directions based on multidimensional decision trend prediction data to cope with possible decision changes and challenges in the future.

[0040] Preferably, the specific steps of step S34 are:

[0041] Step S341: performing control point distribution entropy quantization processing on the multi-dimensional decision control point data to generate a spatial control point distribution entropy index;

[0042] Step S342: performing spatial morphological change amplitude analysis on the control point evolution trajectory data according to the spatial control point distribution entropy index to generate morphological change amplitude data;

[0043] Step S343: performing spatial morphological structure drift analysis on the multidimensional decision space according to the morphological change amplitude data to generate spatial morphological structure drift data;

[0044] Step S344: Calculating the morphological drift rate of the spatial morphological structure drift data based on the time series window data to generate morphological drift rate data;

[0045] Step S345: performing control point evolution analysis on the control point evolution trajectory data according to the morphological drift rate data to generate a decision control point evolution law.

[0046] The present invention can measure the uniformity of control point distribution in the decision space by performing entropy quantization processing on multi-dimensional decision control point data. The spatial control point distribution entropy index reflects the dispersion and concentration of control points in the decision space, and provides a quantitative indicator to evaluate the balance and diversity of the decision space. According to the spatial control point distribution entropy index, the control point evolution trajectory data is analyzed to measure the morphological change amplitude of the decision point in the decision space. The morphological change amplitude data reflects the degree of change of the decision point in space, including changes in position, density, etc., and provides basic data for subsequent morphological drift analysis. According to the morphological change amplitude data, the morphological structure drift analysis of the multi-dimensional decision space is performed. By analyzing the spatial characteristics and patterns in the morphological change amplitude data, the morphological structure drift of the decision space, that is, the change in the relationship between the decision points, can be revealed. This helps to understand the dynamic evolution process of the decision space and the changing laws of the spatial structure. By calculating the spatial morphological structure drift data based on the time series window data, the morphological drift rate data can be obtained. The morphological drift rate refers to the rate and degree of morphological structure drift in the decision space, reflecting the dynamic change intensity of the decision space. The morphological drift rate data provides a measure of the evolution speed of the decision space, which can help discover the rapid changes and potential instability of the decision space. According to the morphological drift rate data, the control point evolution trajectory data is analyzed to reveal the evolution law of the decision control point. By analyzing the relationship between the trajectory changes of the decision control point in the decision space and the morphological drift rate, the correlation between the decision control points, the diversity of the decision points, and the evolution trend can be found. This helps to understand the evolution law of the decision control points and provide reference and guidance for power trading decisions.

[0047] Preferably, the specific steps of step S4 are:

[0048] Step S41: performing optimal solution decision analysis on the multi-dimensional decision trend prediction data to generate configuration optimal solution data;

[0049] Step S42: optimizing the decision structure of the multidimensional decision space by using the configuration optimal solution data to generate a multidimensional decision optimization space;

[0050] Step S43: performing control point association mining on the multi-dimensional decision control point data to generate control point association data;

[0051] Step S44: using the control point association data to perform decision impact intensity analysis on the multidimensional decision space to generate decision impact intensity data;

[0052] Step S45: Perform influence domain coverage calculation on the decision influence intensity data to generate decision influence range data.

[0053] The present invention analyzes multidimensional decision trend prediction data to find the optimal decision combination. This can help decision makers determine the best decision solution among many choices and improve the effect and result of decision making. The multidimensional decision space is optimized by configuring the optimal solution data, and the value range and relationship structure of the decision variables are optimized. By optimizing the decision structure, the efficiency and feasibility of the decision space can be improved, making the decision solution more optimized and reliable. By performing association analysis on multidimensional decision control point data, the correlation and association rules between control points are found. The control point association data can reveal the key factors and influencing factors in the decision space and help decision makers understand the interaction and influence relationship between decision variables. The control point association data is used to analyze the influence intensity and influence direction between decision variables. The decision influence intensity data can help decision makers identify which decision variables have the most significant impact on the final result, so as to better understand the important decision factors in the decision space. The decision influence intensity data is used to calculate the influence range of the decision variables. The influence domain coverage data can help decision makers determine which areas in the decision space are significantly affected by the decision variables, so as to make targeted decision adjustments and optimizations.

[0054] Preferably, the specific steps of step S5 are:

[0055] Step S51: performing decision transfer path analysis on the multi-dimensional decision control point data according to the decision influence range data to generate decision transfer path data;

[0056] Step S52: performing central path detection on the decision transfer path data to generate decision transfer central path data;

[0057] Step S53: performing transmission efficiency analysis on the decision transmission center path data to generate decision transmission efficiency data;

[0058] Step S54: performing dynamic evolution analysis on the decision transmission center path data according to the decision transmission efficiency data to generate decision transmission path evolution data;

[0059] Step S55: performing external entity association analysis on the multi-dimensional decision control point data according to the decision transfer path evolution data to generate entity attribute association data;

[0060] Step S56: performing a correlation network topology analysis on the entity attribute correlation data to generate external topology structure data;

[0061] Step S57: constructing a dynamic decision network for the external topology data to generate an external decision topology network.

[0062] The present invention analyzes the transmission paths between multi-dimensional decision control points through decision influence range data. By analyzing the decision transmission path, the transmission relationship and path flow between decision control points can be understood, which helps to understand the transmission mode of information and influence in the decision process. The decision transmission path data is analyzed to find the central path. The central path represents the decision control point that has a greater impact on the decision result and a more important transmission path. By generating the central path data, decision makers can be helped to identify and focus on those decision control points that have an important impact on the decision result. The transmission efficiency between decision control points is analyzed using the decision transmission central path data. The transmission efficiency data can measure the transmission speed and efficiency of decision information on the central path, help decision makers understand the efficiency of information transmission, and identify the bottleneck and improvement space of information flow. Based on the decision transmission efficiency data, the decision transmission central path data is dynamically evolved. By analyzing the evolution of the decision transmission path, the change trend and evolution law of the path in the decision process can be observed, which helps decision makers understand the dynamic evolution process of the decision. The decision transmission path evolution data is used to analyze the association between multi-dimensional decision control points and external entities. Entity attribute association data can help decision makers understand the degree of association and attribute relationship between decision control points and external entities, so as to better grasp the external factors and constraints in the decision-making process. Analyze entity attribute association data and construct the association network topology between external entities. External topology data can reveal the association relationship, dependency relationship and network topology characteristics between entities, helping decision makers better understand the structure and organization of external entity systems. Based on external topology data, an external decision topology network is constructed. The external decision topology network can model decision control points, entities and their association relationships in the form of a network, helping decision makers to fully grasp the structure and organization of the decision system, identify key nodes and paths, and conduct decision analysis and optimization.

[0063] Preferably, the specific steps of step S6 are:

[0064] Step S61: performing multidimensional feature distribution analysis on the multidimensional decision optimization space to obtain multidimensional feature distribution data;

[0065] Step S62: performing probability distribution calculation on the multi-dimensional feature distribution data to generate multi-dimensional feature probability distribution data;

[0066] Step S63: Perform decision path space analysis on the multi-dimensional feature probability distribution data to generate intrinsic decision path space data;

[0067] Step S64: performing decision path network analysis on the external decision topology network to generate external decision path network data;

[0068] Step S65: reconstructing the internal decision path space data and the external decision path network data to construct a multiple decision space framework;

[0069] Step S66: construct a knowledge graph for the multiple decision space framework and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

[0070] The present invention analyzes the features in the multidimensional decision optimization space to understand the distribution of each feature within different value ranges. Through the multidimensional feature distribution data, statistical information about the features can be obtained to help decision makers understand the distribution range, distribution density and outliers of the feature values. Based on the multidimensional feature distribution data, the probability distribution of each feature within different value ranges is calculated. The multidimensional feature probability distribution data can provide probability information of the feature values ​​to help decision makers understand the probability of occurrence of each feature value, so as to better evaluate the importance and influence of the feature values. The spatial distribution of the decision path is analyzed using the multidimensional feature probability distribution data. The internal decision path space data can provide spatial distribution information about the decision path, helping decision makers understand the distribution of the decision path in the feature space, identify important decision paths and potential decision bottlenecks, and the external decision topology network. The network is analyzed to identify the decision path network. The external decision path network data can provide the connection relationship and transmission path between the decision paths, helping decision makers understand the interaction and influence of different paths in the decision-making process. The internal decision path space data and the external decision path network data are integrated and reconstructed to build a multiple decision space framework. The multiple decision space framework can integrate decision data from different perspectives to form a comprehensive decision space framework, which helps decision makers to comprehensively consider the influence of different factors and levels. Based on the multiple decision space framework, a knowledge graph of transaction decisions is constructed. The transaction decision knowledge graph can integrate and organize the data and information obtained in each step to form a structured knowledge representation form, which can include concepts, attributes, relationships, etc. related to power trading, helping decision makers to better understand and apply the knowledge in the decision space.

[0071] In this specification, a processing system based on power trading decision data is provided, which is used to execute the power trading method based on multi-dimensional data analysis as described above, including:

[0072] The multidimensional decomposition module is used to obtain real-time power transaction data; perform time series analysis on real-time power transaction data to generate power transaction time series data; perform transaction decision analysis on power transaction time series data to generate transaction time series decision data; perform multidimensional non-uniform pulse decomposition on transaction time series decision data using spline decomposition technology to generate multidimensional decision control point data;

[0073] The decision space module is used to perform spatial position analysis on multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; perform multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point spatial coordinate data to construct a multi-dimensional decision space; perform control point temporal change analysis on the multi-dimensional decision space to generate control point temporal change data;

[0074] The evolution analysis module is used to analyze the evolution trajectory of the control point time series change data to generate the control point evolution trajectory data; perform spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law; and predict the evolution trend of the multi-dimensional decision space through the decision control point evolution law to generate multi-dimensional decision trend prediction data;

[0075] The association mining module is used to optimize the decision structure of the multidimensional decision space using the multidimensional decision trend prediction data to generate the multidimensional decision optimization space; to perform control point association mining on the multidimensional decision control point data to generate control point association data; to perform influence domain coverage calculation on the multidimensional decision space using the control point association data to generate decision influence range data;

[0076] The decision network module is used to perform decision transfer path analysis on the multi-dimensional decision control point data according to the decision impact range data to generate decision transfer path data; perform associated network topology analysis on the decision transfer path data to generate external topology structure data; perform dynamic decision network construction on the external topology structure data to generate an external decision topology network;

[0077] The knowledge graph module is used to perform decision path space analysis on the multi-dimensional decision optimization space to generate internal decision path space data; perform decision path network analysis on the external decision topology network to generate external decision path network data; construct a knowledge graph for the internal decision path space data and the external decision path network data, and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

[0078] The present invention obtains real-time power transaction data through a multidimensional decomposition module, which is crucial for timely decision analysis. By performing time series analysis on real-time power transaction data, time series data of power transactions can be generated, which helps to understand the time correlation and trend of transactions. The decision space module performs spatial position analysis on multidimensional decision control point data to generate spatial coordinate data of multidimensional control points, which helps to determine the position of decision control points in the decision space for subsequent analysis and optimization. Based on the spatial coordinate data of multidimensional control points, multidimensional coordinate mapping is performed to construct a multidimensional decision space, which helps to visualize and analyze multiple dimensions of decision control points in a unified space. Time series change analysis of control points in the multidimensional decision space is performed to generate The time series change data of the control points is generated, which helps to understand the changing trends and patterns of the decision control points over time. The evolution analysis module performs evolution trajectory analysis on the time series change data of the control points and generates evolution trajectory data of the control points, which helps to observe the evolution process and trend of the decision control points in the decision space. By analyzing the evolution trajectory data of the control points, spatial morphological structure drift analysis can be performed, that is, to understand the structural changes of the decision control points in the decision space, which helps to discover the potential patterns and association laws in the decision space. According to the evolution laws of the decision control points, the evolution trend of the multidimensional decision space is predicted, and multidimensional decision trend prediction data is generated, which helps to predict and plan the future changes in the decision space. The association mining module uses the multidimensional decision trend prediction data Optimize the decision structure of the multidimensional decision space and generate a multidimensional decision optimization space, which helps to optimize the structure and performance of the decision space and improve the decision effect. By performing control point association mining on the multidimensional decision control point data, the associated data of the control points can be generated, which helps to discover the association and mutual influence between the decision control points. Using the control point association data, the influence domain coverage calculation of the multidimensional decision space is performed to generate decision influence range data, which helps to determine the degree and scope of influence of the decision control point on the decision space. The decision network module performs associated network topology analysis on the decision transmission path data and generates external topology structure data, which helps to reveal the network relationship and topology structure between the decision control points and further understand the overall nature of the decision system. The external topological structure data is used to construct a dynamic decision network and generate an external decision topological network, which helps to establish a network model of the decision system and provide a dynamic perspective and analysis capability in the decision-making process. The knowledge graph module performs decision path space analysis on the multidimensional decision optimization space and generates internal decision path space data, which helps to understand the characteristics and optimization directions of different paths in the decision space. The decision path network analysis is performed on the external decision topological network to generate external decision path network data, which helps to understand the association and influence relationship between different paths in the decision system. Based on the internal decision path space data and the external decision path network data, a transaction decision knowledge graph is constructed, which helps to organize and integrate decision-related knowledge and information and provide decision support and reference. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] FIG1 is a schematic diagram of a process flow of a method for processing power transaction decision data according to the present invention;

[0080] FIG. 2 is a schematic diagram of a detailed implementation process of step S1. DETAILED DESCRIPTION

[0081] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0082] The present application example provides a processing method and system based on power trading decision data. The execution subject of the processing method and system based on power trading decision data includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0083] Please refer to FIG. 1 and FIG. 2 , the present invention provides a processing method based on power transaction decision data, and the processing method based on power transaction decision data includes the following steps:

[0084] Step S1: acquiring real-time power transaction data; performing time series analysis on the real-time power transaction data to generate power transaction time series data; performing transaction decision analysis on the power transaction time series data to generate transaction time series decision data; performing multi-dimensional non-uniform pulse decomposition on the transaction time series decision data using spline decomposition technology to generate multi-dimensional decision control point data;

[0085] Step S2: Perform spatial position analysis on the multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; perform multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point spatial coordinate data to construct a multi-dimensional decision space; perform control point temporal change analysis on the multi-dimensional decision space to generate control point temporal change data;

[0086] Step S3: performing evolution trajectory analysis on the control point time series change data to generate control point evolution trajectory data; performing spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law; performing evolution trend prediction on the multidimensional decision space through the decision control point evolution law to generate multidimensional decision trend prediction data;

[0087] Step S4: Optimize the decision structure of the multidimensional decision space using the multidimensional decision trend prediction data to generate a multidimensional decision optimization space; perform control point association mining on the multidimensional decision control point data to generate control point association data; perform influence domain coverage calculation on the multidimensional decision space using the control point association data to generate decision influence range data;

[0088] Step S5: performing decision transfer path analysis on the multi-dimensional decision control point data according to the decision impact range data to generate decision transfer path data; performing associated network topology analysis on the decision transfer path data to generate external topology structure data; performing dynamic decision network construction on the external topology structure data to generate an external decision topology network;

[0089] Step S6: Perform decision path space analysis on the multidimensional decision optimization space to generate internal decision path space data; perform decision path network analysis on the external decision topology network to generate external decision path network data; construct a knowledge graph for the internal decision path space data and the external decision path network data, and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

[0090] The present invention can obtain accurate power trading time series data by performing time series analysis on real-time power trading data. Such analysis can reveal the trading rules, trends and periodic changes in the power market, and provide basic data for subsequent trading decisions. Trading decision analysis using power trading time series data can generate trading time series decision data, which can be used to evaluate the effectiveness of trading strategies, risk management and profit prediction. By performing spatial position analysis and multidimensional coordinate mapping on multidimensional decision control point data, a multidimensional decision space can be constructed. Such space construction can convert decision problems into multidimensional coordinate spaces, which is convenient for decision analysis and optimization. Time series change analysis on multidimensional decision control point data can obtain control point Time series change data, these data help to understand the evolution trend and key moments of decision control points, and provide a time series basis for decision-making. By performing evolution trajectory analysis and spatial morphological structure drift analysis on the time series change data of control points, the evolution law of decision control points can be generated, which helps to understand the change trend and regularity of decision control points and provide a basis for decision prediction and optimization. Based on the evolution law of decision control points, the evolution trend of multidimensional decision space is predicted, and multidimensional decision trend prediction data can be generated. These data can be used to predict future decision changes and trends and guide decision making. The decision structure of multidimensional decision space is optimized using multidimensional decision trend prediction data, and a multidimensional decision optimization space can be generated. This space can provide more optimized decisions. By mining the association of multi-dimensional decision control points and calculating the coverage of the influence domain, decision influence range data can be generated. These data can reveal the mutual influence and transmission relationship between decisions, help identify important decision factors and potential risks, and conduct path analysis on the decision influence range data to reveal the transmission path between decision control points and the way of decision transmission, helping decision makers understand the transmission mechanism and path selection of decisions. By conducting association network topology analysis on decision transmission path data, external topology structure data can be constructed to show the association relationship and network structure between decision control points, providing decision makers with a global perspective and overall understanding. By conducting association network topology analysis, The external decision topology network organizes and displays the decision control points in the form of a network, providing decision makers with intuitive decision topology structures and related information. The decision path space analysis is performed on the multidimensional decision optimization space, which can reveal different decision paths in the space. This helps to understand the characteristics, advantages and disadvantages of different decision paths, and provide decision makers with diversity and selectivity of decision plans. The external decision path network analysis is based on the decision topology network. By analyzing the network structure and node relationships, the location and connection method of different decision paths in the overall network are identified. The knowledge graph is a structured knowledge representation form. By integrating and modeling the internal decision path space data and the external decision path network data, a transaction decision knowledge graph is constructed.The construction of knowledge graph helps to express and store the information and relevance of decision paths, forming a knowledge graph database, which provides a visual and queryable way for decision makers to more easily obtain and use knowledge and information related to transaction decisions.

[0091] In this embodiment, referring to FIG. 1 , a schematic flow chart of a method for processing power transaction decision data according to the present invention is shown. In this example, the method for processing power transaction decision data according to the present invention includes the following steps:

[0092] Step S1: acquiring real-time power transaction data; performing time series analysis on the real-time power transaction data to generate power transaction time series data; performing transaction decision analysis on the power transaction time series data to generate transaction time series decision data; performing multi-dimensional non-uniform pulse decomposition on the transaction time series decision data using spline decomposition technology to generate multi-dimensional decision control point data;

[0093] In this embodiment, real-time power transaction data is obtained from the power market or exchange. These data may include information such as power prices, supply and demand, and trading behaviors of market participants. The transaction data is timestamped to ensure the time sequence of the data. Time series analysis methods such as trend analysis, periodicity analysis, and seasonality analysis are performed on the transaction data to generate a time series of power transactions. Time is used as the horizontal axis and transaction data (such as price, transaction volume, etc.) is used as the vertical axis. According to the trend, periodicity and other characteristics of the transaction data, the transaction strategy is formulated and optimized. Statistical analysis, machine learning and other methods are used to perform pattern recognition and prediction on the transaction data to assist decision making. The transaction time series decision data is spline interpolated to convert the time series data into a continuous function form. The spline interpolation technology of non-uniform pulses is used to decompose the continuous function into multiple non-uniform pulses. According to the decomposition results, multi-dimensional decision control point data is generated. Each control point contains values ​​of multiple dimensions, reflecting the key parameters or characteristics of the transaction decision.

[0094] Step S2: Perform spatial position analysis on the multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; perform multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point spatial coordinate data to construct a multi-dimensional decision space; perform control point temporal change analysis on the multi-dimensional decision space to generate control point temporal change data;

[0095] In this embodiment, the dimensions and parameters in the multidimensional decision control point data are determined, and based on these dimensions and parameters, the position of each control point is mapped to a multidimensional space to generate spatial coordinate data of the multidimensional control point. Each control point has corresponding coordinates representing its position in the multidimensional space. Based on the spatial coordinate data of the multidimensional control point, the coordinate axis of the multidimensional decision space is determined, and each control point in the multidimensional decision control point data is mapped to a corresponding coordinate point in the multidimensional decision space to construct a multidimensional decision space. The position of each control point in the space corresponds to its value in the multidimensional decision control point data. For each control point in the multidimensional decision space, its change trend over time is observed, and the position change of the control point at different time points is analyzed. Statistical methods, visualization methods, etc. can be used for analysis to generate control point time series change data and record the position change information of each control point at different time points.

[0096] Step S3: performing evolution trajectory analysis on the control point time series change data to generate control point evolution trajectory data; performing spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law; performing evolution trend prediction on the multidimensional decision space through the decision control point evolution law to generate multidimensional decision trend prediction data;

[0097] In this embodiment, time series analysis is performed on the time series change data of the control points to understand the change trend of the control points over time. The evolution trajectory of the control points in the multidimensional decision space is drawn according to the time series change data. Visual tools or algorithms can be used for drawing. The spatial morphological structure changes in the control point evolution trajectory data are analyzed. The spatial positions and relative relationships of different time points are compared. Clustering, classification or machine learning methods are used to identify and analyze the drift laws and patterns in the control point evolution trajectory data, generate evolution law data of the decision control points, describe the evolution trend and change law of the control points in the multidimensional decision space, and use the evolution law data of the control points to establish a prediction model or algorithm to predict the evolution trend in the multidimensional decision space. Considering other influencing factors, such as market changes, policy adjustments, etc., the prediction model is corrected and optimized to generate multidimensional decision trend prediction data to describe the evolution trend and change prediction of the multidimensional decision space in the future time period.

[0098] Step S4: Optimize the decision structure of the multidimensional decision space using the multidimensional decision trend prediction data to generate a multidimensional decision optimization space; perform control point association mining on the multidimensional decision control point data to generate control point association data; perform influence domain coverage calculation on the multidimensional decision space using the control point association data to generate decision influence range data;

[0099] In this embodiment, multidimensional decision trend prediction data is analyzed to identify potential optimization space and improvement opportunities. According to the prediction data, the multidimensional decision space is structurally adjusted, parameters are optimized, or decision rules are improved to generate a multidimensional decision optimization space, which includes the optimized decision structure and parameter settings. The correlation and relevance between the multidimensional decision control point data are analyzed. The association rules, correlation coefficients, clustering and other methods are used to explore the association relationship between the control points, generate control point association data, describe the degree of association and association rules between different control points, determine the degree and scope of influence of each control point on other control points based on the control point association data, calculate the influence domain of the control point, and use distance measurement, weight calculation and other methods to generate decision influence range data to describe the scope and degree of influence of each control point on other control points in the multidimensional decision space.

[0100] Step S5: performing decision transfer path analysis on the multi-dimensional decision control point data according to the decision impact range data to generate decision transfer path data; performing associated network topology analysis on the decision transfer path data to generate external topology structure data; performing dynamic decision network construction on the external topology structure data to generate an external decision topology network;

[0101] In this embodiment, based on the decision influence range data, the influence degree and transmission path of each control point on other control points are determined, the transmission paths between the control points are analyzed, including direct transmission paths and indirect transmission paths, decision transmission path data are generated, the transmission paths and influence relationships between the control points are described, and based on the decision transmission path data, an association network between the control points is constructed, the topological structure characteristics of the association network, such as node degree, clustering coefficient, network connectivity, etc., are analyzed, the module structure or subgraph structure of the association network is explored, and key nodes or subnetworks are identified, and external topological structure data are generated to describe the topological characteristics and structure of the association network formed by the decision transmission path data. Based on the external topological structure data, the initial state of the dynamic decision network is constructed, the time evolution of the decision transmission path data is considered, the node state and connection weight of the dynamic decision network are updated, and the dynamic decision network is evolved and simulated using a network model or algorithm to simulate decision transmission and decision-making process, generate an external decision topological network, and describe the evolution of the decision transmission path data over time and the structure of the dynamic decision network.

[0102] Step S6: Perform decision path space analysis on the multidimensional decision optimization space to generate internal decision path space data; perform decision path network analysis on the external decision topology network to generate external decision path network data; construct a knowledge graph for the internal decision path space data and the external decision path network data, and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

[0103] In this embodiment, based on the multidimensional decision optimization space, the possibility and feasibility of different decision paths are determined, the key nodes and decision nodes of the decision path, and the paths between the nodes are analyzed, the constraints and optimization objectives of the decision path are considered, the optimal or most appropriate decision path is determined, the internal decision path space data is generated, the possible decision paths and related information in the multidimensional decision optimization space are described, based on the external decision topology network, the connections and paths between different decision paths are identified, the network characteristics of the decision path, such as path length, path weight, node association, etc., are analyzed, the priority and influence of the decision path are considered, the main path and the secondary path are determined, the external decision path network data is generated, the connection relationship and related attributes between different decision paths in the external decision topology network are described, the internal decision path space data and the external decision path network data are converted into a representation form suitable for the knowledge graph, such as the form of graph nodes and edges, the nodes of the knowledge graph are established, including decision nodes, decision path nodes and other related nodes, the edges of the knowledge graph are established, the relationships, connections and influences between different nodes are represented, and the knowledge graph technology and algorithms are used to process and analyze the power trading decision data to extract useful information and knowledge.

[0104] In this embodiment, referring to FIG. 2 , which is a flowchart of detailed implementation steps of step S1, in this embodiment, the detailed implementation steps of step S1 include:

[0105] Step S11: obtaining real-time power transaction data;

[0106] Step S12: performing time series analysis on the real-time power transaction data to generate power transaction time series data;

[0107] Step S13: identifying decision factors for the power transaction time series data and generating decision factor data;

[0108] Step S14: performing transaction decision analysis on the power transaction time series data according to the decision factor data to generate transaction time series decision data;

[0109] Step S15: Use spline decomposition technology to perform multi-dimensional non-uniform pulse decomposition on the transaction timing decision data to generate multi-dimensional decision control point data.

[0110] The present invention provides decision makers with the latest market intelligence and transaction information by acquiring real-time power transaction data, enabling them to make timely decision responses. Real-time data reflects the current market conditions and transaction trends, helping decision makers to grasp the best

[0111] By analyzing the time series data of power trading, the data can be converted into the form of time series, making the trends and changes of the data more clearly visible. This helps decision makers to intuitively understand the dynamic changes of trading data and discover the patterns and trends in it. By analyzing the time series data of power trading, the trends and periodic changes in the data can be identified. Decision makers can use this trend information to make predictions in order to better grasp the market trend and make corresponding trading decisions. By analyzing the time series data of power trading, factors that have an important impact on trading decisions can be identified and extracted. These factors may include supply and demand relationships, price fluctuations, seasonal changes, etc., and decision makers can use these factor data as a basis for decision making. By conducting trading decision analysis on the time series data of power trading, decision makers can formulate more optimized trading strategies based on the decision factor data. This helps to maximize the use of market opportunities, reduce risks, and improve trading benefits. Using spline decomposition technology to decompose the trading time series decision data, the data can be analyzed and controlled in multiple dimensions. This helps decision makers to gain an in-depth understanding of all aspects of trading data and conduct decision analysis from multiple perspectives. Non-uniform pulse decomposition is a method that decomposes time series data into pulse signals of different frequencies and amplitudes. Through this decomposition, decision makers can observe and control the volatility and changes of trading data in a more fine-grained manner, thereby formulating decision strategies more accurately.

[0112] In this embodiment, appropriate methods and tools are used to obtain real-time power transaction data from data sources, which may include using API interfaces, data subscription services or data crawling technologies to obtain data, applying appropriate time series analysis methods and technologies to analyze power transaction data, such as time series models, trend analysis, periodicity analysis, etc., which can help identify patterns, trends and periodicity in the data, and generating power transaction time series data based on the results of the time series analysis, which includes transaction time points and corresponding transaction data, such as electricity, price, transaction type, etc. Based on the extracted features, appropriate methods and algorithms are used to identify factors related to power transaction decisions, which may include statistical analysis, machine learning or knowledge of domain experts, and generating decision factor data based on the identification results of decision factors, which includes identified decision factors and their corresponding values, and analyzing and making decisions on power transaction time series data based on defined decision rules, which may involve According to the results of transaction decision analysis, transaction time series decision data are generated by combining decision models, optimization algorithms, risk assessment and other methods, which include specific decision actions, transaction strategies or other relevant decision indicators at each transaction time point. Spline decomposition technology is applied to decompose the transaction time series decision data. Spline decomposition is a mathematical method used to decompose complex time series data into multiple components. On the basis of spline decomposition, multi-dimensional non-uniform pulse decomposition technology is used to further process the transaction time series decision data. This technology can decompose the time series data into multiple dimensions and capture the changes and characteristics of the data in a non-uniform manner. According to the results of multi-dimensional non-uniform pulse decomposition, multi-dimensional decision control point data are generated. These control point data can reflect the key points, change trends and important characteristics of the transaction time series decision data in different dimensions.

[0113] In this embodiment, the specific steps of step S15 are:

[0114] Step S151: segmenting the transaction timing decision data into timing windows to generate timing window data;

[0115] Step S152: extracting the transaction timing decision data by using the timing window data to generate a transaction timing decision sequence;

[0116] Step S153: using the spline decomposition technology to perform multi-dimensional non-uniform pulse decomposition on the transaction timing decision sequence to generate a one-dimensional pulse component data set;

[0117] Step S154: reconstructing a multi-dimensional spatial pulse path for the one-dimensional pulse component data set to generate a multi-dimensional spatial pulse path;

[0118] Step S155: performing pulse direction mutation analysis on the multidimensional space pulse path and marking the multidimensional space pulse mutation point;

[0119] Step S156: extract decision control nodes from the transaction timing decision sequence according to the multi-dimensional space pulse mutation points to generate multi-dimensional decision control point data.

[0120] The present invention divides the continuous time series data into multiple windows by performing time series window segmentation on the transaction time series decision data. Each window represents data within a relatively small time period. This segmentation helps to further analyze and process the data, making the characteristics of the data more clear and controllable. By performing sequence extraction on the time series window data, the data in each window can be arranged in chronological order to form a transaction time series decision sequence. This serialized data structure is more suitable for subsequent analysis and processing, and is convenient for applying various sequence models and algorithms. The transaction time series decision sequence is pulse decomposed by spline decomposition technology, and the sequence can be decomposed into different pulse components. Each pulse component represents a local fluctuation or change in the sequence. This helps to capture important features and change patterns in the sequence. By reconstructing the multidimensional space pulse path, a one-dimensional pulse component data set can be converted into a path in the multidimensional space. This visual presentation method helps to intuitively observe and understand the relationship and evolution trend between different pulse components. By performing pulse direction mutation analysis on the multidimensional space pulse path, the mutation point in the path can be marked. The mutation point represents the location where the direction of the path changes significantly. This helps to identify important turning points and key moments in the transaction decision data. According to the multi-dimensional spatial pulse mutation points, decision control nodes can be extracted from the transaction timing decision sequence. These nodes represent important time points and key decision stages in the decision-making process. Extracting these nodes helps to analyze and control the key links of decision-making. The generation of multi-dimensional decision control point data enables decision analysis and control in various dimensions. By comprehensively analyzing decision control points in different dimensions, more comprehensive and accurate decision guidance can be obtained, improving the effect and accuracy of power trading decisions.

[0121] In this embodiment, the size of the timing window is determined, that is, the time span of the transaction timing decision data contained in each window. The selection of the window size can be determined according to specific needs and data characteristics. The sliding window method is applied to the transaction timing decision data, the window is applied to the starting position of the transaction timing decision data, and the window is slid with a fixed step size according to the set window size. The transaction timing decision data in each timing window data is subjected to sequence extraction. The specific sequence to be extracted can be selected as needed, such as a price sequence, a trading volume sequence, etc. The extracted sequence data are combined to form a transaction timing decision sequence. Each transaction timing decision sequence contains a specific sequence extracted from the timing window data. The spline decomposition technology is applied to decompose the transaction timing decision sequence to extract the non-uniform pulse component in the sequence. The spline decomposition can decompose the sequence into multiple non-uniform pulse components. According to the result of the spline decomposition, a one-dimensional pulse component data set is generated. The data set contains the one-dimensional non-uniform pulse component extracted from the transaction timing decision sequence. The one-dimensional pulse component The data set is used to reconstruct the multidimensional space pulse path, which can be achieved by combining and reconstructing the one-dimensional pulse components in different dimensions. According to the results of the multidimensional space pulse path reconstruction, a multidimensional space pulse path is generated. The multidimensional space pulse path contains the path information of the transaction timing decision data in different dimensions. The pulse direction change analysis is performed on the multidimensional space pulse path, which can be determined by comparing the differences between adjacent pulse points. According to the results of the pulse direction mutation analysis, the pulse mutation points in the multidimensional space pulse path are marked. The pulse mutation points represent the points where the pulse direction changes significantly in the multidimensional space. According to the positions of the multidimensional space pulse mutation points, the decision control nodes of the transaction timing decision sequence are extracted. The decision control nodes are the key points marked in the transaction timing decision sequence corresponding to the pulse mutation points. According to the extracted decision control nodes, multidimensional decision control point data are generated. The data contains the key point information corresponding to the pulse mutation points in the transaction timing decision sequence, which is used for subsequent decision analysis and control.

[0122] In this embodiment, the specific steps of step S2 are:

[0123] Step S21: Perform spatial position analysis on the multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data;

[0124] Step S22: performing multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point space coordinate data to construct a multi-dimensional decision space;

[0125] Step S23: performing cluster analysis on the multi-dimensional decision control point data to generate decision control point cluster data;

[0126] Step S24: dividing the multidimensional decision space into cluster regions according to the decision control point cluster data to generate a multidimensional decision space region;

[0127] Step S25: performing control point time series change analysis on the multi-dimensional decision space region to generate control point time series change data.

[0128] Through spatial position analysis, the present invention can assign specific spatial coordinates to each multidimensional decision control point. These coordinates represent the position of each control point in the multidimensional decision space, which helps to understand and visualize the spatial relationship between the control points. Through multidimensional coordinate mapping, the multidimensional decision control point data can be mapped to the multidimensional decision space, thus establishing a multidimensional space based on decision points for representing and analyzing the multidimensional features of decisions. The construction of the multidimensional decision space enables the analysis of decisions in multiple dimensions. By observing and comparing the distribution and change trends of decision points in different dimensions, more comprehensive and accurate decision information can be obtained, and the effect of power trading decisions can be further optimized. Through cluster analysis, The multidimensional decision control point data can be divided into different clusters. Each cluster represents a group of control points with similar decision characteristics. The generation of clusters helps to discover potential patterns and group behaviors in decision data. According to the decision control point clustering data, the multidimensional decision space can be divided into different regions. Each region contains control points with similar decision characteristics. This spatial division helps to subdivide the decision space into different subspaces, so as to better understand and analyze the characteristics and patterns of decisions. By performing time series change analysis on the control points in the multidimensional decision space region, the time evolution characteristics of the decision can be revealed, which helps to understand the changing trends and patterns of decisions in different time periods and provide a reference for predicting future decision changes.

[0129] In this embodiment, spatial position analysis is performed on the multidimensional decision control point data to determine the position of each decision control point in the multidimensional decision space, which may involve data normalization, standardization, coordinate conversion and other processing. Based on the result of the spatial position analysis, multidimensional control point spatial coordinate data is generated, and the data includes the coordinate information of each decision control point in the multidimensional decision space. Based on the multidimensional control point spatial coordinate data, multidimensional coordinate mapping is performed on the multidimensional decision control point data, which may involve mapping the multidimensional decision control point data to the corresponding coordinate position in the multidimensional decision space. Based on the result of the multidimensional coordinate mapping, a multidimensional decision space is constructed. The multidimensional decision space is a space mapped from the multidimensional decision control point data, which is used to represent the relationship between decision control points in multiple dimensions. A clustering algorithm is applied to the multidimensional decision control point data to classify decision control points with similar characteristics into one category. Commonly used clustering algorithms include K Mean clustering and hierarchical clustering, based on the results of cluster analysis, generate decision control point clustering data, which contains the information of the cluster category or cluster to which each decision control point belongs. According to the decision control point clustering data, the multidimensional decision space is divided into different clustering regions. Each clustering region contains the spatial region where decision control points with similar characteristics are located. According to the results of regional division, a multidimensional decision space region is generated. This data represents the scope and boundary of different clustering regions in the multidimensional decision space. A time series change analysis is performed on the decision control points in the multidimensional decision space region, which may involve the changes, trends and patterns of the control points at different time points. According to the results of the time series change analysis, the control point time series change data is generated. This data contains the time series change information of the decision control points in the multidimensional decision space region, which can be used to analyze and predict the behavior and evolution trend of the control points.

[0130] In this embodiment, the specific steps of step S3 are:

[0131] Step S31: Identify edge control points in the multidimensional decision space area and mark edge decision control points;

[0132] Step S32: performing layout structure analysis on the multidimensional decision space based on the edge decision control points to generate decision space layout structure data;

[0133] Step S33: performing evolution trajectory analysis on the control point time series change data through the multi-dimensional decision space layout structure data to generate control point evolution trajectory data;

[0134] Step S34: performing spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law;

[0135] Step S35: Predict the evolution trend of the multidimensional decision space according to the evolution law of the decision control points to generate multidimensional decision trend prediction data.

[0136] The present invention can determine which decision points are located at the edge of the decision space by identifying edge control points in the multidimensional decision space area. Edge decision points often have special significance and influence in the decision-making process, so marking these points can help identify key decision points. The marking of edge decision control points helps to analyze the boundary characteristics of the decision space. By observing the distribution and changes of edge decision points, the boundary morphology of the decision space can be understood, and the structure and characteristics of the decision space can be further understood. By performing layout structure analysis based on edge decision control points, the overall layout and structural characteristics of the decision space can be revealed, which helps to understand the relative positions and relationships between different decision points in the decision space, and further explore the overall laws and patterns of decision-making. By performing evolution trajectory analysis on the time series change data of control points through multidimensional decision space layout structure data, the control points can be revealed. The changing trajectory of the points in the decision space helps to understand the evolution process of the decision points over time and identify the development trend and change pattern of the decision. By performing spatial morphological structure drift analysis on the control point evolution trajectory data, the morphological structure change law of the decision points in the decision space can be discovered, which helps to understand the relationship evolution between decision points and the dynamic changes of the decision space, revealing the inherent laws and mechanisms of decision-making. Based on the evolution law of the decision control points, the evolution trend of the multidimensional decision space can be predicted, which can help identify potential changes and trends in the decision space, predict possible decision patterns and development directions in advance, and support the generation of decision trend prediction data to provide support for decision making and decision evaluation. Decision makers can adjust decision strategies and directions based on multidimensional decision trend prediction data to cope with possible decision changes and challenges in the future.

[0137] In this embodiment, edge recognition is performed on decision control points in the multidimensional decision space region, which may involve edge detection algorithms, such as the Canny edge detection algorithm or the Sobel operator, etc. According to the result of edge recognition, edge decision control points are marked. These decision control points are located in the edge region of the multidimensional decision space and may have special importance or characteristics. Based on the edge decision control points, a layout structure analysis is performed on the multidimensional decision space, which may involve methods such as graph theory and network analysis, such as the shortest path algorithm, connectivity analysis and centrality measurement, etc. According to the result of the layout structure analysis, decision space layout structure data is generated, which describes the layout relationship and structural characteristics between decision control points in the multidimensional decision space. Using the multidimensional decision space layout structure data, an evolution trajectory analysis is performed on the time series change data of the control points, which may involve trajectory analysis algorithms and time series analysis methods, such as distance-based similarity measurement, clustering algorithm and time series analysis method. Sequence models, etc., generate control point evolution trajectory data based on the results of evolution trajectory analysis, which describes the evolution path and change trend of the control point in the multidimensional decision space, and perform spatial morphological structure drift analysis on the control point evolution trajectory data, which may involve methods such as morphological analysis, clustering analysis and pattern recognition, which are used to identify morphological structure changes in the evolution trajectory, and use the evolution law of decision control points to predict the evolution trend of the multidimensional decision space, which may involve time series prediction methods, machine learning algorithms and model inference technologies, which are used to predict the future evolution trend of decision control points, and generate multidimensional decision trend prediction data based on the results of evolution trend prediction, which describes the future change trend and possible development direction of decision control points in the multidimensional decision space.

[0138] In this embodiment, the specific steps of step S34 are:

[0139] Step S341: performing control point distribution entropy quantization processing on the multi-dimensional decision control point data to generate a spatial control point distribution entropy index;

[0140] Step S342: performing spatial morphological change amplitude analysis on the control point evolution trajectory data according to the spatial control point distribution entropy index to generate morphological change amplitude data;

[0141] Step S343: performing spatial morphological structure drift analysis on the multidimensional decision space according to the morphological change amplitude data to generate spatial morphological structure drift data;

[0142] Step S344: Calculating the morphological drift rate of the spatial morphological structure drift data based on the time series window data to generate morphological drift rate data;

[0143] Step S345: performing control point evolution analysis on the control point evolution trajectory data according to the morphological drift rate data to generate a decision control point evolution law.

[0144] The present invention can measure the uniformity of control point distribution in the decision space by performing entropy quantization processing on multi-dimensional decision control point data. The spatial control point distribution entropy index reflects the dispersion and concentration of control points in the decision space, and provides a quantitative indicator to evaluate the balance and diversity of the decision space. According to the spatial control point distribution entropy index, the control point evolution trajectory data is analyzed to measure the morphological change amplitude of the decision point in the decision space. The morphological change amplitude data reflects the degree of change of the decision point in space, including changes in position, density, etc., and provides basic data for subsequent morphological drift analysis. According to the morphological change amplitude data, the morphological structure drift analysis of the multi-dimensional decision space is performed. By analyzing the spatial characteristics and patterns in the morphological change amplitude data, the morphological structure drift of the decision space, that is, the change in the relationship between the decision points, can be revealed. This helps to understand the dynamic evolution process of the decision space and the changing laws of the spatial structure. By calculating the spatial morphological structure drift data based on the time series window data, the morphological drift rate data can be obtained. The morphological drift rate refers to the rate and degree of morphological structure drift in the decision space, reflecting the dynamic change intensity of the decision space. The morphological drift rate data provides a measure of the evolution speed of the decision space, which can help discover the rapid changes and potential instability of the decision space. According to the morphological drift rate data, the control point evolution trajectory data is analyzed to reveal the evolution law of the decision control point. By analyzing the relationship between the trajectory changes of the decision control point in the decision space and the morphological drift rate, the correlation between the decision control points, the diversity of the decision points, and the evolution trend can be found. This helps to understand the evolution law of the decision control points and provide reference and guidance for power trading decisions.

[0145] In this embodiment, entropy quantization processing is performed on the multi-dimensional decision control point data. Information entropy or other entropy measurement methods can be used to calculate the distribution entropy of the decision control points in each dimension. According to the control point distribution entropy, a spatial control point distribution entropy index is generated. The index represents the uniformity or concentration of the control point distribution in the multi-dimensional decision space. According to the spatial control point distribution entropy index, the spatial morphological change amplitude analysis is performed on the control point evolution trajectory data, which may involve calculating the distance between the control points, the change in direction or the change in morphological features, etc. The morphological change amplitude data is used to perform spatial morphological structure drift analysis on the multi-dimensional decision space. Methods such as morphological analysis, cluster analysis or structural feature extraction can be considered to detect the morphological structure drift phenomenon in the space, and the spatial morphological structure drift data is divided into different time series windows, each window contains drift data within a period of time, and the morphological drift rate is calculated for the spatial morphological structure drift data in each time series window. Appropriate indicators or algorithms, such as morphological structure similarity measurement, drift degree assessment or pattern matching, can be used to calculate the drift rate. According to the result of the morphological drift rate calculation, morphological drift rate data are generated, which describes the rate or change frequency of morphological drift in the multidimensional decision space. The control point evolution trajectory data are analyzed by using the morphological drift rate data. The law, trend or mode of the control point evolution can be judged according to the change of the morphological drift rate. According to the analysis of the control point evolution trajectory data, the decision control point evolution law is generated, which can be a mode, trend or law describing the movement, change or development of the control point in the multidimensional decision space.

[0146] In this embodiment, step S4 includes the following steps:

[0147] Step S41: performing optimal solution decision analysis on the multi-dimensional decision trend prediction data to generate configuration optimal solution data;

[0148] Step S42: optimizing the decision structure of the multidimensional decision space by using the configuration optimal solution data to generate a multidimensional decision optimization space;

[0149] Step S43: performing control point association mining on the multi-dimensional decision control point data to generate control point association data;

[0150] Step S44: using the control point association data to perform decision impact intensity analysis on the multidimensional decision space to generate decision impact intensity data;

[0151] Step S45: Perform influence domain coverage calculation on the decision influence intensity data to generate decision influence range data.

[0152] The present invention analyzes multidimensional decision trend prediction data to find the optimal decision combination. This can help decision makers determine the best decision solution among many choices and improve the effect and result of decision making. The multidimensional decision space is optimized by configuring the optimal solution data, and the value range and relationship structure of the decision variables are optimized. By optimizing the decision structure, the efficiency and feasibility of the decision space can be improved, making the decision solution more optimized and reliable. By performing association analysis on multidimensional decision control point data, the correlation and association rules between control points are found. The control point association data can reveal the key factors and influencing factors in the decision space and help decision makers understand the interaction and influence relationship between decision variables. The control point association data is used to analyze the influence intensity and influence direction between decision variables. The decision influence intensity data can help decision makers identify which decision variables have the most significant impact on the final result, so as to better understand the important decision factors in the decision space. The decision influence intensity data is used to calculate the influence range of the decision variables. The influence domain coverage data can help decision makers determine which areas in the decision space are significantly affected by the decision variables, so as to make targeted decision adjustments and optimizations.

[0153] In this embodiment, the optimal solution decision analysis is performed on the multidimensional decision trend prediction data. This may include using optimization algorithms, decision tree analysis, multi-objective optimization methods, etc. to determine the optimal solution decision within a given decision variable range. According to the results of the optimal solution decision analysis, the configuration optimal solution data is generated. This data describes the optimal decision configuration or solution in the multidimensional decision space. The decision structure of the multidimensional decision space is optimized using the configuration optimal solution data. This may involve adjusting the relationship between decision variables, optimizing the value range or constraint conditions of the decision variables, so as to achieve better decision results. According to the results of the decision structure optimization, a multidimensional decision optimization space is generated. This space describes the value range and relationship of the decision variables under the optimized decision structure. Control point association mining is performed on the multidimensional decision control point data. This may include methods such as association rule mining, correlation analysis, and network analysis to discover the association or mutual influence relationship between decision control points. According to the results of the control point association mining, control point association data is generated. This data describes the degree of association, association type, or association rule between decision control points. The decision influence intensity analysis is performed on the multidimensional decision space using the control point association data. This may involve calculating the strength, degree or range of influence between decision control points to evaluate their impact on the decision outcome. Based on the results of the decision influence strength analysis, decision influence strength data is generated. This data describes the strength or degree of influence of the decision control point on other variables in the multidimensional decision space. Perform an influence domain coverage calculation on the decision influence strength data. This may involve calculating the influence range or influence area corresponding to each decision control point, that is, the value range of other variables affected by it. Based on the results of the influence domain coverage calculation, decision influence range data is generated. This data describes the influence range or influence area corresponding to each decision control point, helping decision makers understand and evaluate the influence range of different decision control points.

[0154] In this embodiment, the specific steps of step S5 are:

[0155] Step S51: performing decision transfer path analysis on the multi-dimensional decision control point data according to the decision influence range data to generate decision transfer path data;

[0156] Step S52: performing central path detection on the decision transfer path data to generate decision transfer central path data;

[0157] Step S53: performing transmission efficiency analysis on the decision transmission center path data to generate decision transmission efficiency data;

[0158] Step S54: performing dynamic evolution analysis on the decision transmission center path data according to the decision transmission efficiency data to generate decision transmission path evolution data;

[0159] Step S55: performing external entity association analysis on the multi-dimensional decision control point data according to the decision transfer path evolution data to generate entity attribute association data;

[0160] Step S56: performing a correlation network topology analysis on the entity attribute correlation data to generate external topology structure data;

[0161] Step S57: constructing a dynamic decision network for the external topology data to generate an external decision topology network.

[0162] The present invention analyzes the transmission paths between multi-dimensional decision control points through decision influence range data. By analyzing the decision transmission path, the transmission relationship and path flow between decision control points can be understood, which helps to understand the transmission mode of information and influence in the decision process. The decision transmission path data is analyzed to find the central path. The central path represents the decision control point that has a greater impact on the decision result and a more important transmission path. By generating the central path data, decision makers can be helped to identify and focus on those decision control points that have an important impact on the decision result. The transmission efficiency between decision control points is analyzed using the decision transmission central path data. The transmission efficiency data can measure the transmission speed and efficiency of decision information on the central path, help decision makers understand the efficiency of information transmission, and identify the bottleneck and improvement space of information flow. Based on the decision transmission efficiency data, the decision transmission central path data is dynamically evolved. By analyzing the evolution of the decision transmission path, the change trend and evolution law of the path in the decision process can be observed, which helps decision makers understand the dynamic evolution process of the decision. The decision transmission path evolution data is used to analyze the association between multi-dimensional decision control points and external entities. Entity attribute association data can help decision makers understand the degree of association and attribute relationship between decision control points and external entities, so as to better grasp the external factors and constraints in the decision-making process. Analyze entity attribute association data and construct the association network topology between external entities. External topology data can reveal the association relationship, dependency relationship and network topology characteristics between entities, helping decision makers better understand the structure and organization of external entity systems. Based on external topology data, an external decision topology network is constructed. The external decision topology network can model decision control points, entities and their association relationships in the form of a network, helping decision makers to fully grasp the structure and organization of the decision system, identify key nodes and paths, and conduct decision analysis and optimization.

[0163] In this embodiment, a decision transmission path analysis is performed based on the decision impact range data, which may involve tracing the decision path, analyzing the transmission relationship between decision control points, etc., to determine the transmission path between the decision control points. According to the result of the decision transmission path analysis, decision transmission path data is generated, which describes the transmission path or transmission order between the decision control points. A central path detection is performed on the decision transmission path data, which may involve methods such as the shortest path algorithm and centrality index analysis in graph theory to determine the path with the most important or core position in the decision transmission path. According to the result of the central path detection, decision transmission central path data is generated, which describes the path with an important or core position in the decision transmission path. A transmission efficiency analysis is performed on the decision transmission central path data, which may involve calculating the information transmission efficiency on the path, the speed or accuracy of the decision transmission, etc., to evaluate the efficiency of the decision transmission. According to the decision transmission efficiency data, a dynamic evolution analysis is performed on the decision transmission central path data, which may involve methods such as time series analysis and dynamic network analysis to identify the evolution trend, change pattern or dynamic characteristics of the decision transmission path. According to the result of the dynamic evolution analysis, decision transmission path evolution data is generated, which describes the path with an important or core position in the decision transmission path. The paper describes the changes of decision transmission paths over time, the evolution trend and the dynamic characteristics of decision transmission paths. According to the decision transmission path evolution data, the multi-dimensional decision control point data is subjected to external entity association analysis, which may involve methods such as entity attribute association analysis and correlation analysis to determine the association relationship between decision control points and external entities. According to the results of external entity association analysis, entity attribute association data is generated, which describes the association relationship and association degree between decision control points and external entities. The entity attribute association data is subjected to association network topology analysis, which may involve network graph analysis, node degree centrality analysis, etc. The method is used to identify key nodes, associated subgroups, etc. in an associated network. According to the results of the topological analysis of the associated network, external topological structure data is generated, which describes the nodes, edges and the topological structure between them in the associated network. According to the external topological structure data, a dynamic decision network is constructed, which may involve steps such as network modeling, identification and connection of decision nodes, so as to construct a dynamic decision network reflecting the external associated relationship. According to the results of the dynamic decision network construction, an external decision topological network is generated, which describes the associated relationship between the decision control point and the external entity, and reflects their dynamic characteristics and topological structure.

[0164] In this embodiment, the specific steps of step S6 are:

[0165] Step S61: performing multidimensional feature distribution analysis on the multidimensional decision optimization space to obtain multidimensional feature distribution data;

[0166] Step S62: performing probability distribution calculation on the multi-dimensional feature distribution data to generate multi-dimensional feature probability distribution data;

[0167] Step S63: Perform decision path space analysis on the multi-dimensional feature probability distribution data to generate intrinsic decision path space data;

[0168] Step S64: performing decision path network analysis on the external decision topology network to generate external decision path network data;

[0169] Step S65: reconstructing the internal decision path space data and the external decision path network data to construct a multiple decision space framework;

[0170] Step S66: construct a knowledge graph for the multiple decision space framework and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

[0171] The present invention analyzes the features in the multidimensional decision optimization space to understand the distribution of each feature within different value ranges. Through the multidimensional feature distribution data, statistical information about the features can be obtained to help decision makers understand the distribution range, distribution density and outliers of the feature values. Based on the multidimensional feature distribution data, the probability distribution of each feature within different value ranges is calculated. The multidimensional feature probability distribution data can provide probability information of the feature values ​​to help decision makers understand the probability of occurrence of each feature value, so as to better evaluate the importance and influence of the feature values. The spatial distribution of the decision path is analyzed using the multidimensional feature probability distribution data. The internal decision path space data can provide spatial distribution information about the decision path, helping decision makers understand the distribution of the decision path in the feature space, identify important decision paths and potential decision bottlenecks, and the external decision topology network. The network is analyzed to identify the decision path network. The external decision path network data can provide the connection relationship and transmission path between the decision paths, helping decision makers understand the interaction and influence of different paths in the decision-making process. The internal decision path space data and the external decision path network data are integrated and reconstructed to build a multiple decision space framework. The multiple decision space framework can integrate decision data from different perspectives to form a comprehensive decision space framework, which helps decision makers to comprehensively consider the influence of different factors and levels. Based on the multiple decision space framework, a knowledge graph of transaction decisions is constructed. The transaction decision knowledge graph can integrate and organize the data and information obtained in each step to form a structured knowledge representation form, which can include concepts, attributes, relationships, etc. related to power trading, helping decision makers to better understand and apply the knowledge in the decision space.

[0172] In this embodiment, a multidimensional feature distribution analysis is performed on the multidimensional decision optimization space. This may involve methods such as statistical analysis and data mining to understand the distribution of each decision variable in the optimization space, including its value range, density, correlation, etc. According to the results of the multidimensional feature distribution analysis, multidimensional feature distribution data is generated. The data describes the distribution of the multidimensional decision variables in the optimization space. Probability distribution calculation is performed on the multidimensional feature distribution data. This may involve probability density estimation methods, such as kernel density estimation, parameterized distribution fitting, etc., to calculate the probability distribution of decision variables. According to the results of the probability distribution calculation, multidimensional feature probability distribution data is generated. The data describes the probability distribution of decision variables in the optimization space. Decision path space analysis is performed on the multidimensional feature probability distribution data. This may involve methods such as path search algorithms and optimization algorithms to determine the decision path space under probability distribution. According to the results of the decision path space analysis, internal decision path space data is generated. The data describes the possible decision path space under a given probability distribution. Decision path network analysis is performed on the external decision topology network. This may involve path search algorithms, network analysis methods, etc., to determine the possible decision paths in the external decision topology network. Based on the results of the decision path network analysis, the external decision path network data is generated. This data describes the possible decision paths in the external decision topology network. The internal decision path space data and the external decision path network data are reframed. This can involve methods such as structural analysis and integration to combine the internal decision path space and the external decision path network to construct a unified multiple decision space framework. Based on the results of the framework reconstruction, a multiple decision space framework is constructed. This framework describes the relationship between the internal decision path space and the external decision path network, providing a basis for subsequent decision analysis and optimization.

[0173] In this embodiment, a processing system based on power trading decision data is provided, which is used to execute the power trading method based on multi-dimensional data analysis as described above, including:

[0174] The multidimensional decomposition module is used to obtain real-time power transaction data; perform time series analysis on real-time power transaction data to generate power transaction time series data; perform transaction decision analysis on power transaction time series data to generate transaction time series decision data; perform multidimensional non-uniform pulse decomposition on transaction time series decision data using spline decomposition technology to generate multidimensional decision control point data;

[0175] The decision space module is used to perform spatial position analysis on multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; perform multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point spatial coordinate data to construct a multi-dimensional decision space; perform control point temporal change analysis on the multi-dimensional decision space to generate control point temporal change data;

[0176] An evolution analysis module is used to perform evolution trajectory analysis on the control point time series change data and generate control point evolution trajectory data;

[0177] Perform spatial morphological structure drift analysis on the control point evolution trajectory data to generate the evolution law of decision control points; predict the evolution trend of multidimensional decision space through the evolution law of decision control points to generate multidimensional decision trend prediction data;

[0178] The association mining module is used to optimize the decision structure of the multidimensional decision space using the multidimensional decision trend prediction data to generate the multidimensional decision optimization space; to perform control point association mining on the multidimensional decision control point data to generate control point association data; to perform influence domain coverage calculation on the multidimensional decision space using the control point association data to generate decision influence range data;

[0179] The decision network module is used to perform decision transfer path analysis on the multi-dimensional decision control point data according to the decision impact range data to generate decision transfer path data; perform associated network topology analysis on the decision transfer path data to generate external topology structure data; perform dynamic decision network construction on the external topology structure data to generate an external decision topology network;

[0180] The knowledge graph module is used to perform decision path space analysis on the multi-dimensional decision optimization space to generate internal decision path space data; perform decision path network analysis on the external decision topology network to generate external decision path network data; construct a knowledge graph for the internal decision path space data and the external decision path network data, and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

[0181] The present invention obtains real-time power transaction data through a multidimensional decomposition module, which is crucial for timely decision analysis. By performing time series analysis on real-time power transaction data, time series data of power transactions can be generated, which helps to understand the time correlation and trend of transactions. The decision space module performs spatial position analysis on multidimensional decision control point data to generate spatial coordinate data of multidimensional control points, which helps to determine the position of decision control points in the decision space for subsequent analysis and optimization. Based on the spatial coordinate data of multidimensional control points, multidimensional coordinate mapping is performed to construct a multidimensional decision space, which helps to visualize and analyze multiple dimensions of decision control points in a unified space. Time series change analysis of control points in the multidimensional decision space is performed to generate The time series change data of the control points is generated, which helps to understand the changing trends and patterns of the decision control points over time. The evolution analysis module performs evolution trajectory analysis on the time series change data of the control points and generates evolution trajectory data of the control points, which helps to observe the evolution process and trend of the decision control points in the decision space. By analyzing the evolution trajectory data of the control points, spatial morphological structure drift analysis can be performed, that is, to understand the structural changes of the decision control points in the decision space, which helps to discover the potential patterns and association laws in the decision space. According to the evolution laws of the decision control points, the evolution trend of the multidimensional decision space is predicted, and multidimensional decision trend prediction data is generated, which helps to predict and plan the future changes in the decision space. The association mining module uses the multidimensional decision trend prediction data Optimize the decision structure of the multidimensional decision space and generate a multidimensional decision optimization space, which helps to optimize the structure and performance of the decision space and improve the decision effect. By performing control point association mining on the multidimensional decision control point data, the associated data of the control points can be generated, which helps to discover the association and mutual influence between the decision control points. Using the control point association data, the influence domain coverage calculation of the multidimensional decision space is performed to generate decision influence range data, which helps to determine the degree and scope of influence of the decision control point on the decision space. The decision network module performs associated network topology analysis on the decision transmission path data and generates external topology structure data, which helps to reveal the network relationship and topology structure between the decision control points and further understand the overall nature of the decision system. The external topological structure data is used to construct a dynamic decision network and generate an external decision topological network, which helps to establish a network model of the decision system and provide a dynamic perspective and analysis capability in the decision-making process. The knowledge graph module performs decision path space analysis on the multidimensional decision optimization space and generates internal decision path space data, which helps to understand the characteristics and optimization directions of different paths in the decision space. The decision path network analysis is performed on the external decision topological network to generate external decision path network data, which helps to understand the association and influence relationship between different paths in the decision system. Based on the internal decision path space data and the external decision path network data, a transaction decision knowledge graph is constructed, which helps to organize and integrate decision-related knowledge and information and provide decision support and reference.

[0182] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0183] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0184] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A processing method based on power trading decision data, characterized in that: The following steps are involved: Step S1: Obtaining real-time power transaction data; Perform time series analysis on real-time power transaction data to generate power transaction time series data; Conducting transaction decision analysis on power transaction time series data to generate transaction time series decision data; Use spline decomposition technology to perform multi-dimensional non-uniform pulse decomposition on transaction timing decision data to generate multi-dimensional decision control point data; Step S2: Perform spatial position analysis on the multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; Based on the multi-dimensional control point spatial coordinate data, multi-dimensional coordinate mapping is performed on the multi-dimensional decision control point data to construct a multi-dimensional decision space; Conduct time series change analysis of control points in multi-dimensional decision space to generate time series change data of control points; Step S3: performing evolution trajectory analysis on the control point time series change data to generate control point evolution trajectory data; performing spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law; performing evolution trend prediction on the multidimensional decision space through the decision control point evolution law to generate multidimensional decision trend prediction data; Step S3 includes the following steps: Step S31: Identify edge control points in the multidimensional decision space area and mark edge decision control points; Step S32: performing layout structure analysis on the multidimensional decision space based on the edge decision control points to generate decision space layout structure data; Step S33: performing evolution trajectory analysis on the control point time series change data through the multi-dimensional decision space layout structure data to generate control point evolution trajectory data; Step S34: Perform spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law; the specific steps of step S34 are: Step S341: performing control point distribution entropy quantization processing on the multi-dimensional decision control point data to generate a spatial control point distribution entropy index; Step S342: performing spatial morphological change amplitude analysis on the control point evolution trajectory data according to the spatial control point distribution entropy index to generate morphological change amplitude data; Step S343: performing spatial morphological structure drift analysis on the multidimensional decision space according to the morphological change amplitude data to generate spatial morphological structure drift data; Step S344: Calculating the morphological drift rate of the spatial morphological structure drift data based on the time series window data to generate morphological drift rate data; Step S345: performing control point evolution analysis on the control point evolution trajectory data according to the morphological drift rate data to generate a decision control point evolution law; Step S35: Predicting the evolution trend of the multidimensional decision space according to the evolution law of the decision control points to generate multidimensional decision trend prediction data; Step S4: Optimize the decision structure of the multidimensional decision space using the multidimensional decision trend prediction data to generate a multidimensional decision optimization space; perform control point association mining on the multidimensional decision control point data to generate control point association data; perform influence domain coverage calculation on the multidimensional decision space using the control point association data to generate decision influence range data; Step S5: performing decision transfer path analysis on the multi-dimensional decision control point data according to the decision impact range data to generate decision transfer path data; performing associated network topology analysis on the decision transfer path data to generate external topology structure data; performing dynamic decision network construction on the external topology structure data to generate an external decision topology network; Step S5 includes the following steps: Step S51: performing decision transfer path analysis on the multi-dimensional decision control point data according to the decision influence range data to generate decision transfer path data; Step S52: performing central path detection on the decision transfer path data to generate decision transfer central path data; Step S53: performing transmission efficiency analysis on the decision transmission center path data to generate decision transmission efficiency data; Step S54: performing dynamic evolution analysis on the decision transmission center path data according to the decision transmission efficiency data to generate decision transmission path evolution data; Step S55: performing external entity association analysis on the multi-dimensional decision control point data according to the decision transfer path evolution data to generate entity attribute association data; Step S56: performing a correlation network topology analysis on the entity attribute correlation data to generate external topology structure data; Step S57: constructing a dynamic decision network for the external topology data to generate an external decision topology network; Step S6: Perform decision path space analysis on the multidimensional decision optimization space to generate internal decision path space data; perform decision path network analysis on the external decision topology network to generate external decision path network data; construct a knowledge graph for the internal decision path space data and the external decision path network data, and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

2. The processing method based on power transaction decision data according to claim 1 is characterized in that: The specific steps of step S1 are: Step S11: Acquire real-time power transaction data; Step S12: performing time series analysis on the real-time power transaction data to generate power transaction time series data; Step S13: identifying decision factors for the power transaction time series data and generating decision factor data; Step S14: performing transaction decision analysis on the power transaction time series data according to the decision factor data to generate transaction time series decision data; Step S15: Use spline decomposition technology to perform multi-dimensional non-uniform pulse decomposition on the transaction timing decision data to generate multi-dimensional decision control point data.

3. The processing method based on power transaction decision data according to claim 2 is characterized in that: The specific steps of step S15 are: Step S151: segmenting the transaction timing decision data into timing windows to generate timing window data; Step S152: extracting the transaction timing decision data by using the timing window data to generate a transaction timing decision sequence; Step S153: using the spline decomposition technology to perform multi-dimensional non-uniform pulse decomposition on the transaction timing decision sequence to generate a one-dimensional pulse component data set; Step S154: reconstructing a multi-dimensional spatial pulse path for the one-dimensional pulse component data set to generate a multi-dimensional spatial pulse path; Step S155: performing pulse direction mutation analysis on the multidimensional space pulse path and marking the multidimensional space pulse mutation point; Step S156: extract decision control nodes from the transaction timing decision sequence according to the multi-dimensional space pulse mutation points to generate multi-dimensional decision control point data.

4. The processing method based on power transaction decision data according to claim 1 is characterized in that: The specific steps of step S2 are: Step S21: Perform spatial position analysis on the multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; Step S22: performing multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point space coordinate data to construct a multi-dimensional decision space; Step S23: performing cluster analysis on the multi-dimensional decision control point data to generate decision control point cluster data; Step S24: dividing the multidimensional decision space into cluster regions according to the decision control point cluster data to generate a multidimensional decision space region; Step S25: performing control point time series change analysis on the multi-dimensional decision space region to generate control point time series change data.

5. The processing method based on power transaction decision data according to claim 1 is characterized in that: The specific steps of step S4 are: Step S41: performing optimal solution decision analysis on the multi-dimensional decision trend prediction data to generate configuration optimal solution data; Step S42: optimizing the decision structure of the multidimensional decision space by using the configuration optimal solution data to generate a multidimensional decision optimization space; Step S43: performing control point association mining on the multi-dimensional decision control point data to generate control point association data; Step S44: using the control point association data to perform decision impact intensity analysis on the multidimensional decision space to generate decision impact intensity data; Step S45: Perform influence domain coverage calculation on the decision influence intensity data to generate decision influence range data.

6. The processing method based on power transaction decision data according to claim 1 is characterized in that: The specific steps of step S6 are: Step S61: performing multidimensional feature distribution analysis on the multidimensional decision optimization space to obtain multidimensional feature distribution data; Step S62: performing probability distribution calculation on the multi-dimensional feature distribution data to generate multi-dimensional feature probability distribution data; Step S63: Perform decision path space analysis on the multi-dimensional feature probability distribution data to generate intrinsic decision path space data; Step S64: performing decision path network analysis on the external decision topology network to generate external decision path network data; Step S65: reconstructing the internal decision path space data and the external decision path network data to construct a multiple decision space framework; Step S66: construct a knowledge graph for the multiple decision space framework and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

7. A processing system based on power trading decision data, characterized in that: The method for executing the processing method based on power transaction decision data according to claim 1 comprises: The multidimensional decomposition module is used to obtain real-time power transaction data; perform time series analysis on real-time power transaction data to generate power transaction time series data; perform transaction decision analysis on power transaction time series data to generate transaction time series decision data; perform multidimensional non-uniform pulse decomposition on transaction time series decision data using spline decomposition technology to generate multidimensional decision control point data; The decision space module is used to perform spatial position analysis on multi-dimensional decision control point data to generate multi-dimensional control point spatial coordinate data; perform multi-dimensional coordinate mapping on the multi-dimensional decision control point data based on the multi-dimensional control point spatial coordinate data to construct a multi-dimensional decision space; perform control point temporal change analysis on the multi-dimensional decision space to generate control point temporal change data; The evolution analysis module is used to analyze the evolution trajectory of the control point time series change data to generate the control point evolution trajectory data; perform spatial morphological structure drift analysis on the control point evolution trajectory data to generate the decision control point evolution law; and predict the evolution trend of the multi-dimensional decision space through the decision control point evolution law to generate multi-dimensional decision trend prediction data; The association mining module is used to optimize the decision structure of the multidimensional decision space using the multidimensional decision trend prediction data to generate the multidimensional decision optimization space; to perform control point association mining on the multidimensional decision control point data to generate control point association data; to perform influence domain coverage calculation on the multidimensional decision space using the control point association data to generate decision influence range data; The decision network module is used to perform decision transfer path analysis on the multi-dimensional decision control point data according to the decision impact range data to generate decision transfer path data; perform associated network topology analysis on the decision transfer path data to generate external topology structure data; perform dynamic decision network construction on the external topology structure data to generate an external decision topology network; The knowledge graph module is used to perform decision path space analysis on the multi-dimensional decision optimization space to generate internal decision path space data; perform decision path network analysis on the external decision topology network to generate external decision path network data; construct a knowledge graph for the internal decision path space data and the external decision path network data, and construct a transaction decision knowledge graph to perform power transaction decision data processing operations.

Images