Cross-department government affair big data business co-processing method based on computing power platform and data fusion

Abstract

The invention relates to the technical field of electric power government affair management, and provides a cross-department government affair big data business co-processing method based on a computing power platform and data fusion, comprising the following steps: S1, deploying a heterogeneous computing power cluster, and establishing a computing power resource dynamic scheduling data processing platform comprising a CPU, a GPU, an FPGA and an edge computing node; s2, acquiring multi-dimensional data of power grid equipment data, power grid load data, industry power consumption data and government policy data; and S3, designing a cross-department data access system based on the zero-trust architecture, and developing a multi-dimensional model of a power grid risk prediction model, an industry energy consumption analysis model and a carbon emission accounting model in power management. The space-time joint probability prediction and conflict priority ranking algorithm is applied to power-government affair emergency disposal, intellectualization and precision of cross-department business collaboration are achieved, power faults with serious consequences are rapidly handled according to the priority level, and the harmfulness and unpredictability of the power faults can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power government affairs management, and particularly to a cross-departmental government affairs big data service collaborative processing method based on a computing power platform and data fusion. Background Art

[0002] Under the background of the deep integration of digital government affairs and the energy Internet, cross-departmental service collaboration has become a key requirement for improving public service efficiency and social governance capabilities. As a core component of the country's infrastructure, the operation status of the power system is closely related to the livelihood guarantee and policy implementation of government services. For example, the government affairs platform needs to obtain power consumption data in real time to support people's livelihood services, and the power system needs to optimize resource allocation relying on the policy orientation of the government affairs. However, the existing technologies face the following core challenges in cross-departmental collaboration.

[0003] First of all, traditional government affairs and power systems each deploy independent computing power facilities, and computing resources such as CPUs and GPUs cannot be dynamically allocated on demand, resulting in an imbalance in computing power utilization in high-concurrency scenarios and making it difficult to support the real-time processing requirements of big data. At the same time, there are problems such as format heterogeneity and semantic ambiguity in multi-source data such as power equipment operation data, power grid load data, and government policy documents. Cross-departmental data sharing only stays at the level of basic field exchange, lacking in-depth correlation analysis of equipment status, load characteristics, and policy orientation, forming "data islands" and resulting in the lack of comprehensiveness and timeliness in collaborative decision-making.

[0004] Secondly, existing cross-departmental data access mostly adopts static permission control, which cannot cope with dynamically changing security threats. At the same time, power-government collaborative scenarios rely on manual experience to formulate solutions, lacking data-driven intelligent prediction and optimization models, and it is difficult to prioritize the handling of faults according to the degree of consequences caused by the faults in a timely manner, resulting in great harm and unpredictability of power faults. Summary of the Invention

[0005] Aiming at the problems existing in the prior art, the purpose of the present invention is to provide a cross-departmental government affairs big data service collaborative processing method based on a computing power platform and data fusion to solve the problems proposed in the above background art.

[0006] To achieve the above purpose, the present invention provides a cross-departmental government affairs big data service collaborative processing method based on a computing power platform and data fusion, including the following steps:

[0007] S1. Deploy a heterogeneous computing power cluster and establish a computing power resource dynamic scheduling data processing platform including CPUs, GPUs, FPGAs, and edge computing nodes;

[0008] S2. Obtain multi-dimensional data of power grid equipment data, power grid load data, industrial electricity consumption data, and government policy data, perform semantic annotation on the data, construct a fusion model of power-government data, establish an associated data structure for equipment operation status, power grid load status, industrial energy consumption characteristics, and policy orientation information, and process all the obtained data;

[0009] S3. Design a cross-departmental data access system based on the zero-trust architecture, develop multi-dimensional models for power grid risk prediction models, industrial energy consumption analysis models, and carbon emission accounting models in power management, and design cross-departmental business collaboration content for power distribution planning and emergency response;

[0010] S4. Based on knowledge graphs, historical data, and data analysis models, construct intelligent optimization recommendations for different decision-making scenarios and develop an intelligent decision support system;

[0011] S5. Establish an evaluation index system for business collaboration effectiveness, optimize system parameters, and continuously optimize and analyze the system model.

[0012] Preferably, in step S2, processing all the obtained data includes the following steps:

[0013] S21. Establish an evaluation index system for data integrity, accuracy, and consistency, and perform standardization processing on the obtained data;

[0014] S22. Design ontology semantic annotation rules, automatically map data objects to standard semantic annotations, and construct a semantic annotation platform that supports human-machine collaboration, providing automatic annotation and manual verification functions;

[0015] S23. Automatically identify entities regarding power equipment, power grid load, industrial policies, and policies from unstructured text, perform relationship extraction, and establish a relationship model between entities;

[0016] S24. Fill data attribute values through data mapping and knowledge reasoning, map entities and relationships to a low-dimensional vector space, and set up a query index mechanism;

[0017] S25. Perform time dimension modeling on power grid operation data and industrial electricity consumption data;

[0018] S26. Perform spatial modeling on power grid equipment, industrial locations, and government plans, based on spatio-temporal proximity and spatio-temporal co-occurrence relationships, and mine spatio-temporal association patterns between data;

[0019] S27. Perform spatio-temporal prediction on power grid load, equipment status, industrial electricity consumption, and government plans, and calculate the joint probability of events in the spatio-temporal domain.

[0020] Preferably, in the step S3, the cross-department business collaboration content includes the following steps:

[0021] S31. Classify and encrypt the data, and set zero-trust access control access permissions;

[0022] S32. Collect the historical data, real-time monitoring data, and environmental data of the power grid equipment and the power grid load, and process them to construct a power grid risk prediction model and an industry energy consumption analysis model. The formula is: , where is the power grid risk prediction value, is the observation vector, is the environmental feature vector, and are both feature vectors, is the activation function, is the bias term, is the predicted industry energy consumption value at time is the weight matrix of the fully connected layer, is the attention weight vector, is the element-wise multiplication, is the gated recurrent unit, is the latent space representation vector, is the historical energy consumption sequence at is the bias vector;

[0023] S33. Based on the prediction results of the power grid risk prediction model and the industry energy consumption analysis model, and considering the power grid transmission loss, construct a carbon emission accounting model. The formula is: , where is the total carbon emission of the actual power generation side, is the industry power consumption at time , is the power generation proportion of the th type of energy in the power grid at time , is the th type of energy's carbon intensity, is the power grid transmission loss rate;

[0024] S34. Based on the prediction results of the multi-dimensional model, and combined with the government plan and the industrial plan, detect the conflict between the power grid plan and them, evaluate the conflict, and sort the conflict resolution priorities. The formula is: , where is the priority calculation value, is the influence degree of the conflict in terms of scale, intensity, and time, is the irreplaceability of the power grid plan , is the conflict resolution cost, To prevent a minimum value with a zero denominator, the value is taken as ;

[0025] S35. According to the conflict resolution priority sorting result, construct an emergency response plan for the power grid planning scheme, and conduct cross-departmental collaborative command to solve the problems existing in the power grid dispatching.

[0026] Preferably, in step S31, the access permission control includes the following steps:

[0027] S311. Classify the power-government data according to the sensitivity of the data, divide the security levels, and encrypt the data at different levels.

[0028] S312. Based on the multi-factor identity access system, construct a dynamic authorization model, monitor abnormal access in real time, and calculate the multi-factor access risk value.

[0029] S313. Dynamically adjust the access permission according to the risk assessment result.

[0030] Preferably, in step S32, the processing of the collected data includes the following steps:

[0031] S321. Clean the collected data, process the outliers, missing values and duplicate values in the data, unify the data format and dimension, and divide the processed data into a training set, a prediction set and a validation set.

[0032] S322. Extract the features related to the risk in the data, evaluate the importance of each feature for risk prediction, and combine and transform the original features to generate new features.

[0033] S323. Perform dimensionality reduction processing on the high-dimensional features and dynamically update the feature engineering strategy.

[0034] S324. According to the characteristics of power grid risk prediction, construct a neural network association model between features and risks, and the formula is: , where is the predicted value based on the score position , that is, the value quantifying the correlation degree between features and risks, is the input feature, is the true value, is the variable to be optimized Take the minimum value parameter of the loss function, that is, obtain the optimal score position predicted value through optimization.

[0035] Preferably, in step S35, solving the problems existing in the power grid dispatching includes the following steps:

[0036] S351. Invoke the power grid GIS system, locate the conflict position, confirm the priority sorting for conflict resolution, generate multiple solutions based on the priority sorting for conflict resolution, establish a cost-benefit matrix for solution implementation, evaluate the solutions, and select the optimal solution;

[0037] S352. Based on the selected solution, establish an inter-departmental joint command center, assign power grid departments, government departments, and external experts according to the solution, establish emergency repair, material allocation, and backup management plans, and complete information sharing and collaborative command;

[0038] S353. Based on the carbon emission accounting model, perform real-time updated accounting of carbon emissions, and allocate carbon quotas according to the government's carbon emission reduction targets and actual situations;

[0039] S354. Regularly evaluate the carbon emission reduction effects of the industry, verify the effectiveness of the solution, formulate policy incentives or restraint measures according to the carbon emission reduction situation of the industry, and conduct data sharing and publicity to improve transparency.

[0040] Preferably, in step S4, the construction of the knowledge graph includes the following steps:

[0041] S41. Collect and label training data for entity recognition, construct a training data set, establish a training model in the power-government field, and fine-tune the data;

[0042] S42. Define the relationship types between entities in the power-government field, construct a training data set for relationship extraction, and establish a relationship extraction model. , where is the feature of the head entity and the feature of the tail entity When, the predicted relationship type is The probability distribution of, is the weight matrix of the relationship classification layer, is the bias vector, is the concatenation symbol;

[0043] S43. Design the entity and relationship attribute system in the power-government field, extract the attribute values of the entities and relationships, unify the data format of the extracted attribute values, process outliers, and fill the processed attribute values into the knowledge graph, and establish a dynamic update mechanism for attribute values to ensure the timeliness of attribute values;

[0044] S44. Use test data to evaluate the performance of the entity recognition model, perform relationship reasoning based on the existing relationships, complete the missing relationships in the knowledge graph, and conduct manual verification.

[0045] Preferably, in step S5, the continuous optimization and analysis of the system model include the following steps:

[0046] S51. Clearly define the power-government collaboration scenarios and core objectives, divide them into four dimensions of data, process, service, and resources, initially select indicators, and then form an evaluation system through screening and weight assignment.

[0047] S52. Classify and formulate data interaction and process scheduling parameter strategies, design an AB test plan for pilot verification, and establish a risk control mechanism.

[0048] S53. Continuously optimize the model through sensitivity analysis, develop an intelligent platform to solidify capabilities, and provide supporting training and institutional guarantees for long-term operation.

[0049] A cross-departmental government affairs big data business collaborative processing system based on the integration of computing power platform and data, which is applied to the cross-departmental government affairs big data business collaborative processing method described in any one of the above, includes:

[0050] A heterogeneous computing power scheduling module, which is used to integrate CPUs, GPUs, FPGAs, and edge computing nodes to build a dynamic scheduling platform.

[0051] A multi-dimensional data fusion module, which is used to converge power grid equipment, load, industrial electricity consumption, and government policy data, and construct a cross-domain data fusion structure through semantic annotation and correlation modeling.

[0052] A security collaborative modeling module, which designs a cross-departmental data security access system based on the zero-trust architecture, develops multi-dimensional models for power grid risk prediction, energy consumption analysis, and carbon emission accounting, and plans the cross-departmental collaborative business logic for power distribution and emergency response.

[0053] An intelligent decision-making support module, which combines knowledge graphs, historical data, and analysis models to construct intelligent optimization recommendation solutions for different scenarios.

[0054] An efficiency evaluation and optimization module, which is used to establish an evaluation index system for business collaboration efficiency, optimize system parameters through data monitoring and analysis, and continuously iterate and improve the model.

[0055] The beneficial effects of the cross-departmental government affairs big data business collaborative processing method provided by the present invention are:

[0056] 1. By integrating CPUs, GPUs, FPGAs, and edge computing nodes to build a dynamic computing power platform, and introducing semantic annotation, spatio-temporal modeling, and knowledge graph technologies, it realizes the deep association and structured processing of multi-source data such as power equipment, load, and policies, breaks through the traditional cross-departmental data island problem, and improves the efficiency of heterogeneous data fusion and analysis accuracy.

[0057] 2. Design a cross-departmental data access control mechanism based on the zero-trust architecture, combine multi-dimensional models such as power grid risk prediction, energy consumption analysis, and carbon emission accounting, build a closed-loop of "data security - business collaboration - intelligent decision-making", apply the spatio-temporal joint probability prediction and conflict priority ranking algorithm to power - government emergency response, realize the intelligence and precision of cross-departmental business collaboration, and in the event of a power network failure, be able to quickly generate a collaborative processing plan for multiple departments and quickly handle power failures with serious consequences according to the priority level, which is conducive to reducing the harmfulness and unpredictability of power failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0059] Figure 1 It is a schematic flow chart of the steps of a cross-departmental government affairs big data business collaboration processing method based on a computing power platform and data fusion provided by the present application;

[0060] Figure 2 It is a schematic diagram of the system modules of a cross-departmental government affairs big data business collaboration processing method based on a computing power platform and data fusion provided by the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] The following will further describe in detail the specific embodiments of the present invention in conjunction with the drawings in the specification and the embodiments. The following embodiments are only used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0062] As Figure 1 - Figure 2 shown, this embodiment proposes a cross-departmental government affairs big data business collaboration processing method based on a computing power platform and data fusion, including the following steps:

[0063] S1. Deploy a heterogeneous computing power cluster and establish a data processing platform for dynamic scheduling of computing power resources including CPUs, GPUs, FPGAs, and edge computing nodes;

[0064] S2. Obtain multi-dimensional data of power grid equipment data, power grid load data, industry electricity consumption data, and government policy data, perform semantic annotation on the data, build a fusion model of power - government data, establish an associated data structure of equipment operation status, power grid load status, industry energy consumption characteristics, and policy orientation information, and process all the obtained data;

[0065] S3. Design a cross-departmental data access system based on the zero-trust architecture, develop a multi-dimensional model for power grid risk prediction, industry energy consumption analysis, and carbon emission accounting models in power management, and design cross-departmental business collaboration content for power distribution planning and emergency response.

[0066] S4. Based on knowledge graphs, historical data, and data analysis models, build intelligent optimization recommendations for different decision-making scenarios and develop an intelligent decision support system.

[0067] S5. Establish an evaluation index system for business collaboration effectiveness, optimize system parameters, and continuously optimize and analyze the system model.

[0068] The present invention constructs a dynamic computing power platform by integrating CPUs, GPUs, FPGAs, and edge computing nodes, and introduces semantic annotation, spatio-temporal modeling, and knowledge graph technologies to achieve deep association and structured processing of multi-source data such as power equipment, loads, and policies, break through the traditional cross-departmental data island problem, and improve the efficiency of heterogeneous data fusion and analysis accuracy.

[0069] In this embodiment, in step S2, processing all the acquired data includes the following steps:

[0070] S21. Establish an evaluation index system for data integrity, accuracy, and consistency, and perform standardization processing on the acquired data.

[0071] S22. Design ontology semantic annotation rules, automatically map data objects to standard semantic annotations, and construct a semantic annotation platform that supports human-computer collaboration, providing automatic annotation and manual verification functions.

[0072] S23. Automatically identify entities related to power equipment, grid loads, industry policies, and regulations from unstructured text, perform relationship extraction, and establish a relationship model between entities. The formula is: , where is the label sequence predicted by the model, is selected from all possible label sequence sets to be the sequence that maximizes the subsequent probability product, is the number of all label sequences, is at the th position of the feature and the previous label under the condition that the current label is the conditional probability;

[0073] S24. Fill data attribute values through data mapping and knowledge reasoning, map entities and relationships to a low-dimensional vector space, and set up a query index mechanism.

[0074] S25. Perform time - dimension modeling on power grid operation data and industrial electricity consumption data. The formula is: , where in the formula, is the observed value of the data variable at time , is the intercept term, is the number of data variables, is the th smooth function corresponding to the independent variable, is the th independent variable's observed value at time , is the random error term at time ;

[0075] S26. Perform spatial modeling on power grid equipment, industrial locations, and government plans. The formula is: , where in the formula, and are the feature vectors of node in the th and th layers of the neural network respectively, is the spatial neighbor set of node , is the variable belonging to the spatial neighbor set , and are the weight matrix and bias term of the th layer respectively, is the activation function. Based on spatio - temporal proximity and spatio - temporal co - occurrence relationships, and mining spatio - temporal association patterns among data, the formula is: , where in the formula, is the predicted value of spatial location at time , , and are the spatial, time latent factor matrices, and the spatial - latent factor correlation tensor respectively, , and are the spatial location attributes and spatial characteristics of time - regular government plans respectively;

[0076] S27. Perform spatio - temporal prediction on power grid load, equipment status, industrial electricity consumption, and government plans, and calculate the joint probability of events in the spatio - temporal domain. The formula is: , where in the formula, is the joint probability value, is the spatial region attribute, is the time window attribute.

[0077] Specifically, by establishing a data quality assessment system and a semantic annotation platform, standardizing the power-government data and modeling the entity relationships, and combining with the mathematical modeling of the spatio-temporal dimension, the spatio-temporal correlation patterns and joint probability distributions among the data are mined to provide a highly consistent and highly available structured data foundation for subsequent models such as power grid risk prediction and industry energy consumption analysis, ensuring the accuracy and timeliness of cross-departmental collaborative analysis.

[0078] In this embodiment, in the step S3, the cross-departmental business collaboration content includes the following steps:

[0079] S31. Classify and encrypt the data, and set zero-trust access control access permissions;

[0080] S32. Collect the historical data, real-time monitoring data, and environmental data of the power grid equipment and the power grid load and process them to construct a power grid risk prediction model and an industry energy consumption analysis model. The formula is: , where is the power grid risk prediction value, is the observation vector, is the environmental feature vector, and are both feature vectors, is the activation function, is the bias term, is the predicted industry energy consumption value at time the weight matrix of the fully connected layer, is the attention weight vector, is the element-wise multiplication, is the gated recurrent unit, is the latent space representation vector, is the historical energy consumption sequence at time steps,

[0081] S33. Based on the prediction results of the power grid risk prediction model and the industry energy consumption analysis model, and considering the power grid transmission loss, construct a carbon emission accounting model. The formula is: , where is the total actual carbon emission on the power generation side, is the industry electricity consumption at time , is the power generation proportion of the th type of energy by the power grid at time , is the th type of energy carbon intensity, is the power grid transmission loss rate;

[0082] S34. Based on the prediction results of the multi-dimensional model and combined with government plans and industrial plans, detect conflicts between the power grid plan and them, evaluate the conflicts, sort the priorities for conflict resolution, and the formula is: , where is the priority calculation value, is the degree of influence of the scale, intensity, and time involved in the conflict, is the power grid planning scheme of irreplaceability, is the cost of conflict resolution, is a minimum value to prevent the denominator from being zero, and the value is ;

[0083] S35. According to the sorted results of the priorities for conflict resolution, construct an emergency response plan for the power grid planning scheme, and conduct cross-departmental collaborative command to solve the problems existing in power grid dispatching.

[0084] Specifically, based on the zero-trust architecture, classify and encrypt data at different levels and dynamically control access rights. Through the power grid risk prediction model, industry energy consumption analysis model, and carbon emission accounting model, realize real-time monitoring of the power grid operation status and cross-departmental collaborative planning. At the same time, through the conflict detection and priority sorting algorithm, ensure the consistency between the power grid plan and government policies, and improve the scientific nature and collaborative efficiency of emergency response.

[0085] In this embodiment, in the step S31, the access right control includes the following steps:

[0086] S311. Classify the power - government data according to the sensitivity of the data, divide the security levels, and encrypt the data at different levels;

[0087] S312. Based on the multi-factor identity access system, construct a dynamic authorization model, and monitor abnormal access in real time, calculate the multi-factor access risk value, and the formula is: , where is the comprehensive risk value, and are respectively the weight coefficient and risk score of the th type of factor, and are respectively the penalty coefficient and boolean value of the th type of risk rule, and are respectively the number of factors and risk rules;

[0088] S313. Dynamically adjust the access rights according to the risk assessment results, and the formula is: , where is the overall environmental security credibility, is the total number of environmental factors, is the normalized safety value of the th environmental factor.

[0089] Specifically, by classifying and encrypting the sensitivity of power-government data at different levels, combining multi-factor authentication and dynamic authorization models, calculating the access risk value in real time and adjusting permissions, a security protection chain of "data classification - risk assessment - dynamic permission adjustment" is formed to effectively prevent risks such as unauthorized access and data leakage in cross-departmental data sharing, ensuring the compliant use and transmission security of sensitive data in collaborative scenarios.

[0090] In this embodiment, in step S35, solving the problems existing in power grid dispatching includes the following steps:

[0091] S351. Invoke the power grid GIS system to locate the conflict location, confirm the priority order for conflict resolution, generate multiple solutions based on the priority order for conflict resolution, establish a cost-benefit matrix for solution implementation, evaluate the solutions, and select the optimal solution;

[0092] S352. Based on the selected solution, establish a cross-departmental joint command center, assign power grid departments, government departments, and external experts according to the solution, and form emergency repair, material allocation, and backup management plans to complete information sharing and collaborative command;

[0093] S353. Based on the carbon emission accounting model, update and account the carbon emissions in real time, and allocate carbon quotas according to the government's carbon emission reduction targets and actual situations. The formula is: , where is the annual carbon quota for the th industry, is the annual output value of the industry, is the average carbon emission intensity in the three years before the adjustment of the industry, is the adjustment coefficient, and , is the annual emission reduction rate target in the same baseline emission;

[0094] S354. Regularly evaluate the carbon emission reduction effect of the industry, verify the effectiveness of the solution, formulate policy incentives or restraint measures according to the carbon emission reduction situation of the industry, and conduct data sharing and publicity to improve transparency.

[0095] Specifically, locate the conflict location through the power grid GIS system and generate multiple sets of solutions, screen the optimal strategy based on the cost-benefit matrix, rely on the cross-departmental joint command center to achieve resource allocation and information sharing, and at the same time use the carbon emission accounting model to dynamically update the carbon quota, combining power grid dispatching with the government's carbon emission reduction target, which not only ensures the stability of power grid operation, but also promotes the green transformation of the industry through data sharing and policy incentives, enhancing the environmental benefits and social transparency of cross-departmental collaboration.

[0096] In this embodiment, in step S4, the construction of the knowledge graph includes the following steps:

[0097] S41. Collect and annotate the training data for entity recognition, construct the training data set, and establish a training model for the power-government field to fine-tune the data. The formula is: , where in the formula is the adjusted parameter value, and are the number of samples and the number of categories of government policies or power failures respectively, and are the true label and the predicted logits value of the sample belonging to the category respectively, is the exponentiation of the predicted score, is the regularization parameter, is the model parameter is the square of the norm of

[0098] S42. Define the relationship types between entities in the power-government field, construct the training data set for relation extraction, and establish a relation extraction model. , where in the formula is the probability distribution of predicting the relation type as when the head entity feature and the tail entity feature are given, is the weight matrix of the relation classification layer, is the bias vector, is the concatenation symbol;

[0099] S43. Design the entity and relation attribute systems in the power-government field, extract the attribute values of the entities and relations, unify the data formats of the extracted attribute values, process the outliers, and fill the processed attribute values into the knowledge graph, and establish a dynamic update mechanism for the attribute values to ensure the timeliness of the attribute values;

[0100] S44. Use the test data to evaluate the performance of the entity recognition model, perform relation reasoning based on the existing relations, complete the missing relations in the knowledge graph, and conduct manual verification.

[0101] Specifically, through the entity recognition model training and relation extraction algorithm, construct the knowledge graph in the power-government field, integrate the multi-dimensional entity relations such as equipment, policies, and energy consumption, and combine the dynamic update of attribute values and the mechanism for reasoning and complementing missing relations to form a structured knowledge network, providing domain knowledge support for the intelligent decision-making support system, enabling the system to generate optimized recommendation solutions for different scenarios based on historical data and graph association relations, and improving the accuracy and interpretability of decisions.

[0102] In this embodiment, in the step S32, the processing of the collected data includes the following steps:

[0103] S321. Clean the collected data, process the outliers, missing values and duplicate values in the data, unify the data format and dimension, and divide the processed data into a training set, a prediction set and a validation set;

[0104] S322. Extract the features related to risks in the data, evaluate the importance degree of each feature for risk prediction, and combine and transform the original features to generate new features;

[0105] S323. Perform dimensionality reduction processing on the high-dimensional features and dynamically update the feature engineering strategy;

[0106] S324. According to the characteristics of power grid risk prediction, construct a neural network association model between features and risks, and the formula is: , where is the predicted value based on the score position , that is, the value quantifying the correlation degree between features and risks, is the input feature, is the true value, is the variable to be optimized Take the minimum value of the loss function for the parameter, that is, obtain the optimal score position predicted value through optimization.

[0107] Specifically, by cleaning the collected data, performing feature engineering processing and neural network modeling, the accuracy and robustness of the power grid risk prediction and the industry energy consumption analysis model are improved. Data cleaning ensures the data quality input into the model. Feature engineering mines the key influencing factors and reduces dimensional redundancy. The neural network model realizes the dynamic prediction of the power grid operation state and the analysis of the energy consumption trend by quantifying the correlation degree between features and risks, provides a scientific basis for cross-departmental collaborative formulation of power distribution plans and emergency response strategies, and enhances the foresight and accuracy of power management.

[0108] In this embodiment, in the step S5, the continuous optimization and analysis of the system model includes the following steps:

[0109] S51. Define the power-government collaborative scenario and the core objectives, divide the four dimensions of data, process, service and resource, and form an evaluation system after initial selection of indicators, screening and weight assignment;

[0110] S52. Classify and formulate data interaction and process scheduling parameter strategies, design an AB test plan for pilot verification, and establish a risk control mechanism;

[0111] S53. Continuously optimize the model through sensitivity analysis, develop intelligent platform solidification capabilities, and provide supporting training and systems to ensure long-term operation.

[0112] Specifically, by establishing a four-dimensional evaluation system covering data, processes, services, and resources, quantifying collaborative efficiency and identifying bottleneck indicators, and optimizing system parameters through AB testing and sensitivity analysis, and relying on intelligent platforms to solidify optimization strategies, and supporting training and assessment systems to ensure long-term operations, a closed-loop management mechanism of "evaluation-optimization-verification-iteration" is formed to continuously improve the performance and user experience of the power-government collaborative system.

[0113] A cross-departmental government big data business collaborative processing system based on computing power platform and data fusion, applied to any one of the above-mentioned cross-departmental government big data business collaborative processing methods based on computing power platform and data fusion, comprising:

[0114] Heterogeneous computing power scheduling module, used to integrate CPU, GPU, FPGA and edge computing nodes to build a dynamic scheduling platform;

[0115] Multi-dimensional data fusion module, which is used to aggregate data on power grid equipment, loads, industry electricity consumption and government policies, and build a cross-domain data fusion structure through semantic annotation and association modeling;

[0116] The security collaborative modeling module designs a cross-departmental data security access system based on a zero-trust architecture, develops multi-dimensional models for grid risk prediction, energy consumption analysis, and carbon emission accounting, and plans cross-departmental collaborative business logic for power distribution and emergency response;

[0117] The intelligent decision support module combines knowledge graphs, historical data and analytical models to build intelligent optimization recommendation solutions for different scenarios;

[0118] The performance evaluation and optimization module is used to establish a business collaboration performance evaluation indicator system, optimize system parameters through data monitoring and analysis, and continuously iterate and improve the model.

[0119] The present invention designs a cross-departmental data access control mechanism based on a zero-trust architecture, combines multi-dimensional models such as power grid risk prediction, energy consumption analysis, and carbon emission accounting, constructs a "data security-business collaboration-intelligent decision-making" closed loop, and applies the spatiotemporal joint probability prediction and conflict priority sorting algorithm to power-government emergency response, thereby realizing the intelligent and precise cross-departmental business collaboration. In the event of a power network failure, it can quickly generate a collaborative processing plan for multiple departments, and quickly handle power failures with serious consequences according to priority levels, which is conducive to reducing the harmfulness and unpredictability of power failures.

[0120] The above embodiments are only used to illustrate the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the embodiments, those of ordinary skill in the art should understand that various combinations, modifications or equivalent substitutions of the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention and should all be covered within the scope of the claims of the present invention.

Claims

1. A cross-departmental government affairs big data business collaborative processing method based on the integration of computing power platform and data, characterized in that, It includes the following steps: S1. Deploy a heterogeneous computing power cluster and establish a dynamic scheduling data processing platform for computing power resources including CPUs, GPUs, FPGAs, and edge computing nodes; S2. Obtain multi-dimensional data of power grid equipment data, power grid load data, industrial electricity consumption data, and government policy data, perform semantic annotation on the data, construct a fusion model of power - government data, establish an associated data structure of equipment operation status, power grid load status, industrial energy consumption characteristics, and policy orientation information, and process all the obtained data; S3. Design a cross - department data access system based on the zero - trust architecture, develop multi - dimensional models of power grid risk prediction models, industrial energy consumption analysis models, and carbon emission accounting models in power management, and design power distribution planning and cross - department business collaboration content for emergency response; S4. Based on knowledge graphs, historical data, and data analysis models, construct intelligent optimization recommendations for different decision - making scenarios and develop an intelligent decision - making support system; S5. Establish an evaluation index system for business collaboration effectiveness, optimize system parameters, and continuously optimize and analyze the system model.

2. A cross-departmental government affairs big data service collaborative processing method based on computing power platform and data fusion according to claim 1, characterized in that, In the step S2, processing all the obtained data includes the following steps: S21. Establish an evaluation index system for data integrity, accuracy, and consistency, and perform standardization processing on the obtained data; S22. Design an ontology - based semantic annotation rule, automatically map data objects to standard semantic annotations, and construct a semantic annotation platform that supports human - machine collaboration, providing automatic annotation and manual verification functions; S23. Automatically identify entities related to power equipment, power grid load, industry policies, and policies from unstructured text, perform relationship extraction, and establish a relationship model between entities; S24. Fill data attribute values through data mapping and knowledge reasoning, map entities and relationships to a low - dimensional vector space, and set up a query index mechanism; S25. Perform time - dimension modeling on power grid operation data and industrial electricity consumption data; S26. Perform spatial modeling on power grid equipment, industrial locations, and government plans, based on spatio - temporal proximity and spatio - temporal co - occurrence relationships, and mine spatio - temporal association patterns between data; S27. Perform spatio - temporal prediction on power grid load, equipment status, industrial electricity consumption, and government plans, and calculate the joint probability of events in the spatio - temporal domain.

3. A cross-departmental government big data service collaborative processing method based on computing power platform and data fusion according to claim 1, characterized in that, In the step S3, the cross - department business collaboration content includes the following steps: S31. Classify and encrypt data at different levels, and set zero - trust access control access permissions; S32. Collect and process the historical data, real-time monitoring data, and environmental data of grid equipment and grid loads, and construct a grid risk prediction model and an industry energy consumption analysis model. The formula is as follows: , where is the grid risk prediction value, is the observation vector, is the environmental feature vector, and are both feature vectors, is the activation function, is the bias term, is the predicted industry energy consumption value at time is the weight matrix of the fully connected layer, is the attention weight vector, is the element-wise multiplication, is the gated recurrent unit, is the latent space representation vector, is the historical energy consumption sequence at time steps, and is the bias vector; S33. Based on the prediction results of the power grid risk prediction model and the industry energy consumption analysis model, and considering the power transmission losses of the power grid, a carbon emission accounting model is constructed. The formula is: , where is the total carbon emissions on the actual power generation side, is the time industry electricity consumption, is the time the power grid's power generation proportion from the type of energy, is the type of energy's carbon intensity, is the power grid power transmission loss rate; S34. Based on the prediction results of the multi-dimensional model, combined with government plans and industrial plans, detect conflicts between the power grid plan and them, evaluate the conflicts, and rank the priority of conflict resolution. The formula is: , where is the priority calculation value, is the degree of influence of the scale, intensity, and time involved in the conflict, is the power grid planning scheme of irreplaceability, is the conflict resolution cost, is a minimum value to prevent the denominator from being zero, and the value is ; S35. According to the conflict resolution priority ranking result, construct an emergency handling plan for the power grid planning scheme, and conduct cross - department collaborative command to solve problems existing in power grid scheduling.

4. A cross-departmental government affairs big data service collaborative processing method based on computing power platform and data fusion according to claim 3, characterized in that, In the step S31, the access permission control includes the following steps: S311. Classify power - government data according to the sensitivity of the data, divide security levels, and encrypt the data at different levels; S312. Based on a multi - factor identity access system, construct a dynamic authorization model, monitor abnormal access in real - time, and calculate the multi - factor access risk value; S313. Dynamically adjust access permissions according to the risk assessment results.

5. A cross-departmental government affairs big data service collaborative processing method based on computing power platform and data fusion according to claim 3, characterized in that In the step S32, processing the collected data includes the following steps: S321. Clean the collected data, process outliers, missing values and duplicate values in the data, unify the data format and dimension, and divide the processed data into a training set, a prediction set and a validation set; S322. Extract the features related to risks in the data, evaluate the importance of each feature for risk prediction, and combine and transform the original features to generate new features; S323. Perform dimensionality reduction on high-dimensional features and dynamically update the feature engineering strategy; S324. According to the characteristics of power grid risk prediction, a neural network correlation model of features and risks is constructed, and the formula is: , where is the predicted value based on the fractional bit , that is, the quantization feature and risk correlation degree value, is the input feature, is the true value, is the parameter value for minimizing the loss function for the variable to be optimized , that is, the optimal fractional bit predicted value is obtained through optimization.

6. A cross-departmental government affairs big data service collaborative processing method based on computing power platform and data fusion according to claim 3, characterized in that, In the step S35, solving the problems existing in power grid dispatching includes the following steps: S351. Invoke the power grid GIS system, locate the conflict positions, confirm the priority order for conflict resolution, generate multiple solutions according to the priority order for conflict resolution, establish a cost-benefit matrix for solution implementation, evaluate the solutions, and select the optimal solution; S352. Establish a cross-departmental joint command center according to the selected solution, assign power grid departments, government departments and external experts according to the solution, establish repair, material allocation and backup management plans, and complete information sharing and coordinated command; S353. Based on the carbon emission accounting model, perform real-time updated accounting of carbon emissions, and allocate carbon quotas according to the government's carbon emission reduction targets and actual situations; S354. Regularly evaluate the carbon emission reduction effect of the industry, verify the effectiveness of the solution, formulate policy incentives or restraint measures according to the carbon emission reduction situation of the industry, and conduct data sharing and publicity to improve transparency.

7. A cross-departmental government affairs big data service collaborative processing method based on computing power platform and data fusion according to claim 1, characterized in that: In the step S4, the construction of the knowledge graph includes the following steps: S41. Collect and label the training data for entity recognition, construct a training data set, and establish a training model in the power-e-government field to fine-tune the data; S42. Define the relationship types between entities in the power-government domain, construct a training dataset for relation extraction, and establish a relation extraction model. , where is the head entity feature and the tail entity feature , when predicting the relationship type as the probability distribution of is the weight matrix of the relation classification layer, is the bias vector, is the concatenation symbol; S43. Design the entity and relationship attribute system in the power-e-government field, extract the attribute values of entities and relationships, unify the data format of the extracted attribute values, process outliers, and fill the processed attribute values into the knowledge graph, and establish a dynamic update mechanism for attribute values to ensure the timeliness of attribute values; S44. Use the test data to evaluate the performance of the entity recognition model, perform relationship reasoning based on the existing relationships, complete the missing relationships in the knowledge graph, and conduct manual verification.

8. A cross-departmental government affairs big data service collaborative processing method based on computing power platform and data fusion according to claim 1, characterized in that In the step S5, the continuous optimization and analysis of the system model includes the following steps: S51. Define the power-e-government collaboration scenarios and core objectives, divide the four dimensions of data, process, service and resources, initially select indicators and then form an evaluation system through screening and weight assignment; S52. Classify and formulate data interaction and process scheduling parameter strategies, design an AB test plan for pilot verification, and establish a risk control mechanism; S53. Continuously optimize the model through sensitivity analysis, develop an intelligent platform to solidify the capabilities, and provide supporting training and institutional guarantees for long-term operation.

9. A cross-departmental government affairs big data service collaborative processing system based on the integration of computing power platform and data, which is applied to a cross-departmental government affairs big data service collaborative processing method described in any one of claims 1-8, and is characterized in that, Including: A heterogeneous computing power scheduling module for integrating CPUs, GPUs, FPGAs and edge computing nodes to build a dynamic scheduling platform; A multi-dimensional data fusion module for aggregating power grid equipment, load, industrial electricity consumption and government policy data, and constructing a cross-domain data fusion structure through semantic annotation and association modeling; The secure collaborative modeling module designs a cross-departmental data security access system based on the zero-trust architecture, develops multi-dimensional models for power grid risk prediction, energy consumption analysis, and carbon emission accounting, and plans the cross-departmental collaborative business logic for power distribution and emergency response; The intelligent decision support module constructs intelligent optimization recommendation solutions for different scenarios by combining knowledge graphs, historical data, and analysis models; The efficiency evaluation and optimization module is used to establish an evaluation index system for business collaboration efficiency, optimize system parameters through data monitoring and analysis, and continuously iterate and improve the model.

Images