A multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises

Abstract

This invention relates to the field of safety monitoring and emergency command technology for power generation enterprises, specifically a multi-source heterogeneous data fusion safety monitoring and emergency command system for power generation enterprises. The system includes a data aggregation and fusion unit, an intelligent monitoring and early warning unit, an emergency command and decision-making unit, a simulation exercise and evaluation unit, and a visualization unit. This invention achieves real-time generation and iterative optimization of emergency response plans through a dynamic intelligent simulation engine. Combined with cross-departmental collaborative scheduling and process tracking, it significantly improves the efficiency and accuracy of emergency response. Furthermore, based on digital twins, it constructs high-fidelity dynamically coupled accident scenarios, conducts human-machine collaborative adaptive exercises, and implements multi-dimensional quantitative evaluation, enabling continuous verification and iterative optimization of emergency plans. This significantly enhances the emergency response capabilities and safety risk prevention and control level of power generation enterprises.

Description

Technical Field

[0001] This invention relates to the field of safety monitoring and emergency command technology for power generation enterprises, specifically a multi-source heterogeneous data fusion safety monitoring and emergency command system for power generation enterprises. Background Technology

[0002] Power generation enterprises are industrial enterprises whose core business is the production and supply of electricity. They convert primary energy sources such as coal, hydropower, wind power, solar power, and nuclear power into electricity and then transmit this electricity to the power grid or supply it directly to users. They are a core component of the power industry system. Because their production processes involve the operation of equipment under high temperature, high pressure, and high load, and because the energy conversion chain is complex, any sudden events such as equipment failure, cyberattacks, or safety accidents can not only cause power generation interruptions and economic losses, but may also trigger chain reactions, threatening the stability of the regional power grid and public safety. Therefore, emergency management capability is a core indicator for measuring the safe operation level of power generation enterprises.

[0003] However, in the existing technology, the emergency management model of traditional power generation enterprises still has many pain points that urgently need to be addressed: emergency response plans are mostly static documents prepared in advance, which are difficult to match the complex and ever-changing scenarios of emergencies; the risk identification and disposal instruction transmission links are long, resulting in delayed response; emergency drills are mostly based on preset scripts, and the scenario simulation is out of touch with actual production conditions, greatly reducing the effectiveness of the drills; and the plans and disposal processes lack effective verification and optimization mechanisms, making it impossible for enterprises to achieve closed-loop iterative improvement of their emergency response capabilities.

[0004] Based on this, the present invention provides a multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises. This invention realizes the real-time generation and iterative optimization of emergency response plans through a dynamic intelligent simulation engine. Combined with cross-departmental collaborative scheduling and process tracking, it significantly improves the efficiency and accuracy of emergency response. Furthermore, based on digital twins, it constructs high-fidelity dynamically coupled accident scenarios, conducts human-machine collaborative adaptive drills, and implements multi-dimensional quantitative assessments to achieve continuous verification and iterative optimization of emergency plans, thereby significantly improving the emergency response capabilities and safety risk prevention and control level of power generation enterprises.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises, comprising a data aggregation and fusion unit, an intelligent monitoring and early warning unit, an emergency command and decision-making unit, a simulation exercise and evaluation unit, and a visualization display unit, wherein: The data aggregation and fusion unit is used to collect multi-source heterogeneous data of different types and protocols within the power generation enterprise, preprocess and correlate the collected data, and generate a unified multi-source fused data resource. The intelligent monitoring and early warning unit: Based on the fused data, it monitors the multi-dimensional security status of power generation enterprises' production and operation, network security, and physical security in real time, identifies abnormal risks, and issues early warning prompts; The emergency command and decision-making unit is used to generate and optimize emergency response plans in real time through a dynamic intelligent simulation engine based on multi-source fusion data, early warning information and contingency plans when an emergency occurs, and to execute adaptive cross-departmental collaborative command and dispatch. The simulation exercise and evaluation unit: Based on multi-source fusion data, digital twin and dynamic intelligent inference technology, it constructs dynamically coupled accident scenarios, carries out human-machine collaborative adaptive emergency drills, and drives emergency plans through multi-dimensional quantitative evaluation; The visualization display unit is used to present the enterprise's security status, data fusion results, early warning information, emergency response progress, and simulation exercise and evaluation results in a visual format.

[0007] The data aggregation and fusion unit includes a multi-source data access module, a data preprocessing module, and a correlation and fusion module, wherein: The multi-source data access module is used to adapt to heterogeneous data interfaces of different types and protocols of power generation enterprises, and to uniformly collect and access various types of data in production, security and management. The data preprocessing module is used to clean, convert, remove outliers, and filter redundant data from the collected raw data. The association and fusion module is used to construct a multi-source data association rule base, perform deep association and fusion of cross-type and cross-source data, and generate unified multi-source fused data resources.

[0008] The intelligent monitoring and early warning unit includes a multi-dimensional status monitoring module, an abnormal risk identification module, and a hierarchical early warning module, wherein: The multi-dimensional status monitoring module: Based on fused data, it performs real-time monitoring and status awareness of key indicators of power generation enterprise production operation, network security, and physical security. The abnormal risk identification module is used to identify explicit and implicit abnormal risks in multi-dimensional monitoring data by using preset safety thresholds and intelligent detection models. The graded early warning module is used to generate and push graded and categorized early warning information based on the risk level and the scope of impact.

[0009] The emergency command and decision-making unit includes a dynamic intelligent simulation engine module, an emergency plan management module, a cross-departmental collaborative dispatch module, and a response process tracking module, wherein: The dynamic intelligent simulation engine module is used to integrate multi-source fusion data and early warning information, and to generate and dynamically optimize emergency response plans in real time through intelligent algorithms. The emergency response plan management module is used to store, retrieve, and update various emergency response plans. The cross-departmental collaborative dispatch module is used to share information and issue instructions across departments and levels during emergency command, supporting adaptive collaborative command. The process tracking module is used to track the execution progress of the emergency response plan in real time, provide feedback on the response effect, and support the immediate adjustment of the plan.

[0010] The dynamic intelligent simulation engine module integrates multi-source fusion data and early warning information, and uses intelligent algorithms to generate and dynamically optimize emergency response plans in real time. The specific operation is as follows: A1: Integrate multi-source fusion data and early warning information to extract core feature parameters of emergency event type, risk level, impact range and real-time resource status; A2: Based on the core feature parameters, the preset plan matching algorithm is invoked to match the optimal plan fragment from the emergency plan management module and generate an initial emergency response plan; A3: Simulate the execution process of the initial plan through a multi-agent simulation algorithm, and output the inference results of resource consumption, handling timeliness, and risk control effect; A4: Compare the simulation results with the preset safety thresholds and disposal targets. If there are deviations, adjust the scheme parameters through the gradient descent algorithm and iteratively generate the optimal emergency response scheme.

[0011] The specific formulas involved in A1-A4 are as follows: The formula for quantifying risk level is: ; Among them, R is the comprehensive risk quantification value of sudden events; S is the basic risk level score; S is the quantitative value of the scope of impact; T is the emergency response urgency coefficient; These are the weighting coefficients; The formula for resource availability is: ; in, Real-time availability of personnel, equipment, and materials for emergency response; This represents the current amount of available resources. This represents the total quantity of this type of resource; The resource decay coefficient is t; t is the duration the resource has been occupied. The preset contingency plan matching algorithm is invoked to match the optimal contingency plan fragment from the emergency plan management module (32) to generate an initial emergency response plan. The cosine similarity matching formula is as follows: ; in, For feature vector similarity; This is the core feature vector of the current emergency. The feature vector is a fragment of a historical contingency plan; n is the dimension of the feature vector. The comprehensive deviation assessment formula is: ; in, To comprehensively address deviations; To output the results of the deduction; This refers to the preset target value for the corresponding indicator; For deviation weights; The objective function of the gradient descent algorithm is: ; Constraints: ; in, These are the parameters of the scheme to be optimized; For parameters The reasonable range of values; Let be the scheduling amount for the k-th type of resource; is the total available quantity of this type of resource; m is the number of resource types.

[0012] The simulation exercise and evaluation unit includes a digital twin scenario construction module, a human-machine collaborative exercise module, a multi-dimensional quantitative evaluation module, and a contingency plan optimization and driving module, wherein: The digital twin scenario construction module: Based on multi-source fusion data and digital twin technology, it constructs a simulated accident scenario that is dynamically coupled with the actual enterprise scenario; The human-machine collaborative drill module is used to support adaptive emergency drills that combine automatic system triggering with manual intervention. The multi-dimensional quantitative evaluation module is used to quantitatively evaluate the exercise process and results from the dimensions of handling procedures, system response, and personnel operation. The contingency plan optimization driving module is used to iteratively optimize the existing emergency plans based on the quantitative evaluation results.

[0013] The digital twin scenario construction module, based on multi-source fusion data and digital twin technology, constructs simulated accident scenarios that are dynamically coupled with the actual enterprise scenario. The specific operation is as follows: B1: Collect the geometric parameters and attribute information of the physical elements of power generation enterprises' equipment, pipelines, security areas, and production workshops to construct a 1:1 scale digital twin basic model; B2: By linking the equipment operation data, environmental perception data, and security monitoring data from the multi-source fusion data to the corresponding elements of the digital twin basic model through a data mapping algorithm, a real-time mapping relationship between physical quantities and virtual model parameters is established. The specific formula for the data mapping algorithm is as follows: ; Where k is the scaling factor and b is the correction factor; B3: Based on historical accident data and real-time risk warning information, determine the fault type, initial location and impact chain of the simulated accident, inject accident disturbance parameters into the digital twin basic model, and generate the initial simulated accident scenario; B4: The deviation between the virtual scene and the physical scene is calculated in real time by evaluating the coupling degree C. When C exceeds the preset threshold, the virtual scene parameters are corrected by using real-time data fusion from multiple sources. The specific formula for evaluating the coupling degree is as follows: ; in, Let be the virtual parameter of the i-th element. For the corresponding physical quantity, n is the number of elements.

[0014] The multi-dimensional quantitative evaluation module quantifies the exercise process and results from the dimensions of handling procedures, system response, and personnel operations. The specific operation is as follows: C1: Determine the core evaluation indicators for each dimension, including the process dimension (process compliance rate, step omission rate); the system response dimension (response latency, resource scheduling accuracy); and the personnel operation dimension (operation accuracy, number of misoperations). C2: Extract the raw data of each evaluation indicator from the simulation exercise process log, including the execution record of the exercise steps, system response timing data, and personnel operation behavior data; C3: Quantitatively scores each indicator using a preset formula, specifically including: ① Compliance rate of the handling process: ; ②System response efficiency score: ; ③ Personnel operation accuracy: ; C4: Calculate the comprehensive score based on the weights of each dimension's indicators, using the following formula: ; in, The report assigns weights to the handling process, system response, and personnel operation dimensions, and outputs a quantitative evaluation report that includes scores for each dimension and optimization suggestions.

[0015] The visualization unit includes a multi-source data visualization module, an emergency information display module, and a drill evaluation result display module, wherein: The multi-source data visualization module is used to intuitively present multi-source fused data and enterprise security status in the form of charts and dashboards. The emergency information display module is used to display early warning information, emergency response progress, and resource allocation status in real time. The exercise evaluation results display module is used to present the scenario process of the simulation exercise, the quantitative evaluation results, and suggestions for optimizing the contingency plan.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention enables the real-time generation and iterative optimization of emergency response plans through a dynamic intelligent simulation engine. Combined with cross-departmental collaborative scheduling and process tracking, it significantly improves the efficiency and accuracy of emergency response. Furthermore, it constructs high-fidelity dynamically coupled accident scenarios based on digital twins, conducts human-machine collaborative adaptive drills, and implements multi-dimensional quantitative assessments to achieve continuous verification and iterative optimization of emergency plans, thereby significantly improving the emergency response capabilities and safety risk prevention and control level of power generation enterprises. Attached Figure Description

[0017] Figure 1 This is a system diagram of a multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises according to the present invention.

[0018] Figure 2 This invention presents a flowchart of the construction and modification process for a digital twin scenario in a multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises.

[0019] Figure 3 This invention presents a flowchart of the hierarchical push and display process for early warning information in a multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises.

[0020] Explanation of icon numbers: 1. Data Aggregation and Fusion Unit; 11. Multi-Source Data Access Module; 12. Data Preprocessing Module; 13. Correlation and Fusion Module; 2. Intelligent Monitoring and Early Warning Unit; 21. Multi-Dimensional Status Monitoring Module; 22. Anomaly Risk Identification Module; 23. Hierarchical Early Warning Module; 3. Emergency Command and Decision-Making Unit; 31. Dynamic Intelligent Simulation Engine Module; 32. Emergency Plan Management Module; 33. Cross-Departmental Collaborative Dispatch Module; 34. Handling Process Tracking Module; 4. Simulation Exercise and Evaluation Unit; 41. Digital Twin Scenario Construction Module; 42. Human-Machine Collaborative Exercise Module; 43. Multi-Dimensional Quantitative Evaluation Module; 44. Plan Optimization Driven Module; 5. Visualization Unit; 51. Multi-Source Data Visualization Module; 52. Emergency Information Display Module; 53. Exercise Evaluation Result Display Module. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. Example

[0022] like Figures 1-3 As shown, this embodiment provides a multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises, including a data aggregation and fusion unit 1, an intelligent monitoring and early warning unit 2, an emergency command and decision-making unit 3, a simulation exercise and evaluation unit 4, and a visualization display unit 5. Specifically: the data aggregation and fusion unit 1 is used to collect multi-source heterogeneous data of different types and protocols within the power generation enterprise, preprocess and fused the collected data to generate a unified multi-source fused data resource; the intelligent monitoring and early warning unit 2, based on the fused data, monitors the multi-dimensional security status of the power generation enterprise's production operation, network security, and physical security in real time, identifies abnormal risks, and... The system includes: 1) Issuing early warning alerts; 2) Emergency command and decision-making unit 3: Used to generate and optimize emergency response plans in real time through a dynamic intelligent simulation engine based on multi-source fusion data, early warning information, and contingency plans during emergencies, and to execute adaptive cross-departmental collaborative command and dispatch; 3) Simulation exercise and evaluation unit 4: Based on multi-source fusion data, digital twins, and dynamic intelligent simulation technology, it constructs dynamically coupled accident scenarios, conducts adaptive emergency drills with human-machine collaboration, and drives emergency plans through multi-dimensional quantitative evaluation; 4) Visualization unit 5: Used to present the enterprise's safety status, data fusion results, early warning information, emergency response progress, and simulation exercise and evaluation results in a visual format.

[0023] It should be noted that the system uses the data aggregation and fusion unit 1 as a unified data foundation to support the intelligent monitoring and early warning unit 2 in realizing multi-dimensional risk perception and early warning. The early warning information triggers the emergency command and decision-making unit 3 to conduct dynamic simulation and collaborative handling. Its plans and processes are continuously verified and optimized in the digital twin environment through the simulation exercise and evaluation unit 4. Finally, the entire link status, process and results are integrated and presented by the visualization display unit 5.

[0024] In this embodiment, it should also be noted that the data aggregation and fusion unit 1 includes a multi-source data access module 11, a data preprocessing module 12, and an association and fusion module 13, wherein: the multi-source data access module 11 is used to adapt to heterogeneous data interfaces of different types and protocols of power generation enterprises, and to uniformly collect and access multiple types of data in production, security, and management; the data preprocessing module 12 is used to clean, convert formats, remove outliers, and filter redundant data on the collected raw data; the association and fusion module 13 is used to build a multi-source data association rule base, to deeply associate and fuse cross-type and cross-source data, and to generate unified multi-source fused data resources.

[0025] It should be noted that the multi-source data access module 11 collects heterogeneous data in a unified manner, the data preprocessing module 12 cleans and standardizes the data, and then the association and fusion module 13 achieves deep fusion of cross-source and cross-type data based on the association rule base, so as to jointly build a high-quality, semantically consistent unified multi-source fusion data resource.

[0026] Furthermore, it should be noted that the multi-source data access module 11 has the following protocol adaptation and collection scope: It has a built-in parsing engine for multiple types of protocols, including industrial control protocols (such as Modbus, OPC UA, IEC 61850), security equipment protocols (such as ONVIF, GB / T 28181), and management system protocols (such as HTTP / HTTPS, JDBC). It supports full coverage collection of production operation data (unit power, voltage, temperature), network security data (traffic logs, intrusion alarms), physical security data (video streams, access control records), environmental perception data (wind speed, humidity, dust concentration), and management business data (inspection records, work order information). At the same time, it is configured with a breakpoint resume and data caching mechanism. When the data transmission link is interrupted, the cached data can be automatically retransmitted after the link is restored, ensuring the integrity and continuity of data collection.

[0027] The data preprocessing module 12 employs a dual-mode cleaning process: rule-based cleaning and intelligent cleaning. Rule-based cleaning removes obviously invalid data (such as abnormal data with values ​​exceeding the equipment's range or invalid data with incorrect timestamps) based on preset thresholds. Intelligent cleaning identifies hidden abnormal data (such as slowly drifting unit operating parameters) using the isolated forest algorithm. The format conversion stage follows the data standardization specifications of power generation enterprises, uniformly converting raw data in different formats (such as binary, JSON, and CSV) into structured data tables and adding a unique identifier code (including data source, collection time, equipment number, etc.) to each data entry. Redundant data filtering is based on the similarity calculation of data features, removing repeatedly collected data from the same source to reduce the computational load of subsequent fusion processing. The preprocessed data must meet the quality standards of "outlier percentage ≤ 0.5% and data integrity ≥ 99%".

[0028] The rule base construction and fusion algorithm of the association and fusion module 13: The construction of the multi-source data association rule base is based on the core principle of "spatiotemporal consistency + business relevance": Spatiotemporal consistency rules establish data associations based on timestamps and spatial locations (such as equipment installation locations and monitoring areas) (e.g., associating operational data of the same unit at the same time with security data); business relevance rules establish data associations based on the production business logic of power generation enterprises (e.g., associating unit maintenance work order data with the corresponding equipment's operating status data). The association rule base supports dynamic updates and can be supplemented according to new data types and business needs. The data fusion stage adopts a layered fusion strategy: first, weighted fusion is performed on data of the same type and dimension (e.g., multi-sensor monitoring data of the same equipment), and then semantic fusion is performed on cross-type and cross-dimensional data (e.g., fusion of unit operation data with environmental temperature and humidity data), finally generating a unified multi-source fused data resource containing four-dimensional information of "equipment-time-indicator-status".

[0029] In this embodiment, it should also be noted that the intelligent monitoring and early warning unit 2 includes a multi-dimensional status monitoring module 21, an anomaly risk identification module 22, and a graded early warning module 23, wherein: the multi-dimensional status monitoring module 21: based on fused data, performs real-time monitoring and status perception of key indicators of power generation enterprise production operation, network security, and physical security; the anomaly risk identification module 22: is used to identify explicit and implicit anomaly risks in multi-dimensional monitoring data through preset safety thresholds and intelligent detection models; the graded early warning module 23: is used to generate and push graded and classified early warning information according to the risk level and impact range.

[0030] It should be noted that the multi-dimensional status monitoring module 21 collects and integrates data in real time and senses key safety status; the abnormal risk identification module 22 identifies explicit and implicit risks based on monitoring results using thresholds and intelligent models; and the graded early warning module 23 generates and pushes differentiated early warning information based on risk level and impact range.

[0031] Furthermore, it should be noted that the dual-layer identification mechanism and model training method of the abnormal risk identification module 22 adopts a dual-layer abnormal risk identification mechanism of "threshold discrimination + intelligent model" to balance the rapid identification of explicit risks and the accurate mining of implicit risks. The first layer is explicit risk threshold discrimination: based on the power generation enterprise's equipment operation and maintenance standards and safety management specifications, rigid safety thresholds are set for each monitoring indicator (such as the upper limit of the unit's main steam temperature and the abnormal threshold of network traffic). When the monitoring data exceeds the threshold, it is directly judged as an explicit abnormal risk. The second layer is implicit risk intelligent model identification: for implicit risks without clear thresholds, such as equipment degradation and network attack precursors, a Long Short-Term Memory (LSTM) network model based on fused data training is adopted. The model training dataset covers the power generation enterprise's historical operating data, fault records, and risk event reports for the past 3 years. By learning the changing trends and correlation patterns of indicators, implicit risk signals such as slow drift of unit parameters and abnormal fluctuations in network traffic are identified. In addition, the model supports online self-learning and can dynamically optimize model parameters based on newly added risk event data to improve the accuracy of implicit risk identification.

[0032] The graded early warning module 23 classifies early warning information into four levels based on risk level and impact scope, and provides differentiated push strategies and response mechanisms: Level 1: Minor anomaly warning (e.g., a slight deviation of an auxiliary equipment parameter from the threshold, with no risk to production safety), pushed only to the corresponding shift's maintenance personnel via system pop-up; Level 2: General risk warning (e.g., abnormal local network traffic, non-core parameters of a single device exceeding the standard), pushed to workshop management personnel and maintenance supervisors via system pop-up + SMS; Level 3: Major risk warning (e.g., core unit parameters approaching the threshold, security alarms in key areas), pushed to the enterprise's safety management department and emergency command center via system pop-up + SMS + voice call; Level 4: Emergency warning (e.g., precursors to unit failure, network attack intrusion, fire alarms), pushed to enterprise management, emergency command team, and relevant regulatory departments via system pop-up + SMS + voice call + audible and visual alarms. Meanwhile, each warning message includes key information such as the risk type, scope of impact, handling recommendations, and responsible personnel, ensuring effective communication and rapid response to the warning information.

[0033] In this embodiment, it should also be noted that the emergency command and decision-making unit 3 includes a dynamic intelligent simulation engine module 31, an emergency plan management module 32, a cross-departmental collaborative scheduling module 33, and a handling process tracking module 34. The dynamic intelligent simulation engine module 31 is used to integrate multi-source fusion data and early warning information, and to generate and dynamically optimize emergency response plans in real time through intelligent algorithms. The specific operation is as follows: A1: Integrate multi-source fusion data and early warning information to extract core feature parameters such as the type of emergency, risk level, scope of impact, and real-time resource status; the risk level quantification formula is: ; Among them, R is the comprehensive risk quantification value of sudden events; S is the basic risk level score; S is the quantitative value of the scope of impact; T is the emergency response urgency coefficient; These are the weighting coefficients; The formula for resource availability is: ; in, Real-time availability of personnel, equipment, and materials for emergency response; This represents the current amount of available resources. This represents the total quantity of this type of resource; The resource decay coefficient is t; t is the duration the resource has been occupied. A2: Based on core feature parameters, a preset contingency plan matching algorithm is invoked to match the optimal contingency plan fragment from the emergency plan management module 32, generating an initial emergency response plan; the preset contingency plan matching algorithm is invoked to match the optimal contingency plan fragment from the emergency plan management module 32, generating an initial emergency response plan, using the cosine similarity matching formula: ; in, For feature vector similarity; This is the core feature vector of the current emergency. The feature vector is a fragment of a historical contingency plan; n is the dimension of the feature vector. A3: Simulate the execution process of the initial plan through a multi-agent simulation algorithm, and output the inference results of resource consumption, handling timeliness, and risk control effect; A4: Compare the simulation results with the preset safety thresholds and response targets. If deviations exist, adjust the scheme parameters using the gradient descent algorithm and iteratively generate the optimal emergency response scheme. The comprehensive deviation evaluation formula is: ; in, To comprehensively address deviations; To output the results of the deduction; This refers to the preset target value for the corresponding indicator; For deviation weights; The objective function of the gradient descent algorithm is: ; Constraints: ; in, These are the parameters of the scheme to be optimized; For parameters The reasonable range of values; Let be the scheduling amount for the k-th type of resource; is the total available quantity of this type of resource; m is the number of resource types.

[0034] Emergency Plan Management Module 32: Used to store, retrieve and update various emergency plans; Cross-departmental Collaborative Dispatch Module 33: Used to share information and issue instructions across departments and levels during emergency command, supporting adaptive collaborative command; Handling Process Tracking Module 34: Used to track the execution progress of emergency handling plans in real time, provide feedback on handling effects and support immediate adjustments to the plans.

[0035] It should be noted that the emergency plan management module 32 is a plan knowledge base. The dynamic intelligent simulation engine module 31 extracts features from multi-source fusion data and early warning information, matches plan fragments, simulates and iteratively optimizes to generate the optimal disposal plan. The cross-departmental collaborative scheduling module 33 executes multi-level instruction distribution and resource coordination accordingly, while the disposal process tracking module 34 provides real-time feedback on the execution status and effect.

[0036] Furthermore, it should be noted that, and The value can be a non-negative integer; The value range is [0, 0.1], and is determined according to the resource type of the power generation enterprise; the value of t is ≥ 0.

[0037] The cosine similarity matching formula in A2 is as follows: The preset similarity threshold is 0.8. When Sim≥0.8, the corresponding plan fragment is directly matched. When 0.5≤Sim<0.8, multiple high-similarity plan fragments are merged to generate a composite plan. When Sim<0.5, manual intervention is triggered to formulate a temporary plan.

[0038] Constraints of the multi-agent simulation algorithm in A3: The multi-agent simulation algorithm needs to construct three types of agents for emergency response of power generation enterprises, namely, command and decision-making agent, resource scheduling agent, and on-site execution agent. The simulation time step is set to 5 minutes, and the simulation duration is not less than the standard duration for handling emergencies.

[0039] The real-time feedback mechanism and plan adjustment triggering conditions of the handling process tracking module 34 are achieved by acquiring handling progress data through a dual channel of "manual reporting + automatic system collection," enabling real-time awareness of the plan's execution status. Automatically collected data includes equipment operating parameters, security monitoring footage, and resource occupancy status; manually reported data includes on-site handling progress, encountered problems, and personnel availability, with reporting nodes linked to key steps in the handling plan.

[0040] The triggering conditions for scheme adjustments are set into two categories: ① Deviation trigger, when the comprehensive deviation assessment value... When the deviation data is automatically pushed to the dynamic intelligent simulation engine module 31, the scheme parameter optimization is initiated; ② Event triggering: when an unexpected situation occurs on site (such as risk spread exceeding the expected range or resource supply interruption), the on-site personnel initiate an adjustment application, which is reviewed by the command center and triggers the scheme reconstruction to ensure that the disposal plan always fits the actual scenario.

[0041] In this embodiment, it should also be noted that the simulation exercise and evaluation unit 4 includes a digital twin scenario construction module 41, a human-machine collaborative exercise module 42, a multi-dimensional quantitative evaluation module 43, and a contingency plan optimization driving module 44. Specifically: the digital twin scenario construction module 41 constructs a simulated accident scenario dynamically coupled with the actual enterprise scenario based on multi-source fusion data and digital twin technology. The specific operations are as follows: B1: Collect the geometric parameters and attribute information of the physical elements of the power generation enterprise's equipment, pipelines, security areas, and production workshops to construct a 1:1 scale digital twin basic model; B2: Associate the equipment operation data, environmental perception data, and security monitoring data from the multi-source fusion data with the corresponding elements of the digital twin basic model through a data mapping algorithm to establish a real-time mapping relationship between physical quantities and virtual model parameters. The specific formula for the data mapping algorithm is: ; Where k is the scaling factor and b is the correction factor; B3: Based on historical accident data and real-time risk warning information, determine the fault type, initial location and impact chain of the simulated accident, inject accident disturbance parameters into the digital twin basic model, and generate the initial simulated accident scenario; B4: The deviation between the virtual scene and the physical scene is calculated in real time by evaluating the coupling degree C. When C exceeds the preset threshold, the virtual scene parameters are corrected by using real-time data fusion from multiple sources. The specific formula for evaluating the coupling degree is as follows: ; in, Let be the virtual parameter of the i-th element. For the corresponding physical quantity, n is the number of elements.

[0042] Human-Machine Collaborative Drill Module 42: Used to support adaptive emergency drills combining automatic system triggering and manual intervention; Multi-Dimensional Quantitative Evaluation Module 43: Used to quantitatively evaluate the drill process and results from the dimensions of handling procedures, system response, and personnel operation; Specific operations are as follows: C1: Determine the core evaluation indicators for each dimension, including the handling procedure dimension (process compliance rate, step omission rate); the system response dimension (response delay time, resource scheduling accuracy); and the personnel operation dimension (operation accuracy, number of misoperations); C2: Extract the raw data of each evaluation indicator from the simulation drill process log, including drill step execution records, system response timing data, and personnel operation behavior data; C3: Quantitatively score each indicator using preset formulas, specifically including: ① Compliance rate of the handling process: ; ②System response efficiency score: ; ③ Personnel operation accuracy: ; C4: Calculate the comprehensive score based on the weights of each dimension's indicators, using the following formula: ; in, The system assigns weights to the handling process, system response, and personnel operation dimensions, and outputs a quantitative assessment report containing scores for each dimension and optimization suggestions. The contingency plan optimization-driven module 44 is used to iteratively optimize existing emergency plans based on the quantitative assessment results.

[0043] It should be noted that the digital twin scenario construction module 41 builds a high-fidelity, dynamically coupled accident simulation environment based on multi-source fusion data. The human-machine collaborative exercise module 42 supports adaptive exercises that combine automatic triggering and manual intervention in this environment. The multi-dimensional quantitative evaluation module 43 conducts index-based scoring and comprehensive evaluation of the entire exercise process in three dimensions: process, system, and personnel. The emergency plan optimization driving module 44 drives the structured iterative optimization of the emergency plan based on the evaluation results.

[0044] Furthermore, it should be noted that the modeling constraints in B1 are: "The digital twin basic model must reproduce the key physical elements of the power generation enterprise at a 1:1 scale, including generator sets, power transmission and transformation equipment, security monitoring points, emergency access routes, etc., with a geometric error of ≤0.5 meters, and the attribute information must be consistent with the physical equipment ledger."

[0045] The coupling degree evaluation formula in B4 has a preset coupling degree threshold of 0.05. When C > 0.05, the system automatically starts the virtual scene correction process, and calls multi-source fusion real-time data to update the virtual model parameters according to the principle of "core equipment priority and key area priority". After the correction is completed, the coupling degree needs to be recalculated until C ≤ 0.05 to ensure the dynamic consistency between the virtual accident scene and the physical scene.

[0046] The scoring constraint in C3 is: "System response efficiency score" The value range is [0, 100]. hour, When this ratio > 1, ".

[0047] In this embodiment, it should also be noted that the visualization display unit 5 includes a multi-source data visualization module 51, an emergency information display module 52, and a drill evaluation result display module 53, wherein: the multi-source data visualization module 51 is used to intuitively present multi-source fused data and enterprise security status in the form of charts and dashboards; the emergency information display module 52 is used to display early warning information, emergency response progress, and resource scheduling status in real time; and the drill evaluation result display module 53 is used to present the scenario process of the simulation drill, quantitative evaluation results, and contingency plan optimization suggestions.

[0048] It should be noted that the multi-source data visualization module 51 presents the enterprise's integrated data and overall security situation, the emergency information display module 52 focuses on the early warning, handling progress and resource scheduling dynamics of emergencies, and the exercise evaluation results display module 53 reviews and displays the simulation exercise process, quantitative evaluation results and suggestions for plan optimization.

[0049] Furthermore, it should be noted that the emergency information display module 52 displays information in a hierarchical manner: a differentiated display strategy is adopted according to the warning level (notification level, warning level, alarm level, emergency level)—notification level information is presented as a gray text bar at the bottom of the interface; warning level information is pushed out in a yellow pop-up window with a slight prompt sound; alarm level information is displayed at the top in an orange pop-up window, with the corresponding area's situation map highlighted simultaneously; emergency level information pops up in a red full-screen pop-up window, triggering the sound and light alarm and automatically displaying the core handling steps of the emergency plan.

[0050] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0051] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises, characterized in that, It includes a data aggregation and fusion unit (1), an intelligent monitoring and early warning unit (2), an emergency command and decision-making unit (3), a simulation exercise and evaluation unit (4), and a visualization display unit (5), wherein: The data aggregation and fusion unit (1) is used to collect multi-source heterogeneous data of different types and protocols within the power generation enterprise, preprocess and correlate the collected data, and generate a unified multi-source fusion data resource. The intelligent monitoring and early warning unit (2) is based on the fused data to monitor the multi-dimensional security status of power generation enterprises’ production and operation, network security, and physical security in real time, identify abnormal risks and issue early warning prompts. The emergency command and decision-making unit (3) is used to generate and optimize emergency response plans in real time through a dynamic intelligent simulation engine based on multi-source fusion data, early warning information and contingency plans when an emergency occurs, and to execute adaptive cross-departmental collaborative command and dispatch. The simulation exercise and evaluation unit (4) is based on multi-source fusion data, digital twin and dynamic intelligent simulation technology to construct a dynamically coupled accident scenario, carry out human-machine collaborative adaptive emergency drills, and drive the emergency plan through multi-dimensional quantitative evaluation. The visualization unit (5) is used to present the enterprise's security status, data fusion results, early warning information, emergency response progress, and simulation exercise and evaluation results in a visual form.

2. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 1, characterized in that, The data aggregation and fusion unit (1) includes a multi-source data access module (11), a data preprocessing module (12), and a correlation and fusion module (13), wherein: The multi-source data access module (11) is used to adapt to heterogeneous data interfaces of different types and protocols of power generation enterprises, and to uniformly collect and access multiple types of data in production, security and management. The data preprocessing module (12) is used to clean, convert, remove outliers and filter redundant data from the collected raw data. The association and fusion module (13) is used to construct a multi-source data association rule library, perform deep association and fusion of cross-type and cross-source data, and generate unified multi-source fusion data resources.

3. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 1, characterized in that, The intelligent monitoring and early warning unit (2) includes a multi-dimensional status monitoring module (21), an abnormal risk identification module (22), and a hierarchical early warning module (23), wherein: The multi-dimensional status monitoring module (21) is based on fused data to monitor and perceive the key indicators of power generation enterprise production operation, network security and physical security in real time. The abnormal risk identification module (22) is used to identify explicit and implicit abnormal risks in multi-dimensional monitoring data by using a preset safety threshold and an intelligent detection model. The graded early warning module (23) is used to generate and push graded and classified early warning information according to the risk level and the scope of impact.

4. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 1, characterized in that, The emergency command and decision-making unit (3) includes a dynamic intelligent simulation engine module (31), an emergency plan management module (32), a cross-departmental collaborative dispatch module (33), and a handling process tracking module (34), wherein: The dynamic intelligent simulation engine module (31) is used to integrate multi-source fusion data and early warning information, and to generate and dynamically optimize emergency response plans in real time through intelligent algorithms. The emergency response plan management module (32) is used to store, retrieve, and update various emergency response plans; The cross-departmental collaborative scheduling module (33) is used to share information and issue instructions across departments and levels during emergency command, supporting adaptive collaborative command. The process tracking module (34) is used to track the execution progress of the emergency response plan in real time, provide feedback on the response effect, and support the immediate adjustment of the plan.

5. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 4, characterized in that, The dynamic intelligent simulation engine module (31) integrates multi-source fusion data and early warning information, and uses intelligent algorithms to generate and dynamically optimize emergency response plans in real time. The specific operation is as follows: A1: Integrate multi-source fusion data and early warning information to extract core feature parameters of emergency event type, risk level, impact range and real-time resource status; A2: Based on the core feature parameters, the preset plan matching algorithm is called to match the optimal plan fragment from the emergency plan management module (32) and generate an initial emergency response plan; A3: Simulate the execution process of the initial plan through a multi-agent simulation algorithm, and output the inference results of resource consumption, handling timeliness, and risk control effect; A4: Compare the simulation results with the preset safety thresholds and disposal targets. If there are deviations, adjust the scheme parameters through the gradient descent algorithm and iteratively generate the optimal emergency response scheme.

6. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 5, characterized in that, The specific formulas involved in A1-A4 are as follows: The formula for quantifying risk level is: ; Among them, R is the comprehensive risk quantification value of sudden events; S is the basic risk level score; S is the quantitative value of the scope of impact; T is the emergency response urgency coefficient; These are the weighting coefficients; The formula for resource availability is: ; in, Real-time availability of personnel, equipment, and materials for emergency response; This represents the current amount of available resources. This represents the total quantity of this type of resource; The resource decay coefficient is t; t is the duration the resource has been occupied. The preset contingency plan matching algorithm is invoked to match the optimal contingency plan fragment from the emergency plan management module (32) to generate an initial emergency response plan. The cosine similarity matching formula is as follows: ; in, For feature vector similarity; This is the core feature vector of the current emergency. The feature vector is a fragment of a historical contingency plan; n is the dimension of the feature vector. The comprehensive deviation assessment formula is: ; in, To comprehensively address deviations; To output the results of the deduction; This refers to the preset target value for the corresponding indicator; For deviation weights; The objective function of the gradient descent algorithm is: ; Constraints: ; in, These are the parameters of the scheme to be optimized; For parameters The reasonable range of values; Let be the scheduling amount for the k-th type of resource; is the total available quantity of this type of resource; m is the number of resource types.

7. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 1, characterized in that, The simulation exercise and evaluation unit (4) includes a digital twin scenario construction module (41), a human-machine collaborative exercise module (42), a multi-dimensional quantitative evaluation module (43), and a contingency plan optimization driving module (44), wherein: The digital twin scenario construction module (41) constructs a simulated accident scenario that is dynamically coupled with the actual scenario of the enterprise based on multi-source fusion data and digital twin technology. The human-machine collaborative training module (42) is used to support adaptive emergency drills that combine automatic system triggering with manual intervention. The multi-dimensional quantitative evaluation module (43) is used to quantitatively evaluate the exercise process and results from the dimensions of handling procedures, system response, and personnel operation. The contingency plan optimization driving module (44) is used to iteratively optimize the existing emergency plan based on the quantitative evaluation results.

8. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 7, characterized in that, The digital twin scenario construction module (41) constructs a simulated accident scenario that is dynamically coupled with the actual scenario of the enterprise based on multi-source fusion data and digital twin technology. The specific operation is as follows: B1: Collect the geometric parameters and attribute information of the physical elements of power generation enterprises' equipment, pipelines, security areas, and production workshops to construct a 1:1 scale digital twin basic model; B2: By linking the equipment operation data, environmental perception data, and security monitoring data from the multi-source fusion data to the corresponding elements of the digital twin basic model through a data mapping algorithm, a real-time mapping relationship between physical quantities and virtual model parameters is established. The specific formula for the data mapping algorithm is as follows: ; Where k is the scaling factor and b is the correction factor; B3: Based on historical accident data and real-time risk warning information, determine the fault type, initial location and impact chain of the simulated accident, inject accident disturbance parameters into the digital twin basic model, and generate the initial simulated accident scenario; B4: The deviation between the virtual scene and the physical scene is calculated in real time by evaluating the coupling degree C. When C exceeds the preset threshold, the virtual scene parameters are corrected by using real-time data fusion from multiple sources. The specific formula for evaluating the coupling degree is as follows: ; in, Let be the virtual parameter of the i-th element. For the corresponding physical quantity, n is the number of elements.

9. The multi-source heterogeneous data fusion security monitoring and emergency command system for power generation enterprises according to claim 8, characterized in that, The multi-dimensional quantitative evaluation module (43) quantifies the exercise process and results from the dimensions of handling procedures, system response, and personnel operation. The specific operation is as follows: C1: Determine the core evaluation indicators for each dimension, including the process dimension (process compliance rate, step omission rate); the system response dimension (response latency, resource scheduling accuracy); and the personnel operation dimension (operation accuracy, number of misoperations). C2: Extract the raw data of each evaluation indicator from the simulation exercise process log, including the execution record of the exercise steps, system response timing data, and personnel operation behavior data; C3: Quantitatively scores each indicator using a preset formula, specifically including: ① Compliance rate of the handling process: ; ②System response efficiency score: ; ③ Personnel operation accuracy: ; C4: Calculate the comprehensive score based on the weights of each dimension's indicators, using the following formula: ; in, The report assigns weights to the handling process, system response, and personnel operation dimensions, and outputs a quantitative evaluation report that includes scores for each dimension and optimization suggestions.

10. The power generation enterprise multi-source heterogeneous data fusion security monitoring and emergency command system according to claim 1, characterized in that, The visualization unit (5) includes a multi-source data visualization module (51), an emergency information display module (52), and a drill evaluation result display module (53), wherein: The multi-source data visualization module (51) is used to intuitively present multi-source fused data and enterprise security status in the form of charts and dashboards; The emergency information display module (52) is used to display early warning information, emergency response progress and resource scheduling status in real time. The exercise evaluation results display module (53) is used to present the scenario process of the simulation exercise, the quantitative evaluation results, and the suggestions for optimizing the contingency plan.

Images