Subway safety risk and emergency capability assessment method and device based on scene construction

By constructing subway accident scenario simulation and digital twin technology, the deficiencies in accident scenario construction and emergency response capability assessment in the subway system have been resolved, global linkage analysis and dynamic emergency capability assessment have been achieved, and the emergency preparedness and response capabilities of subway safe operations have been improved.

CN120725430APending Publication Date: 2025-09-30CHENGDU INST OF URBAN SAFETY & EMERGENCY MANAGEMENT
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202510773783.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-11
Publication Date
2025-09-30

AI Technical Summary

Technical Problem

Existing technologies lack technical support for constructing accident scenarios for subway systems and emergency response capability assessments, resulting in incomplete and inaccurate subway safety risk assessments and making it difficult to effectively improve the pertinence and effectiveness of emergency preparedness.

Method used

By constructing a subway accident scenario simulation, combining digital twin technology and reinforcement learning algorithms, a time-space synchronized digital twin is generated, the accident evolution path is simulated, risk area division and emergency task configuration are carried out, emergency response capabilities are evaluated, and an emergency response capability assessment model is constructed.

Benefits of technology

It has achieved the ability to conduct global linkage analysis of the subway system and respond to changes in dynamic factors, improved the pertinence and applicability of emergency preparedness, reduced computing power consumption, and improved the speed of emergency response and the accuracy of emergency capability assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120725430A_ABST
    Figure CN120725430A_ABST
Patent Text Reader

Abstract

The invention discloses a subway safety risk and emergency capability assessment method and device based on scene construction. The method comprises the following steps: acquiring and preprocessing subway accident data; subway risk area accident scene construction is carried out based on the preprocessed subway accident data; performing risk assessment on each risk area of the subway based on the accident scene of the risk area of the subway; determining a major risk event according to the risk assessment result; configuring a corresponding emergency task according to the major risk event, and performing emergency drilling according to the emergency task to obtain drilling data of each drilling role in the drilling process; and based on the drilling data of each drilling role and the emergency preparation data of the subway system, carrying out emergency capability evaluation to obtain an emergency capability evaluation result. The method has the remarkable effects that the emergency capability evaluation model is constructed to evaluate the emergency capability, so that the overall emergency capability of the urban subway system is improved, and a powerful guarantee is provided for safe operation of the subway.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of subway operation management, and in particular to a scenario-based subway safety risk and emergency response capability assessment method and device. Background Art

[0002] In recent years, major cities have increased subway construction to address growing traffic pressures. Subway systems are relatively closed and independent, offering greater safety than other modes of transportation. However, due to increasing passenger volume and the increasing pressure on equipment, safety incidents are becoming increasingly common. Furthermore, the impact of safety incidents is becoming more severe. With the development of subway technology, safety requirements are constantly increasing. Numerous studies have found that safety assessment models in the subway sector, both domestically and internationally, hold great application value. Leveraging safety assessment systems, they can effectively enhance the prevention and control of safety incidents, enabling dynamic and timely detection and control of safety incidents. Based on the assessment results, targeted safety precautions can be proposed, ensuring the timely and safe operation of subways.

[0003] While a wealth of research has been conducted on assessing threats to subway operational safety from various perspectives, few studies offer comprehensive, systematic analysis encompassing all hazards, processes, and areas of concern. Furthermore, the sheer number of hazards and hazards involved in subway operations hinders their widespread application due to data collection difficulties.

[0004] Scenario building is a method that uses accident cases to create a comprehensive description of emergencies and provide organizations with emergency preparedness objectives. By building subway accident scenarios and matching them with the actual risks of subway operations, we can effectively improve the relevance and effectiveness of emergency preparedness, ultimately enhancing and improving emergency preparedness capabilities.

[0005] However, the existing technology lacks relevant technical support for the construction of accident scenarios for subway systems, and also lacks the assessment of emergency response capabilities in different accident scenarios. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the purpose of the present invention is to provide a scenario-based subway safety risk and emergency response capability assessment method and device. By constructing a subway accident scenario simulation to deduce the development of the accident, and conducting emergency response capability drills based on different accident scenarios, it can provide strong protection for the safe operation of the subway.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is as follows: In the first aspect, the present invention proposes a scenario-based subway safety risk and emergency response capability assessment method, which includes the following steps: Step 1: Collect and pre-process subway accident data; Step 2: Combined with scenario construction theory, construct subway risk area accident scenarios based on pre-processed subway accident data; Step 3: Conduct risk assessment on each risk area of ​​the subway based on the accident scenario of the subway risk area; Step 4: Obtain the hazard sources and influencing factors of each risk area of ​​the subway, and determine major risk events based on the risk assessment results; Step 5: Configure corresponding emergency tasks according to major risk events, conduct emergency drills based on the emergency tasks, and obtain drill data for each drill role during the drill; Step 6: Based on the drill data of each drill role and the emergency preparedness data of the subway system, an emergency capability assessment is performed to obtain an emergency capability assessment result.

[0008] Furthermore, the pre-processing of subway accident data in step 1 specifically includes: Step 1.1: Integrate and identify subway accident data from different sources to determine problematic data in the accident data; Step 1.2: Process the problematic data based on pre-determined data cleaning rules; Step 1.3: Based on a predetermined data format standard, standardize the data format of the accident data after the problem data processing; Step 1.4: Based on the accident research and analysis of the subway system, the accident data after data format standardization is sequentially subjected to time series continuity check and data consistency check to obtain pre-processed subway accident data with continuity and consistency in time series.

[0009] Furthermore, the construction of subway risk area accident scenarios based on the pre-processed subway accident data in step 2 specifically includes: Step 2.1: extract common features and individual features based on the preprocessed subway accident data, and generate a spatiotemporally aligned feature vector matrix based on the common features and individual features; Step 2.2: Construct a three-dimensional dynamic scene based on digital twin technology, load the eigenvector matrix in real time, and generate a digital twin synchronized in time and space; Step 2.3: Analyze the accident process mechanism of the subway system based on the digital twin to determine the hazard factors that may cause the accident; Step 2.4: Define the spatiotemporal boundaries, construct the event evolution logic chain, set key parameters, use the reinforcement learning algorithm to simulate the accident evolution path, and generate the scenario construction element model library required to meet the subway system accident scenarios; Step 2.5: Divide subway accidents into different locations to obtain multiple risk areas; Step 2.6: Collect risk data and emergency preparedness data from multiple risk areas, match them with the element model library, and construct accident scenarios for subway risk areas.

[0010] Furthermore, the steps of extracting common features and individual features based on the pre-processed accident data in step 2.1, and generating a spatiotemporally aligned feature vector matrix based on the common features and individual features specifically include: Extract common features using common feature extraction network; Extracting personality features using a personality feature extraction network; Normalize the extracted common features and individual features; The normalized common features and individual features are combined into the feature vector matrix using a feature fusion unit.

[0011] Furthermore, the reinforcement learning algorithm is used in step 2.4 to simulate the accident evolution path and generate a scenario construction element model library required to meet the subway system accident scenario, specifically including: Determine the initial accident node based on the set key parameters; Using a reinforcement learning algorithm to simulate the accident evolution path, and using a scenario entropy weighting algorithm to dynamically adjust the time step, to determine all possible evolution paths for each of the initial accident nodes; Determine all possible turning points of each evolution path of the subway system under the hazard factor; Determining a subsequent accident node corresponding to the initial accident node according to the initial accident node, all possible evolution paths of the initial accident node, and turning points of each evolution path; The element model library is constructed based on the initial accident node, the evolution path, the turning point and the subsequent event node.

[0012] Furthermore, in step 2.6, risk data and emergency preparedness data of multiple risk areas are collected and matched with the element model library to construct accident scenarios for subway risk areas, specifically including: Collect risk data and emergency preparedness data for multiple risk areas, where the risk data includes risk type, risk medium, on-site emergency resources, and surrounding environment information; the emergency preparedness data includes emergency group and responsibility information, firefighting force information, emergency linkage information, and emergency resource information; By matching risk data with data in the factor model library, we can determine the occurrence processes of several accidents with the highest similarity and construct the accident scenario stages; By matching risk data, emergency preparedness data and data in the factor model library, the accident expansion process of several accidents with the highest similarity is determined, and the accident scenario expansion stage is constructed; By matching risk data, emergency preparedness data and data in the factor model library, the secondary disaster processes of several accidents with the highest similarity are determined, and the accident scenario continuation stage is constructed; By matching risk data, emergency preparedness data and data in the factor model library, the accident termination process, number of casualties and accident losses of several accidents with the highest similarity are determined, and the accident scenario subsidence stage is constructed.

[0013] Furthermore, in step 3, risk assessment of each risk area of ​​the subway is conducted based on the accident scenario of the subway risk area, specifically including: Step 3.1: Select risk events in each risk area of ​​the subway system and conduct risk identification; Step 3.2: Build a risk assessment indicator system for various risk events; Step 3.3: Conduct risk assessment for each risk event and obtain risk level scores.

[0014] Furthermore, in step 4, the hazard sources and influencing factors of each risk area of ​​the subway are obtained, and major risk events are determined based on the risk assessment results; Step 4.1, matching corresponding scenario events in the accident scenarios of the subway risk areas according to the hazard sources of the subway risk areas; Step 4.2, according to the location of the scenario event in the subway risk area accident scenario; Step 4.3: Optimize and adjust the position of the scenario event in the subway risk area accident scenario based on the influencing factors; Step 4.4: Determine the risk level of the scenario event after the position adjustment based on the risk level score, and screen out major risk events based on the risk level.

[0015] Furthermore, in step 6, based on the drill data of each drill role and the emergency preparedness data of the subway system, an emergency capability assessment is performed to obtain the emergency capability assessment results, which specifically include: Step 6.1: Determine the emergency preparedness level score of each risk area based on the emergency preparedness data of the subway system; Step 6.2: Determine the emergency response capability level score of each risk area based on the drill data of each drill role; Step 6.3: Determine the weights corresponding to the emergency preparedness level score and the emergency response capability level score; Step 6.4: Establish an emergency response capability assessment model based on the determined weights. The expression is:

[0016] in, represents the emergency capability assessment model, For the i The rating of each emergency preparedness item, is the weight of the emergency preparedness rating score, For the j The grade score of emergency handling matters during each emergency drill, is the weight of the emergency response capability grade score, M and N The total number of emergency preparedness items and emergency handling items respectively; Step 6.5: Use the emergency capability assessment model to conduct emergency capability assessment and obtain emergency capability assessment results.

[0017] In a second aspect, the present invention proposes a scenario-based subway safety risk and emergency response capability assessment system for implementing the method described in the first aspect, including: Data acquisition module, used to collect and pre-process subway accident data; Scenario construction module, which is used to construct accident scenarios in subway risk areas based on pre-processed subway accident data in combination with scenario construction theory; Risk assessment module, used to conduct risk assessment of each risk area in the subway based on accident scenarios in the subway risk areas; A risk event determination module is used to obtain the hazard sources and influencing factors of each risk area in the subway and determine major risk events based on the risk assessment results; The drill data acquisition module is used to configure the corresponding emergency tasks according to the major risk events, conduct emergency drills according to the emergency tasks, and obtain the drill data of each drill role during the drill; The emergency capability assessment module is used to conduct emergency capability assessment based on the drill data of each drill role and the emergency preparedness data of the subway system to obtain emergency capability assessment results.

[0018] The remarkable effects of the present invention are: 1. The present invention first effectively integrates the collected multi-source data to construct accident scenarios for subway risk areas, then conducts risk assessment based on the constructed subway risk accident scenarios, and through emergency tasks under different major risk events, conducts on-site simulations with actual work backgrounds and drill roles to evaluate the emergency response capabilities of various subway risk areas, quantify the emergency preparedness of the urban subway system and its actual capabilities in performing emergency tasks, and constructs an emergency response capability assessment model to evaluate its emergency response capabilities, thereby improving the overall emergency response capabilities of the urban subway system and providing strong guarantees for the safe operation of the subway.

[0019] 2. In the process of constructing subway accidents, the present invention first effectively integrates the multi-source data used for scenario construction, effectively ensuring the authenticity of the constructed accident scenario, and can also realize global linkage analysis, can cope with changes in dynamic factors, and meet the processing capabilities of complex events; at the same time, the use of a dynamic deduction method overcomes the defects of traditional accident scenario construction methods that rely on fixed plans and lack real-time dynamic adjustment capabilities, and can construct a complete indicator system to measure and describe, which is more targeted and applicable; in addition, when using the reinforcement learning algorithm to simulate the accident evolution path, the scenario entropy weight algorithm is used to dynamically adjust the deduction granularity, and the computing power consumption is reduced by 50%, which greatly improves the response speed of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a structural diagram of the system of the present invention. DETAILED DESCRIPTION

[0021] The specific implementation manner and working principle of the present invention will be further described in detail below with reference to the accompanying drawings.

[0022] Example 1: like Figure 1 As shown, the embodiment of the present invention proposes a scenario-based subway safety risk and emergency response capability assessment method, the specific steps are as follows: Step 1: Collect and pre-process subway accident data; In some implementations, the collected subway accident data includes basic accident data, accident development data, emergency rescue process information, and accident consequence data. Specifically: The basic accident data include the accident point, accident type, accident medium, number of casualties and accident losses; the accident point refers to the location where the accident occurred; the accident type refers to fire, explosion, leakage, etc.; the accident medium refers to the type of medium that caused the accident, including but not limited to gasoline, liquefied gas and various chemical substances; the number of casualties refers to the number of deaths and injuries caused by the accident; the accident losses refer to the direct and indirect economic losses caused by the accident.

[0023] The accident development and evolution data include the accident occurrence process, accident expansion process, secondary disaster process and accident termination process; The emergency rescue process information includes the initial grassroots response, the first rescue dispatch, rescue force reinforcement, joint forces, emergency rescue duration and emergency resource requirements; The accident consequence data include legal liability and environmental impact.

[0024] In the specific implementation process, before processing the multi-source accident data, the collected accident data is first preprocessed to deal with missing values ​​and outliers in the data to ensure the continuity and consistency of the data in the time series. In other words, the specific steps for preprocessing subway accident data are as follows: Step 1.1: Integrate and identify subway accident data from different sources to determine problematic data in the accident data; wherein the problematic data includes missing values, outliers, and duplicate values; That is, accident data will be collected from different sources (such as sensors, monitoring, databases, etc.), and the data from these different sources will be integrated into a unified database for subsequent processing and analysis.

[0025] Step 1.2: Process the problematic data based on pre-determined data cleaning rules, wherein missing values ​​are processed based on interpolation, outliers are processed based on pre-determined statistical rules, and duplicate values ​​are deleted based on data fluctuations. Identify possible quality issues in the data, such as missing values, outliers, duplicate values, etc., and determine the standards and rules for data cleaning based on the characteristics of the data and business rules. Specifically: handle missing values, that is, use interpolation methods (such as linear interpolation, polynomial interpolation, etc.) to estimate missing values; if the missing data point has little impact on the overall analysis, the data point can be deleted directly; in addition, missing values ​​can be filled based on historical data or business rules. Handle outliers, that is, by setting thresholds, using statistical methods (such as standard deviation method, interquartile range method, etc.) or machine learning algorithms; identified outliers can be corrected, deleted or marked; Step 1.3: Based on a predetermined data format standard, the accident data after the problem data processing is standardized. That is, the data is converted into a unified format and unit to facilitate subsequent analysis and comparison, such as formatting the date and time, and standardizing the numerical values. Step 1.4: Based on the accident research and analysis of the subway system, the accident data after data format standardization is sequentially checked for time series continuity and data consistency. This obtains pre-processed subway accident data that is continuous and consistent in time series. This is to perform data quality checks and verifications to ensure that the cleaned data meets the analysis requirements. Visualization tools or statistical methods can also be used to check the distribution, trends, and correlations of the data.

[0026] In addition, the process and results of data cleaning can be recorded to facilitate subsequent data tracing and problem troubleshooting, and the cleaned data can be backed up to prevent data loss or damage.

[0027] Through the above-mentioned data preprocessing process, not only can the missing values, outliers and duplicate values ​​in the collected accident data be removed, thereby reducing the data volume and improving data processing efficiency, but also the negative impact of the above-mentioned data on the construction of accident scenarios can be eliminated, which helps to improve the accuracy and reliability of the accident scenarios constructed by the present invention.

[0028] Step 2: Combined with scenario construction theory, construct subway risk area accident scenarios based on pre-processed subway accident data; In some specific implementations, the specific steps for constructing a subway risk area accident scenario are: Step 2.1: extract common features and individual features based on the preprocessed subway accident data, and generate a spatiotemporally aligned feature vector matrix based on the common features and individual features; In some implementations, generating a spatiotemporally aligned eigenvector matrix may be implemented as follows: Extract common features using common feature extraction network; Extracting personality features using a personality feature extraction network; Normalize the extracted common features and individual features; The normalized common features and individual features are combined into the feature vector matrix using a feature fusion unit.

[0029] In some preferred embodiments, the common feature extraction network and the individual feature extraction network each include two FC+BN+GLU units, each unit is connected through a residual network and is finally normalized.

[0030] Through the above-mentioned feature combination process, not only can the key information in the accident data be captured, but also the multi-source data used for scenario construction can be effectively integrated, effectively ensuring the authenticity of the constructed accident scenario. It can also realize global linkage analysis, respond to changes in dynamic factors, and meet the processing capabilities of complex events.

[0031] Step 2.2: Construct a three-dimensional dynamic scene based on digital twin technology, load the eigenvector matrix in real time, and generate a digital twin synchronized in time and space; Step 2.3: Analyze the accident process mechanism of the subway system based on the digital twin to determine the hazard factors that may cause the accident; Hazards are rare or extreme events in the natural or man-made environment that can adversely affect human life, property, or various activities, reaching the point of causing a disaster. Examples include heavy rain and floods, droughts, tropical cyclones, storm surges, frost, low temperatures, hail, tsunamis, earthquakes, landslides, and mudslides.

[0032] The severity of a natural disaster is determined by the hazard-causing factors, the disaster-prone environment, and the affected objects. When a hazard-causing factor affects human society and causes disaster, it is called a hazard. When it does not affect human society, or when it does, or when the benefits far outweigh the harm, such as earthquakes and flash floods in uninhabited areas, or torrential rains that primarily alleviate droughts, these hazard-causing factors are simply called natural phenomenon variations.

[0033] Therefore, this embodiment analyzes the mechanism of the accident process of the disaster-prone object to determine the disaster-causing factors that may cause the accident from the characteristic vector matrix, and then uses the disaster-causing factors closely related to the disaster-prone object to simulate the accident evolution path and construct accident scenarios, which can be more targeted and applicable.

[0034] Step 2.4: Define the spatiotemporal boundaries, construct the event evolution logic chain, set key parameters, use the reinforcement learning algorithm to simulate the accident evolution path, and generate the scenario construction element model library required to meet the subway system accident scenarios; In this example, the key parameters include accident development rate, diffusion radius, personnel density, wind direction, wind speed, temperature, humidity, and equipment size.

[0035] Based on the above key parameters, this step uses a reinforcement learning algorithm to simulate the accident evolution path and generate a scenario construction element model library required to meet the accident scenario of the disaster-sustaining object. Specifically, it includes: Step 2.4.1. Determine the initial accident node based on the set key parameters; Step 2.4.2: Use the DDPG algorithm to simulate the accident evolution path, and use the scenario entropy weight algorithm to dynamically adjust the time step to determine all possible evolution paths for each initial accident node; The mathematical expression of the scenario entropy weight algorithm is as follows: in, is the maximum time step, S is the scenario complexity index, α is the attenuation coefficient.

[0036] Step 2.4.3, determining all possible turning points of each evolution path of the disaster-prone object under the disaster-causing factor; Step 2.4.4: Determine a subsequent accident node corresponding to the initial accident node based on the initial accident node, all possible evolution paths of the initial accident node, and turning points of each evolution path; Step 2.4.5: Construct the element model library based on the initial accident node, the evolution path, the turning point and the subsequent event node.

[0037] This embodiment uses a scenario entropy weight algorithm to dynamically adjust the deduction granularity, reducing computing power consumption by 50%, and greatly improving the response speed of this method.

[0038] Step 2.5: Divide subway accidents into different locations to obtain multiple risk areas; Step 2.6: Collect risk data and emergency preparedness data from multiple risk areas, match them with the element model library, and construct accident scenarios for subway risk areas.

[0039] In some preferred embodiments, the risk data includes risk type, risk medium, on-site emergency resources and surrounding environment information; The emergency preparedness data includes enterprise emergency grouping and responsibility information, enterprise firefighting force information, enterprise emergency linkage information and enterprise emergency resource information.

[0040] In this example, based on the aforementioned risk data and emergency preparedness data, the specific steps for constructing an accident scenario are as follows: Step 2.6.1. By matching the risk data with the data in the factor model library, determine the occurrence processes of several accidents with the highest similarity and construct the accident scenario occurrence stage; Step 2.6.2: By matching the risk data, emergency preparedness data, and data in the factor model library, determine the accident expansion processes of several accidents with the highest similarity and construct the accident scenario expansion stage; Step 2.6.3: By matching the risk data, emergency preparedness data, and data in the factor model library, determine the secondary disaster processes of several accidents with the highest similarity and construct the accident scenario continuation stage; Step 2.6.4: By matching the risk data, emergency preparedness data, and data in the factor model library, determine the accident termination process, number of casualties, and accident losses of several accidents with the highest similarity, and construct the accident scenario subsidence stage.

[0041] In the process of matching risk data, emergency preparedness data and data in the factor model library, the data types included in the risk data and emergency preparedness data are matched one by one with the factor models in the factor model library, and the correlation between the data is calculated. Based on the calculated correlation, the factors formed by similar accident data are screened, and finally the accident scenario is constructed with the screened factors.

[0042] Based on the accident scenario construction process described above, dynamic deduction is achieved, overcoming the shortcomings of traditional methods that rely on fixed plans and lack the ability to adjust them in real time. It can build a complete indicator system for measurement and description, making it more targeted and applicable. At the same time, it can also achieve global linkage analysis, respond to dynamic factors and meet the processing capabilities of complex events.

[0043] Step 3: Conduct risk assessment on each risk area of ​​the subway based on the accident scenario of the subway risk area; During the specific implementation process, the specific steps for risk assessment of each risk area in the subway include: Step 3.1: Select risk events in each risk area of ​​the subway system and conduct risk identification; In one embodiment of the present invention, a risk region is selected as the assessment target. Key risk events for that region are identified based on the region's natural geographical conditions, accident statistics, and emergency management-related information. The key risk events identified in this embodiment include seven categories: fire, flood, structural damage, power failure, signal system failure, terrorist attack, and passenger unrest.

[0044] Step 3.2: Build a risk assessment indicator system for various risk events; In one embodiment of the present invention, a risk assessment indicator system is constructed based on literature data and the actual situation of the selected risk area for seven key risk events: fire, flood, structural damage, power failure, signal system failure, terrorist attack, and passenger riot.

[0045] Step 3.3: Conduct risk assessment for each risk event and obtain risk level scores.

[0046] In one embodiment of the present invention, this step uses the hierarchical analysis method to invite experts to score the importance of risk indicators of seven types of risk events, scaling from 1 to 9. The larger the number, the greater the importance. A judgment matrix is ​​constructed, the consistency of the judgment matrix is ​​judged, and the risk indicator weight values ​​of the seven types of risk events are calculated.

[0047] Step 4: Obtain the hazard sources and influencing factors of each risk area of ​​the subway, and determine major risk events based on the risk assessment results; In some embodiments, the specific steps for determining a major risk event based on the risk assessment results are as follows: Step 4.1, matching corresponding scenario events in the accident scenarios of the subway risk areas according to the hazard sources of the subway risk areas; Step 4.2, according to the location of the scenario event in the subway risk area accident scenario; Step 4.3: Optimize and adjust the position of the scenario event in the subway risk area accident scenario based on the influencing factors; Step 4.4: Determine the risk level of the scenario event after the position adjustment based on the risk level score, and screen out major risk events based on the risk level.

[0048] In one embodiment of the present invention, this step performs an existing risk level evaluation based on an existing risk level evaluation standard according to a calculation result of an existing risk level score.

[0049] In one embodiment of the present invention, the existing risk level evaluation standard is that the existing risk level is set to five levels, such as level 1, level 2, level 3, level 4, and level 5. It is understood that this is only a specific form of representation of the five levels and is not limited to the above method. For example, in other embodiments, low, lower, medium, higher, and high may be used to represent the risk. It is also understood that in other embodiments, other numbers of risk levels may be set.

[0050] Major risk events are determined based on risk levels. In this embodiment, risk events at existing risk levels four and five are defined as major risk events, and three risk events, namely fire, flood, and terrorist attack, are screened as major risk events.

[0051] This step determines the risk level of the scenario event after the location adjustment through the risk level score, so that major risk events that have a serious impact on the safe operation of the subway can be screened out according to the risk level, which will help to conduct emergency drills based on major risk events and accurately evaluate the actual emergency response capabilities of each risk area.

[0052] Step 5: Configure corresponding emergency tasks according to major risk events, conduct emergency drills based on the emergency tasks, and obtain drill data for each drill role during the drill; According to the emergency tasks that can be configured for major risk events, the subway system's drill roles are configured in the subway risk area accident scenario. The subway system's drill roles may include: the personnel, vehicles, communication control systems, materials, and equipment.

[0053] After completing the configuration of the drill roles, you can use desktop simulation drills or on-site emergency simulation drills to record the planning and implementation process of each drill role, including the purpose of the drill, the form of the drill, the steps of the drill and the response process.

[0054] This step configures corresponding emergency tasks based on major risk events and conducts emergency drills based on the emergency tasks, so as to accurately assess the actual emergency response capabilities of each risk area based on the drill data of each drill role, thereby helping to improve the accuracy of the emergency response capability assessment results.

[0055] Step 6: Based on the drill data of each drill role and the emergency preparedness data of the subway system, an emergency capability assessment is performed to obtain an emergency capability assessment result.

[0056] During the implementation of this embodiment, the specific steps for performing emergency response capability assessment and obtaining the emergency response capability assessment result are as follows: Step 6.1: Determine the emergency preparedness level score of each risk area based on the emergency preparedness data of the subway system; Step 6.2: Determine the emergency response capability level score of each risk area based on the drill data of each drill role; Step 6.3: Determine the weights corresponding to the emergency preparedness level score and the emergency response capability level score; Step 6.4: Establish an emergency response capability assessment model based on the determined weights. The expression is:

[0057] in, represents the emergency capability assessment model, For the i The rating of each emergency preparedness item, is the weight of the emergency preparedness rating score, For the j The grade score of emergency handling matters during each emergency drill, is the weight of the emergency response capability grade score, M and N Respectively, the total number of emergency preparedness items and emergency handling items; in the specific implementation process, the + =1 and < , specifically, The value of is 0.45, The value of is 0.55; This embodiment performs a weighted evaluation of the emergency preparedness status and the actual emergency drill situation, which not only comprehensively considers the preparedness status and the actual drill situation, making the emergency capability assessment result more accurate, but also gives greater weight to the actual drill situation, thereby paying more attention to the emergency response capability when a risk accident actually occurs, so as to better conform to the actual situation of the subway emergency capability assessment, and the reliability of the obtained emergency capability assessment result is also higher.

[0058] Step 6.5: Use the emergency capability assessment model to conduct emergency capability assessment and obtain emergency capability assessment results.

[0059] Example 2: See attached Figure 2 The embodiment of the present invention proposes a scenario-based subway safety risk and emergency response capability assessment system for implementing the specific steps of the method described in Example 1. The system includes: Data acquisition module, used to collect and pre-process subway accident data; Scenario construction module, which is used to construct accident scenarios in subway risk areas based on pre-processed subway accident data in combination with scenario construction theory; Risk assessment module, used to conduct risk assessment of each risk area in the subway based on accident scenarios in the subway risk areas; A risk event determination module is used to obtain the hazard sources and influencing factors of each risk area in the subway and determine major risk events based on the risk assessment results; The drill data acquisition module is used to configure the corresponding emergency tasks according to the major risk events, conduct emergency drills according to the emergency tasks, and obtain the drill data of each drill role during the drill; The emergency capability assessment module is used to conduct emergency capability assessment based on the drill data of each drill role and the emergency preparedness data of the subway system to obtain emergency capability assessment results.

[0060] To sum up, the present invention first effectively integrates the collected multi-source data to construct accident scenarios for subway risk areas, then conducts risk assessment based on the constructed remaining accident scenarios for subway risks, and through emergency tasks under different major risk events, conducts on-site simulations with actual work backgrounds and exercise roles to evaluate the emergency response capabilities of various risk areas in the subway, quantifies the emergency preparedness of the urban subway system and its actual capabilities in performing emergency tasks, and constructs an emergency response capability assessment model to evaluate its emergency response capabilities, thereby improving the overall emergency response capabilities of the urban subway system and providing strong guarantees for the safe operation of the subway.

[0061] The technical solution provided by the present invention is introduced in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, the present invention can also be improved and modified in a number of ways, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A scenario-based subway safety risk and emergency response capability assessment method, characterized by: The steps include: Step 1: Collect and pre-process subway accident data; Step 2: Combined with scenario construction theory, construct subway risk area accident scenarios based on pre-processed subway accident data; Step 3: Conduct risk assessment on each risk area of ​​the subway based on the accident scenario of the subway risk area; Step 4: Obtain the hazard sources and influencing factors of each risk area of ​​the subway, and determine major risk events based on the risk assessment results; Step 5: Configure corresponding emergency tasks according to major risk events, conduct emergency drills based on the emergency tasks, and obtain drill data for each drill role during the drill; Step 6: Based on the drill data of each drill role and the emergency preparedness data of the subway system, an emergency capability assessment is performed to obtain an emergency capability assessment result.

2. The scenario-based subway safety risk and emergency response capability assessment method according to claim 1 is characterized in that: The preprocessing of subway accident data in step 1 specifically includes: Step 1.1: Integrate and identify subway accident data from different sources to determine problematic data in the accident data; Step 1.2: Process the problematic data based on pre-determined data cleaning rules; Step 1.3: Based on a predetermined data format standard, standardize the data format of the accident data after the problem data processing; Step 1.4: Based on the accident research and analysis of the subway system, the accident data after data format standardization is sequentially subjected to time series continuity check and data consistency check to obtain pre-processed subway accident data with continuity and consistency in time series.

3. The scenario-based subway safety risk and emergency response capability assessment method according to claim 1 is characterized in that: The construction of subway risk area accident scenarios based on pre-processed subway accident data in step 2 specifically includes: Step 2.1: extract common features and individual features based on the preprocessed subway accident data, and generate a spatiotemporally aligned feature vector matrix based on the common features and individual features; Step 2.2: Construct a three-dimensional dynamic scene based on digital twin technology, load the eigenvector matrix in real time, and generate a digital twin synchronized in time and space; Step 2.3: Analyze the accident process mechanism of the subway system based on the digital twin to determine the hazard factors that may cause the accident; Step 2.4: Define the spatiotemporal boundaries, construct the event evolution logic chain, set key parameters, use the reinforcement learning algorithm to simulate the accident evolution path, and generate the scenario construction element model library required to meet the subway system accident scenarios; Step 2.5: Divide subway accidents into different locations to obtain multiple risk areas; Step 2.6: Collect risk data and emergency preparedness data from multiple risk areas, match them with the element model library, and construct accident scenarios for subway risk areas.

4. The scenario-based subway safety risk and emergency response capability assessment method according to claim 3 is characterized in that: The steps described in step 2.1 are to extract common features and individual features from the pre-processed accident data, and generate a spatiotemporal aligned feature vector matrix based on the common features and individual features. Specifically, the steps include: Extract common features using common feature extraction network; Extracting personality features using a personality feature extraction network; Normalize the extracted common features and individual features; The normalized common features and individual features are combined into the feature vector matrix using a feature fusion unit.

5. The scenario-based subway safety risk and emergency response capability assessment method according to claim 3 is characterized in that: The reinforcement learning algorithm described in step 2.4 is used to simulate the accident evolution path and generate a scenario construction element model library required to meet the subway system accident scenario. Specifically, it includes: Determine the initial accident node based on the set key parameters; Using a reinforcement learning algorithm to simulate the accident evolution path, and using a scenario entropy weighting algorithm to dynamically adjust the time step, to determine all possible evolution paths for each of the initial accident nodes; Determine all possible turning points of each evolution path of the subway system under the hazard factor; Determining a subsequent accident node corresponding to the initial accident node according to the initial accident node, all possible evolution paths of the initial accident node, and turning points of each of the evolution paths; The element model library is constructed based on the initial accident node, the evolution path, the turning point and the subsequent event node.

6. The scenario-based subway safety risk and emergency response capability assessment method according to claim 3 is characterized in that: In step 2.6, risk data and emergency preparedness data of multiple risk areas are collected and matched with the element model library to construct accident scenarios for subway risk areas. Specifically, the following steps are involved: Collect risk data and emergency preparedness data for multiple risk areas, where the risk data includes risk type, risk medium, on-site emergency resources, and surrounding environment information; the emergency preparedness data includes emergency group and responsibility information, firefighting force information, emergency linkage information, and emergency resource information; By matching risk data with data in the factor model library, we can determine the occurrence processes of several accidents with the highest similarity and construct the accident scenario stages; By matching risk data, emergency preparedness data and data in the factor model library, the accident expansion process of several accidents with the highest similarity is determined, and the accident scenario expansion stage is constructed; By matching risk data, emergency preparedness data and data in the factor model library, the secondary disaster processes of several accidents with the highest similarity are determined, and the accident scenario continuation stage is constructed; By matching risk data, emergency preparedness data and data in the factor model library, the accident termination process, number of casualties and accident losses of several accidents with the highest similarity are determined, and the accident scenario subsidence stage is constructed.

7. The scenario-based subway safety risk and emergency response capability assessment method according to claim 1 is characterized in that: In step 3, the risk assessment of each subway risk area based on the accident scenario of the subway risk area specifically includes: Step 3.1: Select risk events in each risk area of ​​the subway system and conduct risk identification; Step 3.2: Build a risk assessment indicator system for various risk events; Step 3.3: Conduct risk assessment for each risk event and obtain risk level scores.

8. The scenario-based subway safety risk and emergency response capability assessment method according to claim 1 is characterized in that: In step 4, the hazard sources and influencing factors of each risk area of ​​the subway are obtained, and major risk events are determined based on the risk assessment results; Step 4.1, matching corresponding scenario events in the accident scenarios of the subway risk areas according to the hazard sources of the subway risk areas; Step 4.2, according to the location of the scenario event in the subway risk area accident scenario; Step 4.3: Optimize and adjust the position of the scenario event in the subway risk area accident scenario based on the influencing factors; Step 4.4: Determine the risk level of the scenario event after the position adjustment based on the risk level score, and screen out major risk events based on the risk level.

9. The scenario-based subway safety risk and emergency response capability assessment method according to claim 1 is characterized in that: In step 6, based on the drill data of each drill role and the emergency preparedness data of the subway system, an emergency capability assessment is conducted to obtain the following results: Step 6.1: Determine the emergency preparedness level score of each risk area based on the emergency preparedness data of the subway system; Step 6.2: Determine the emergency response capability level score of each risk area based on the drill data of each drill role; Step 6.3: Determine the weights corresponding to the emergency preparedness level score and the emergency response capability level score; Step 6.4: Establish an emergency response capability assessment model based on the determined weights. The expression is: in, represents the emergency capability assessment model, For the i The rating of each emergency preparedness item, is the weight of the emergency preparedness rating score, For the j The grade score of emergency handling matters during each emergency drill, is the weight of the emergency response capability grade score, M and N The total number of emergency preparedness items and emergency handling items respectively; Step 6.5: Use the emergency capability assessment model to conduct emergency capability assessment and obtain emergency capability assessment results.

10. A scenario-based subway safety risk and emergency response capability assessment system, used to implement the method described in any one of claims 1 to 9, characterized in that: include: Data acquisition module, used to collect and pre-process subway accident data; Scenario construction module, which is used to construct accident scenarios in subway risk areas based on pre-processed subway accident data in combination with scenario construction theory; Risk assessment module, used to conduct risk assessment of each risk area in the subway based on accident scenarios in the subway risk areas; A risk event determination module is used to obtain the hazard sources and influencing factors of each risk area in the subway and determine major risk events based on the risk assessment results; The drill data acquisition module is used to configure the corresponding emergency tasks according to the major risk events, conduct emergency drills according to the emergency tasks, and obtain the drill data of each drill role during the drill; The emergency capability assessment module is used to conduct emergency capability assessment based on the drill data of each drill role and the emergency preparedness data of the subway system to obtain emergency capability assessment results.

Citation Information

Cited By

  • Accident scene construction method and system for accident scene construction

    CN120745386A

  • Emergency broadcast scheduling command platform

    CN121261826A

  • Urban rail transit operation safety management method and system

    CN122616964A