Emergency rescue strategy and system for emergencies in tunnel

Abstract

The invention discloses an emergency rescue strategy and system for emergencies in a tunnel, and belongs to the field of intelligent expressways and traffic control intelligence. The method comprises the following steps: acquiring historical emergency data and a historical management and control strategy; preprocessing the historical emergency data and the historical management and control strategy to obtain preprocessed data; acquiring entities, relationships and attributes from the preprocessed data; constructing a knowledge graph according to the entities, the relationships and the attributes; acquiring current emergency information; and obtaining a final emergency rescue scheme from the knowledge graph according to the current emergency information. According to the invention, by constructing the (special) long tunnel emergency rescue knowledge graph, systematic integration and efficient utilization of historical data and experience knowledge are realized, and rapid, accurate and intelligent information support and application services are provided for tunnel management departments in emergency rescue work.

Description

Technical Field

[0001] The present invention relates to the field of smart highways and intelligent traffic control, and in particular to an emergency rescue strategy and system for sudden incidents in tunnels. Background Art

[0002] With the rapid advancement of urbanization and the sharp increase in the number of cars, the safety and emergency response capabilities of (extra) long tunnels, as key nodes in the highway network, have become the focus of public attention. When emergencies such as vehicle rear-end collisions and fires occur in (extra) long tunnels, the limited space in the tunnel and the rapid accumulation of smoke make rescue more difficult, which not only causes serious casualties, but also leads to long-term traffic congestion, causing significant impacts on the social economy.

[0003] Faced with these severe challenges, although the tunnel management department has accumulated a large number of emergency handling records and valuable experience through long-term efforts, these data are often scattered in various departments or systems, making it difficult to form a unified and systematic knowledge system. For example, in a fire accident in a (special) long tunnel, the fire department was unfamiliar with the location of the facilities and passages in the tunnel, resulting in low initial rescue efficiency. Similarly, when directing traffic, the traffic management department also faces the problem of insufficient understanding of the accident site, making it difficult to quickly formulate an effective traffic control plan.

[0004] In order to change this situation and improve the efficiency of handling emergencies and emergency response capabilities in tunnels, it is urgent to introduce more advanced information technology. In recent years, knowledge graphs, as an important achievement in the field of artificial intelligence, have shown great application value in many fields such as medicine, finance, and education. Taking the medical field as an example, by building a disease knowledge graph, doctors can quickly query all possible diseases, treatment plans, and historical cases related to patient symptoms, so as to make more accurate diagnosis and treatment decisions.

[0005] In the field of emergency rescue for emergencies in tunnels, knowledge graphs can also play an important role. By constructing a knowledge graph for emergency rescue for emergencies in (extra) long tunnels, key information such as various types of emergencies (such as fires, traffic accidents, natural disasters, etc.), emergency response processes, resource scheduling plans, historical case analysis, etc. are organized and stored in the form of graphs. For example, when a fire accident occurs, by querying the knowledge graph, you can quickly understand the distribution of fire-fighting facilities in the tunnel, the location of the nearest escape route, and the experience and lessons learned from handling similar fire accidents in history. This information will greatly improve the decision-making efficiency and accuracy of rescue personnel, and provide strong support for quickly controlling the fire and evacuating personnel.

[0006] In addition, the knowledge graph also supports deep semantic analysis and reasoning functions. In an emergency, the knowledge graph can be quickly retrieved and intelligently reasoned through computer technology and artificial intelligence algorithms, which can simulate the human thinking process to analyze and solve complex problems. For example, when a fire occurs in a tunnel, the system quickly calls the module on fire emergency response in the knowledge graph. By analyzing real-time fire data, smoke diffusion, and the layout of fire-fighting facilities in the tunnel, the system can simulate the trend of fire spread and automatically recommend the best evacuation route and fire-fighting equipment use plan. At the same time, the system can also generate the optimal rescue resource scheduling plan based on historical fire cases and current resource distribution, including the deployment of fire trucks, rescue teams, and medical resources, to ensure efficient and orderly rescue operations. Summary of the invention

[0007] The purpose of the present invention is to provide an emergency rescue strategy and system for sudden incidents in tunnels, and to construct an efficient, intelligent and comprehensive emergency rescue strategy for sudden incidents in (extra) long tunnels, so as to significantly improve the emergency response speed and rescue efficiency of tunnels when facing sudden situations such as fires, traffic accidents, and natural disasters. The present invention utilizes advanced technologies such as knowledge graphs, comprehensively considers factors such as the level of the sudden incident, the location of the incident, the status of the incident, and the distribution of electromechanical facilities, and automatically generates a scientific and reasonable emergency rescue strategy to ensure the timely evacuation of personnel at the scene of the incident and the efficient assembly of rescue forces. Traffic flow sensing facilities and traffic control facilities are deployed in (extra) long tunnels to build an emergency rescue system for sudden incidents in tunnels.

[0008] In order to solve the above technical problems, the present invention provides an emergency rescue strategy for sudden incidents in a tunnel, comprising the following steps:

[0009] Obtain historical emergency data and historical control strategies;

[0010] Preprocess historical emergency event data and historical control strategies to obtain preprocessed data;

[0011] Extract entities, relations, and attributes from preprocessed data;

[0012] Build a knowledge graph based on entities, relationships, and attributes;

[0013] Get current emergency information;

[0014] Based on the current emergency information and knowledge graph, an emergency rescue plan is obtained.

[0015] Preferably, the preprocessing includes data cleaning, data formatting and key information extraction.

[0016] Preferably, obtaining entities, relations and attributes from the preprocessed data specifically includes the following steps:

[0017] Obtaining entities from preprocessed data based on natural language processing;

[0018] Identify relationships between entities;

[0019] Extract the attributes of an entity.

[0020] Preferably, the entities include event type, location, driving guidance route and rescue guidance route;

[0021] The attributes include the severity of the emergency and the performance parameters of the inducing equipment.

[0022] Preferably, a knowledge graph is constructed based on entities, relationships and attributes, specifically including the following steps:

[0023] Performing data integration on the preprocessed data and preprocessing the conflicting data to obtain integrated data;

[0024] Build a knowledge graph based on the integrated data.

[0025] Preferably, data integration is performed on the preprocessed data, specifically comprising the following steps:

[0026] The historical emergency data in the pre-processed data are uniformly formatted and standardized to obtain integrated data.

[0027] Preferably, the historical emergency data includes real-time data of the tunnel monitoring system, alarm system records, historical case database, expert interview records and policy and regulation database;

[0028] The emergency includes the initial stage, rescue stage and recovery stage;

[0029] The control strategies include initial control strategies, rescue period control strategies and recovery period control strategies.

[0030] Preferably, an emergency rescue plan is obtained based on the current emergency information and knowledge graph, which specifically includes the following steps:

[0031] Extract keywords from current emergency information to obtain current emergency keywords;

[0032] According to the current emergency keywords, the current emergency level is obtained;

[0033] Obtain real-time data on the current emergency scene;

[0034] According to the real-time data of the current emergency scene, the current emergency status is obtained;

[0035] Perform similarity matching from the knowledge graph based on the current emergency keywords, current emergency level, and current emergency status to obtain historical cases;

[0036] According to the current emergency information and the real-time data of the current emergency scene, the location and direction of the incident are obtained;

[0037] Get rescue evacuation routes based on the location and direction of the incident;

[0038] Get emergency rescue plans based on historical cases and rescue evacuation routes;

[0039] Rank and recommend emergency rescue plans.

[0040] Preferably, the current emergency information includes tunnel outfield facility sensing data and emergency overview;

[0041] The emergency rescue plan includes response measures for the current emergency, personnel evacuation routes and rescue routes.

[0042] The present invention also provides an emergency rescue system for sudden incidents in a tunnel, comprising:

[0043] Historical data acquisition module, used to obtain historical emergency data and historical control strategies;

[0044] A preprocessing module is used to preprocess historical emergency event data and historical control strategies to obtain preprocessed data;

[0045] Recognition module, used to obtain entities, relations and attributes from preprocessed data;

[0046] Knowledge graph construction module, used to build knowledge graphs based on entities, relationships and attributes;

[0047] Current data acquisition module, used to obtain current emergency information;

[0048] The emergency rescue plan generation module is used to obtain the emergency rescue plan based on the current emergency information and knowledge graph.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] By building a knowledge graph for emergency rescue of (extra) long tunnels, we can achieve the systematic integration and efficient use of historical data and experience knowledge, and provide fast, accurate and intelligent information support and application services for tunnel management departments in emergency rescue work. This will not only greatly improve the safe operation level and emergency response capabilities of tunnels, but also inject new vitality into the development of smart highways and intelligent traffic control. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings.

[0052] Figure 1 It is a schematic diagram of the division of the tunnel unit control section;

[0053] Figure 2 It is the control strategy when a Level I emergency occurs in control section 1;

[0054] Figure 3a This is a schematic diagram of the deployment of control equipment in the early stage of a Level I emergency;

[0055] Figure 3b This is a schematic diagram of the deployment of rescue and control equipment for Level I emergency incidents;

[0056] Figure 3c This is a schematic diagram of the deployment of Level I emergency recovery and control equipment;

[0057] Figure 4 It is the knowledge graph framework for emergency rescue of sudden incidents in tunnels;

[0058] Figure 5 Emergency rescue strategy recommendation workflow based on knowledge graph;

[0059] Figure 6 The operation mechanism of the decision-making system based on the knowledge graph of emergency rescue for sudden incidents in tunnels. DETAILED DESCRIPTION

[0060] Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention, so the present invention is not limited to the specific implementation disclosed below.

[0061] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0062] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0063] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0064] Embodiment 1: A strategy for emergency rescue of emergencies in tunnels, integrating core functions such as intelligent perception monitoring, emergency knowledge graph management, event classification response, and regional control induction. Through real-time monitoring of tunnel status by high-definition cameras, sensors, etc., a dynamic knowledge graph is constructed in combination with big data and AI technology to achieve rapid classification and precise response of events. The system divides the control area according to the tunnel structure and event location, flexibly adjusts traffic control, effectively guides vehicle evacuation, reduces secondary disasters, and comprehensively improves the emergency management level and rescue efficiency of tunnels. First, according to the severity and impact range of the emergency, it is divided into levels I to IV, and differentiated response measures are implemented. Taking level I events as an example, a detailed traffic control strategy is formulated to achieve rapid evacuation of vehicles by flexibly utilizing the opening of the central dividing strip and the cross tunnel for vehicles, while optimizing the rescue vehicle path, using the non-emergency tunnel as an alternative route, and reducing the rescue risk. The recovery stage focuses on orderly restoration of traffic order to ensure that the tunnel resumes normal traffic as soon as possible. This strategy system covers a total of 36 specific emergency rescue scenarios and 108 sets of induction strategies, ensuring the comprehensiveness and efficiency of emergency rescue of emergencies in tunnels. Secondly, a knowledge graph of emergency rescue for emergencies in tunnels was constructed. The construction of the knowledge graph went through key links such as data collection and preprocessing, knowledge extraction and fusion. By integrating multi-source data such as tunnel monitoring, alarm systems, historical cases, expert interviews, and policies and regulations, NLP and machine learning technologies were used to accurately extract and integrate key information such as events, locations, and induced routes to form a structured knowledge base. This graph not only provides rich emergency knowledge resources, but also maintains its efficiency and practicality through continuous optimization and updating. Finally, based on the above knowledge graph, an emergency rescue strategy recommendation process was designed. This process can receive the perception of tunnel outfield facilities and emergency event overviews in real time, quickly analyze event information, and construct or update the knowledge graph. By quickly matching existing cases or deeply analyzing event scenarios, the system can automatically generate highly targeted control strategies. In the case of no direct matching cases, the system will start the special situation handling mechanism or add new cases to the knowledge graph to ensure the flexibility and accuracy of emergency response. At the same time, the process also integrates functions such as emergency keyword extraction, on-site situation assessment, and precise location of the incident, providing accurate navigation and scientific decision-making support for rescue operations, significantly improving the speed of emergency response and the effectiveness of rescue measures. This technical solution provides a comprehensive, efficient, and intelligent solution for emergency response in long highway tunnels through hierarchical response, regional control, state definition, and the construction and application of knowledge graphs.

[0065] 1. Prerequisites

[0066] Emergency Classification

[0067] Based on the national "Fire Statistics Management Regulations", the Ministry of Public Security's "Notice on Revision of Road Traffic Accident Classification Standards" and the "Highway Traffic Emergency Response Plan" issued by the Ministry of Transport, highway (extra) long tunnel emergencies are divided into four levels: Level I (particularly dangerous), Level II (dangerous), Level III (general), and Level IV (minor) according to the severity of their disastrous consequences and the scope of impact.

[0068] Table 1 Classification of emergencies in highway tunnels

[0069]

[0070]

[0071] Division of control areas

[0072] In order to ensure the implementation of precise control strategies according to the location of the emergency, the (extra) long tunnel is divided into three control areas: ① Control Area 1: the area between the entrance and the first vehicle tunnel; ② Control Area 2: the middle section (excluding the first and last tunnel areas); ③ Control Area 3: the area between the last vehicle tunnel and the exit. See the schematic diagram of the tunnel unit control section division for details. Figure 1 .

[0073] Emergency Status Definition

[0074] In order to ensure that control strategies can be formulated in a timely and efficient manner when an emergency occurs in the tunnel, three states are defined according to different nodes of the development of the emergency, and corresponding control strategies are formulated accordingly. The emergency states are shown in Table 2.

[0075] Table 2 Definition of emergency status

[0076]

[0077] 2. Relationship between emergency rescue scenarios and control strategies - taking level I emergencies as an example

[0078] The impact range of a Level I emergency event is the entire tunnel, so the traffic control strategy should be studied based on the tunnel entrance. The control strategy for a Level I emergency event occurring in control section 1 is shown in Figure 2 , the schematic diagram of control equipment deployment is shown in Figure 3a , Figure 3b , Figure 3c .

[0079] (1) Initial stage of an emergency

[0080] In the early stage of an emergency, timely and effective control strategies can quickly minimize the impact of the emergency and prevent the situation from further deteriorating. At the same time, it can ensure that vehicles and passengers in the emergency can evacuate the dangerous area in a timely and safe manner to reduce casualties. Figure 2 shown.

[0081] Driving route①

[0082] Vehicle location: Upstream of tunnel entrance

[0083] Control strategy: For vehicles that have not yet entered the tunnel but are approaching the tunnel entrance, traffic instructions or instructions from on-site personnel are used to guide these vehicles to turn around using the opening in the central median and leave the tunnel entrance area.

[0084] Implementation purpose: To reduce the number of vehicles entering the tunnel, free up passage for rescue vehicles, and prevent vehicles from entering dangerous areas without knowing it.

[0085] Driving route②

[0086] Vehicle location: Tunnel entrance and upstream of the incident

[0087] Control strategy: For vehicles that have entered the tunnel but have not yet reached the point where the emergency occurs, they are induced to turn around in a safe area inside the tunnel and exit the tunnel through the central median opening.

[0088] Implementation purpose: To quickly evacuate vehicles close to dangerous areas in the tunnel to reduce potential casualties and property losses.

[0089] Driving route ③

[0090] Vehicle location: upstream of the tunnel entrance

[0091] Control strategy: For vehicles that have not yet reached the tunnel entrance tunnel, guide them to use the tunnel entrance tunnel to turn around and leave the tunnel.

[0092] Implementation purpose: To ensure that vehicles can leave the tunnel quickly to avoid increased risks due to long-term detention, while also providing convenience for rescue vehicles and personnel.

[0093] Driving route④

[0094] Vehicle location: upstream of the tunnel exit

[0095] Control strategy: For vehicles that have passed through the tunnel entrance tunnel but have not yet reached the tunnel exit tunnel, they can be induced to use the exit tunnel to turn around or directly leave the tunnel, depending on the specific situation.

[0096] Implementation purpose: To ensure that vehicles can leave the tunnel safely and orderly, reduce traffic pressure in the tunnel, and create favorable conditions for subsequent rescue work.

[0097] (2) Emergency rescue

[0098] During the rescue period of an emergency, the focus is on organizing rescue operations quickly, orderly, and efficiently to minimize casualties and property losses. The rescue management and control strategy for Level I emergencies is shown in Figure 3.

[0099] Rescue Route ①

[0100] Vehicle location: driving in from the emergency cave entrance for rescue.

[0101] Rescue strategy: Upon receiving the rescue order, the rescue vehicle will start immediately and drive to the emergency cave entrance as quickly as possible to ensure that it arrives at the scene as soon as possible.

[0102] Implementation purpose: Through timely and effective rescue actions, quickly control the development of emergencies and prevent the situation from further deteriorating.

[0103] Rescue Route ②

[0104] Vehicle location: Entering the rescue from the non-emergency hole (see Figure 4-6 b and c routes).

[0105] Rescue strategy: Upon receiving the rescue order, the rescue vehicle enters through the non-emergency tunnel entrance and uses the cross tunnel as an alternative route for the rescue vehicle to enter the emergency scene.

[0106] Implementation purpose: By choosing a non-emergency entrance to enter, rescue vehicles can avoid being directly exposed to the dangerous environment of the emergency scene, thereby reducing the safety risks of personnel and vehicles during the rescue process.

[0107] (3) Emergency recovery

[0108] During the recovery period of an emergency, the focus is on fully restoring traffic order and infrastructure functions. At this stage, the control strategies for the emergency tunnel and the non-emergency tunnel in the double-hole tunnel are different. For the tunnel where the emergency occurs, the control strategy of "driving route ①" can be adopted. For the tunnel where the non-emergency occurs, the control strategy is as follows.

[0109] Driving route⑤

[0110] Vehicle location: Traveling in a non-emergency tunnel.

[0111] Control strategy: When there is no emergency, the traffic light at the tunnel entrance will show "yellow", and the driver should drive normally according to the "positive↓negative×" displayed on the lane indicator.

[0112] Similarly, the relationship between the emergency rescue scenarios and control strategies for Level II (dangerous), Level III (general), and Level IV (minor) emergencies in (extra) long tunnels can be obtained. Finally, we can obtain four levels of emergencies, namely, Level I, II, III, and IV, three locations of occurrence, namely, the front, middle, and back, and three stages of emergencies, namely, the initial stage, rescue, and recovery, with a total of 36 emergency rescue scenarios and 108 induction strategies.

[0113] 3. Knowledge graph construction

[0114] The construction of the knowledge graph of emergency rescue in tunnels covers three levels: data collection and preprocessing, knowledge extraction, and knowledge fusion. The framework is as follows: Figure 4 shown.

[0115] (1) Data collection and preprocessing

[0116] 1) Data Source

[0117] To ensure the comprehensiveness and accuracy of the data, relevant data will be collected from multiple channels, including real-time data from tunnel monitoring systems, alarm system records, historical case libraries, expert interview records, and emergency data such as policy and regulations libraries. These data sources will together form the data foundation for the construction of the knowledge graph.

[0118] 2) Data preprocessing

[0119] After data collection is completed, a series of preprocessing work is carried out to ensure the quality and availability of the data. Specific steps include: data cleaning (removing duplicate, erroneous or irrelevant information), data formatting (unifying data format and units), key information extraction (extracting key information such as event type, time of occurrence, location, and scope of impact), etc. Preprocessing work lays a solid foundation for subsequent knowledge extraction and fusion.

[0120] (2) Knowledge Extraction

[0121] 1) Entity Recognition

[0122] Using advanced natural language processing (NLP) technology and machine learning algorithms, various entities related to emergency rescue of sudden incidents in tunnels are automatically identified from pre-processed data. These entities include but are not limited to event type, location, driving guidance route, rescue guidance route, etc. Through entity recognition technology, basic node elements are provided for the construction of knowledge graphs.

[0123] 2) Relationship Extraction

[0124] On the basis of entity recognition, the relationship between entities is further analyzed and the corresponding relationship data is extracted. These relationship data will be used as edge elements of the knowledge graph to connect various entity nodes to form a complete network structure. In the process of relationship extraction, factors such as the directionality, strength, and attributes of the relationship will be considered to fully express the complex relationship between entities.

[0125] 3) Attribute extraction

[0126] In addition to entities and relationships, attribute information of entities will also be extracted to enrich the content of the knowledge graph. Attribute information includes but is not limited to the severity of the incident, performance parameters of the induction equipment, etc. These attribute information will help to more deeply understand the entity characteristics and rescue needs and provide a strong basis for subsequent decision support.

[0127] (3) Knowledge Fusion

[0128] 1) Data Integration

[0129] After completing the knowledge extraction, the data from different data sources such as the real-time data of the tunnel monitoring system, the alarm system records, the historical case library, the expert interview records, the policy and regulations library, etc. are integrated, and unified formatting and standardized processing (including data cleaning, data conversion, data mapping and other steps) are carried out to form a unified knowledge graph representation; including data cleaning, data conversion, data mapping and other steps. A unified data model and representation method will be used in the integration process to ensure seamless connection and sharing of data. At the same time, the quality of the integrated data will be evaluated to ensure that it meets the actual application needs.

[0130] 2) Conflict Resolution

[0131] Data conflicts may occur during the data integration process. To solve this problem, expert review, majority voting and other methods will be used to adjudicate conflicting data to ensure the accuracy and reliability of the final data. At the same time, a corresponding conflict resolution mechanism will be established to deal with similar problems that may arise in the future.

[0132] A: The specific process implemented through rules / data, etc.:

[0133] Step 1: Determine the criteria for identifying data conflicts. The definition rule is: if two data sources give different values ​​for the data at the same time point, then the data may conflict. Implementation method: Use SQL queries, data cleaning tools or custom scripts to scan the data in the database or data warehouse and mark potential conflicting data.

[0134] Step 2: Design rules or models to resolve conflicts. Establish clear rules: If the timestamp of the data is more recent, the latest data is preferred. Model training: Collect conflict cases and corresponding resolution results in historical data as training data, use these data to train the classifier, and use the classifier to predict the solution to the data conflict.

[0135] Step 3: Verify and adjust the solution. Randomly select a portion of automatically resolved conflict cases for manual review and use accuracy to evaluate the performance of the rules or models. Based on the verification results and performance evaluation, iteratively optimize the rules or models.

[0136] 3) Knowledge graph optimization

[0137] After the knowledge graph is built, it will be continuously optimized and improved. Specific measures include adding new entities and relationships, updating attribute values, and optimizing graph structures. Through continuous optimization and improvement, we ensure that the knowledge graph can keep up with the changes in actual application needs and always maintain its efficiency and accuracy.

[0138] 4. Emergency rescue strategy recommendation based on knowledge graph

[0139] (1) Workflow

[0140] The workflow of emergency rescue strategy recommendation based on knowledge graph can be described in detail as follows. The workflow diagram of emergency rescue strategy recommendation based on knowledge graph is shown in Figure 5 .

[0141] Step 1: The whole process starts from the "Start" step.

[0142] Step 2: Input two key information: "Tunnel off-site facility perception" and "Emergency overview". This information is the basis for subsequent analysis and processing.

[0143] Step 3: Based on the input of "tunnel external facility perception" and "emergency overview", the system records "emergency information". This information describes in detail the key elements of the event, such as the nature, scale, and location.

[0144] Step 4: Create or update a "knowledge graph" based on the emergency information. The knowledge graph is a structured knowledge base used to store and represent various information related to emergencies, including historical cases, response strategies, resource distribution, etc.

[0145] Step 5: First check whether there is any information or case that directly corresponds to the current emergency. This is a quick matching process to find out whether there is a ready-made solution or reference case.

[0146] If the corresponding information is found, the process continues downward.

[0147] If the corresponding information is not found, proceed to the "Whether to enter special situation processing" stage.

[0148] Step 6: This step is a decision point to determine whether the current event needs to be treated as a special case. This is usually based on factors such as the complexity of the event, the urgency, or the lack of relevant information in the knowledge graph.

[0149] If the decision is made to enter special circumstances, a range of special responses or further analysis may be required.

[0150] If it is decided not to enter special case processing, the process returns to further analysis of the knowledge graph.

[0151] Step 7: Match according to the specific "scene" of the emergency. The scene refers to the specific environment, conditions or circumstances in which the incident occurs.

[0152] Step 8: Check again whether there is any information or case matching the current scenario.

[0153] If a corresponding scenario is found, continue to the next steps.

[0154] If no corresponding scenario is found, the "matching failure additional case" mechanism is triggered.

[0155] Step 9: Add additional cases when matching fails: When no information that fully matches the current event or scenario is found in the knowledge graph, the system will start this mechanism. This means adding the current event as a new case to the knowledge graph so that it can be used as a reference when similar events occur in the future.

[0156] Step 10: Based on the information in the knowledge graph and the specific circumstances of the current event, the system generates and displays the corresponding "control strategies". These strategies include key information such as response measures for the current event, evacuation routes, and rescue routes.

[0157] Step 11: The entire process ends with "End", indicating the completion of the emergency rescue strategy recommendation work.

[0158] Through the above steps, the emergency rescue strategy recommendation workflow based on knowledge graph can efficiently handle emergencies and provide decision makers with scientific and reasonable response strategies.

[0159] (2) Reasoning Mechanism

[0160] The emergency information obtained by the workflow and the relevant information in the knowledge graph are used as input for the reasoning mechanism. The reasoning mechanism uses the rules and experience in the knowledge graph to perform logical reasoning and judgment based on this input information, thereby deriving the corresponding emergency rescue strategy. Therefore, the reasoning mechanism provides a scientific basis and intellectual support for the decision-making recommendation steps in the workflow.

[0161] The decision-making system based on the knowledge graph of emergency rescue in tunnels is a highly integrated, intelligent and dynamically adjusted system that can quickly respond to emergencies and provide strong support for rescue operations. The system operation process is shown in Figure 6 The operation mechanism of the decision-making system based on the knowledge graph of emergency rescue in tunnels can be elaborated in the following steps:

[0162] Step 1: Emergency triggering and preliminary analysis:

[0163] When an emergency occurs in the tunnel, the system will be activated immediately.

[0164] First, through the "emergency event keyword extraction" technology, keywords related to the event are quickly screened out from a large amount of information in order to make a preliminary judgment on the nature of the event.

[0165] Then, based on these keywords, the system further analyzes and determines the "event level" to provide preliminary guidance for subsequent rescue operations.

[0166] Step 2: On-site situation assessment and scene matching:

[0167] The system will use sensors, cameras and other equipment to collect real-time data on site, including but not limited to videos, images, sounds, etc.

[0168] Based on these data, the system conducts real-time assessment of the "event status" to understand the development trend of the event.

[0169] At the same time, the system will match historical cases and preset scenarios in the "knowledge graph" to find similar or related cases of event levels and event status to provide support for subsequent decision-making.

[0170] Search the knowledge graph for similar or related historical cases, which provide references and lessons for formulating specific emergency rescue plans;

[0171] Step 3: Event location identification and positioning:

[0172] In order to formulate rescue plans more accurately, the system needs to determine the specific location of the incident.

[0173] Through "induced system matching", the system can identify the location and direction of the incident in the tunnel based on the current emergency information and the real-time data of the current emergency scene. By accurately identifying the location and direction of the incident, the system can plan the optimal rescue route and evacuation route, thereby ensuring the efficiency and accuracy of the rescue operation.

[0174] Furthermore, by combining information such as "traffic lights", "lane indicators" and "signs inside the tunnel", the system can accurately locate specific locations and provide accurate navigation for rescue teams.

[0175] Step 4: Emergency rescue plan formulation and decision-making recommendations:

[0176] After collecting all the necessary information, the system will formulate a specific "emergency rescue plan" based on the rules and experience in the "knowledge graph" and the actual situation on site.

[0177] Based on historical cases, combined with current actual conditions and rescue needs, a more specific and feasible emergency rescue plan will be developed. This plan will take into account various possible factors, including the configuration of the rescue team, the demand for rescue materials, the choice of rescue routes, etc.

[0178] Finally, the system will rank and recommend multiple possible rescue plans based on the evaluation results, providing scientific and reasonable advice to decision makers.

[0179] Step 5: Solution evaluation and optimization:

[0180] Before making a decision, the system will re-evaluate the recommended rescue plan to ensure that it is optimal based on the criterion of "reasonableness".

[0181] If any deficiencies are found in the plan, the system will make timely adjustments and optimizations to ensure the smooth progress of the rescue operation.

[0182] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules, modules or units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units, modules or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0183] The units may or may not be physically separated, and the components displayed as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0184] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0185] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above-mentioned functions defined in the method of the present invention are executed. It should be noted that the above-mentioned computer-readable medium of the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of an electrical, magnetic, optical, electromagnetic, infrared segment, or semiconductor, or any combination of the above.

[0186] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0187] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An emergency rescue strategy for sudden incidents in tunnels, characterized in that: The following steps are involved: Obtain historical emergency data and historical control strategies; Preprocess historical emergency event data and historical control strategies to obtain preprocessed data; Extract entities, relations, and attributes from preprocessed data; Build a knowledge graph based on entities, relationships, and attributes; Get current emergency information; Based on the current emergency information and knowledge graph, an emergency rescue plan is obtained.

2. The emergency rescue strategy for sudden incidents in tunnels according to claim 1 is characterized by: The preprocessing includes data cleaning, data formatting and key information extraction.

3. The emergency rescue strategy for sudden incidents in tunnels according to claim 2 is characterized in that: Obtaining entities, relations, and attributes from preprocessed data includes the following steps: Obtaining entities from preprocessed data based on natural language processing; Identify relationships between entities; Extract the attributes of an entity.

4. The emergency rescue strategy for sudden incidents in tunnels according to claim 3 is characterized by: The entities include event type, location, driving guidance route and rescue guidance route; The attributes include the severity of the emergency and the performance parameters of the inducing equipment.

5. The emergency rescue strategy for sudden incidents in tunnels according to claim 4 is characterized in that: According to entities, relationships and attributes, a knowledge graph is constructed, which includes the following steps: Performing data integration on the preprocessed data and preprocessing the conflicting data to obtain integrated data; Build a knowledge graph based on the integrated data.

6. The emergency rescue strategy for sudden incidents in tunnels according to claim 5 is characterized in that: The preprocessed data is integrated, which specifically includes the following steps: The historical emergency data in the pre-processed data are uniformly formatted and standardized to obtain integrated data.

7. The emergency rescue strategy for sudden incidents in tunnels according to claim 6 is characterized by: The historical emergency data includes real-time data from the tunnel monitoring system, alarm system records, historical case databases, expert interview records, and policy and regulation databases; The emergency includes the initial stage, rescue stage and recovery stage; The control strategies include initial control strategies, rescue period control strategies and recovery period control strategies.

8. The emergency rescue strategy for sudden incidents in tunnels according to claim 7 is characterized in that: According to the current emergency information and knowledge graph, the emergency rescue plan is obtained, which includes the following steps: Extract keywords from current emergency information to obtain current emergency keywords; According to the current emergency keywords, the current emergency level is obtained; Obtain real-time data on the current emergency scene; According to the real-time data of the current emergency scene, the current emergency status is obtained; Perform similarity matching from the knowledge graph based on the current emergency keywords, current emergency level, and current emergency status to obtain historical cases; According to the current emergency information and the real-time data of the current emergency scene, the location and direction of the incident are obtained; Get rescue evacuation routes based on the location and direction of the incident; Get emergency rescue plans based on historical cases and rescue evacuation routes; Sort and recommend emergency rescue plans.

9. The emergency rescue strategy for sudden incidents in tunnels according to claim 8, characterized in that: The current emergency information includes tunnel outfield facility sensing data and emergency overview; The emergency rescue plan includes response measures for the current emergency, personnel evacuation routes and rescue routes.

10. An emergency rescue system for sudden incidents in a tunnel, used to implement the emergency rescue strategy for sudden incidents in a tunnel as claimed in any one of claims 1 to 9, characterized in that: include: Historical data acquisition module, used to obtain historical emergency data and historical control strategies; A preprocessing module is used to preprocess historical emergency event data and historical control strategies to obtain preprocessed data; Recognition module, used to obtain entities, relations and attributes from preprocessed data; Knowledge graph construction module, used to build knowledge graphs based on entities, relationships and attributes; Current data acquisition module, used to obtain current emergency information; The emergency rescue plan generation module is used to obtain the emergency rescue plan based on the current emergency information and knowledge graph.

Images