Mine emergency rescue multi-level remote cooperative command training method and system

Abstract

The invention discloses a mine emergency rescue multi-level remote cooperative command training method and system, and belongs to the technical field of emergency rescue training, and the method comprises the steps: integrating a disaster physical engine based on a mine three-dimensional model, simulating the dynamic change of a disaster situation, and constructing a digital twinborn model of a dynamic disaster scene; the method comprises the following steps: generating tactical suggestions and rescue behavior options according to mine emergency rescue terms and a historical case library, generating different simulated rescue schemes according to the tactical suggestions and the rescue behavior options by a digital twin model, evaluating the rescue schemes by using a tactical evaluation algorithm, generating a quantitative comparison report, and deploying a master control engine; according to the method, global logic and data are managed in a unified manner, a multi-terminal cooperative sand table interaction tactical is generated according to a quantitative comparison report, cross-department disaster situations and resource states can be obtained in real time, the feasibility of a plan is improved, tactical decision and execution are synchronized and unified, and the plan drilling and reproduction cost is reduced.

Description

Technical Field

[0001] This invention belongs to the field of emergency rescue training technology, specifically, it relates to a multi-level remote collaborative command training method and system for mine emergency rescue. Background Technology

[0002] Mine emergency rescue is a high-risk operation that requires rescuers to possess strong comprehensive abilities in physical fitness, psychological resilience, knowledge, and experience. This is crucial for ensuring safe and efficient rescue operations and minimizing accident losses. Conducting rescue drills and training is a fundamental way to improve the capabilities of rescue personnel. The main types of accidents requiring mine emergency rescue include natural environmental disasters such as underground water inrush, fire, gas explosions, coal dust explosions, and roof collapses.

[0003] Existing mine emergency rescue command and training systems mostly involve single individuals or local teams, which cannot support collaborative decision-making and real-time simulations among multi-level command departments and rescue personnel in different locations. Different departments use independent simulation systems with inconsistent data formats, making it impossible to form a global situational awareness. Command training processes cannot be synchronized across platforms, making it difficult to simulate the collaborative processes between on-site command, regional command, and the general command, resulting in a disconnect between tactical decision-making and execution. Furthermore, emergency drills and reenactments are costly, time-consuming, and have low fidelity. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a multi-level, remote collaborative command training method for mine emergency rescue, comprising the following steps:

[0005] Acquire dynamic disaster scenario data in mines, and construct digital twin models and three-dimensional models of disaster evolution for dynamic disaster scenarios in mines;

[0006] Based on mine emergency rescue regulations and historical case database, tactical suggestions and rescue action options are generated. Using a digital twin model of the dynamic mine disaster scenario, different rescue plans are generated based on the tactical suggestions and rescue action options. The rescue plans are evaluated using a tactical evaluation algorithm, and a quantitative comparison report is generated.

[0007] Based on the quantitative comparison report, a multi-terminal collaborative sand table interactive tactic was generated, and mine emergency rescue training was conducted based on the multi-terminal collaborative sand table interactive tactic.

[0008] Furthermore, dynamic disaster scenario data in mines includes: three-dimensional geological models of the mine, roadway topology maps, equipment layout data, and roadway ventilation network parameters;

[0009] The specific method for constructing a digital twin model of a dynamic disaster scenario in a mine is as follows:

[0010] Based on the three-dimensional geological model of the mine, the topology map of the roadway, and the equipment layout data, a digital twin sand table that supports real-time rendering of disaster visualization effects is constructed.

[0011] The location of the fire source is determined, and the fire spread path and temperature distribution map are dynamically calculated based on the principle of fluid dynamics. Combined with the parameters of the roadway ventilation network, a fire spread model is constructed.

[0012] The location of toxic gas release is determined, and the diffusion range of toxic gas is simulated using the finite element method based on the gas concentration gradient, CO concentration gradient and ventilation parameters to construct a gas diffusion model.

[0013] Several fire source locations and several toxic gas release locations are set on the digital twin sand table. The fire spread model and gas diffusion model are assembled on the digital twin sand table. Rescue personnel, rescue equipment and virtual sensors are deployed on the digital twin sand table to obtain a digital twin model of the dynamic disaster scenario in the mine.

[0014] By setting different disaster intensities and environmental parameters, a three-dimensional model of disaster evolution is generated based on a digital twin model of a dynamic disaster scenario in a mine.

[0015] Disaster intensity parameters are used to quantify the initial state and development trend of a disaster, including: initial ignition temperature, ignition power, type of combustible material, initial fire spread rate, initial concentration of toxic gas release, gas release rate, initial gas diffusion pressure, and initial range of disaster impact.

[0016] Environmental parameters are the structural or dynamic environmental conditions of mine roadways that affect the evolution of disasters, including: roadway height, width, length, number of roadway corners and branching paths, roadway wind speed, air volume, ventilation fan operating status, ventilation network resistance coefficient, and humidity in the roadway;

[0017] On the digital twin sandbox, drag and drop to adjust the location of the fire source, set the initial disaster intensity, modify environmental parameters, generate a 3D model of disaster evolution, and drag and drop to deploy rescue personnel, equipment, and virtual sensors. Real-time mapping of resource status to the global view includes the number and allocation of rescue personnel, equipment, and virtual sensors.

[0018] Furthermore, based on mine emergency rescue regulations and a historical case database, tactical suggestions and rescue action options are generated. The specific method is as follows:

[0019] Acquire historical cases of mine emergency rescue and build a historical case database;

[0020] The mine emergency rescue clauses are transformed into a structured decision tree, and key decision nodes are selected by combining historical case databases to generate tactical suggestions and rescue behavior options.

[0021] Furthermore, tactical evaluation algorithms are used to assess different simulated rescue plans and generate quantitative comparison reports. The specific method is as follows:

[0022] Based on time constraints, different rescue plans are generated using task Gantt charts. The first and second curves of key decision nodes are monitored in real time. The first curve is the time deviation curve generated by the difference between the actual arrival time of the rescue team at the key location and the planned time. The second curve is the completion curve of the key rescue tasks, which include fire fighting, gas emission and personnel search and rescue.

[0023] A warning is issued when the arrival time deviation of the rescue team and the completion plan of the rescue mission exceed the preset threshold;

[0024] Using a digital twin model of a dynamic mine disaster scenario, different rescue plans were simulated to obtain simulation results for different rescue plans;

[0025] The tactical evaluation algorithm is used to evaluate the simulation results of different rescue plans. The tactical evaluation algorithm adopts a multi-dimensional weighted comprehensive evaluation algorithm, including: time efficiency dimension, resource consumption dimension, rescue outcome dimension and risk occurrence dimension.

[0026] Time efficiency includes the deviation between the total time of the rescue mission and the preset minimum time;

[0027] Resource consumption includes: the total amount of rescue equipment, personnel, and materials required for the rescue mission;

[0028] The rescue results include: the survival rate of trapped personnel and the disaster control rate;

[0029] Risk occurrence includes: the incidence rate of new risk events;

[0030] The deviation between the total time of the rescue mission and the preset minimum time, the total input of rescue equipment, personnel and materials, the survival rate of trapped people, the disaster control rate and the incidence of new risk events are normalized to make the corresponding values ​​of each dimension uniform. Then, weights are set for each dimension, and the comprehensive score of the rescue plan is calculated by weighted summation.

[0031] A quantitative comparison report of different rescue plans is generated based on the comprehensive score of the rescue plan and the deviation between the actual value of each dimension and the preset threshold. The report includes execution efficiency score, resource consumption statistics and deviation analysis.

[0032] Furthermore, the multi-terminal collaborative sandbox interaction tactic is a multi-level collaborative tactic, including: central cloud, regional edge cloud and field terminal.

[0033] Furthermore, mine emergency rescue training is conducted based on multi-terminal collaborative sand table interactive tactics. The specific methods are as follows:

[0034] The command center serves as the central cloud, where a main control engine is deployed to uniformly manage the global disaster evolution 3D model, resource status, and the time-series logic of rescue plans. Based on quantitative comparison reports, rescue tactical instructions are generated and sent to the regional edge cloud.

[0035] The rescue site is used as a regional edge cloud to receive rescue tactical instructions sent by the central cloud and feed back on on-site execution data. It runs a three-dimensional model of disaster evolution and adjusts the rescue tactical instructions in real time based on real-time environmental data on site.

[0036] Rescue team members serve as on-site terminals, wearing lightweight client equipment. Through these lightweight clients, they receive real-time adjusted rescue tactical instructions, receive feedback on their real-time location and on-site environmental data, and update the environmental and personnel data in the three-dimensional disaster evolution model in real time.

[0037] Furthermore, the command center adopts a heterogeneous intelligent agent collaboration mechanism, which synchronizes the operation events of each key decision node through a distributed message queue; it uses an LSTM time series prediction model based on data from virtual sensors in a digital twin sandbox to predict the evolution trend of the disaster within a preset time and generate early warning information, which is then pushed to various field terminals.

[0038] Furthermore, the central cloud also includes a role-based access control model for defining multi-level command roles and operational permissions;

[0039] The role-based access control (RBAC) permission model defines multi-level command roles and operational permissions. Specifically, it establishes a dynamic permission allocation mechanism. Using the RBAC permission model and dynamic permission allocation mechanism, different roles and permissions are assigned to on-site terminals at different stages of mine emergency rescue training. Multi-level command roles and operational permissions are defined, including command, rescue, and logistics layers. This ensures that the operational permissions and data of personnel at different levels are synchronized during mine emergency rescue training, enabling multi-role remote collaborative operations at the command, rescue, and logistics levels.

[0040] Furthermore, the dynamic permission allocation mechanism includes:

[0041] In the initial phase of mine emergency rescue training, a role-based access control model was used to assign basic permissions according to preset roles, and tactical arrows and warning zone markers were drawn on the digital twin model and shared in real time.

[0042] During the simulation of mine rescue plans, permissions are escalated based on event triggers. When multiple users modify the status of the same device simultaneously, a role-based access control model is used to arbitrate the validity of the operation according to role priority.

[0043] On the other hand, the present invention provides a multi-level remote collaborative command and training system for mine emergency rescue, including: a digital twin module, a report generation module, and a tactical generation module;

[0044] The digital twin module is used to acquire dynamic disaster scenario data in mines and construct a digital twin model of the dynamic disaster scenario in mines;

[0045] The report generation module is used to generate tactical recommendations and rescue action options based on mine emergency rescue regulations and historical case databases. The digital twin model generates simulations of different rescue plans based on the tactical recommendations and rescue action options, uses tactical evaluation algorithms to evaluate the rescue plans, and generates a quantitative comparison report.

[0046] The tactical generation module is used to generate multi-terminal collaborative sand table interactive tactics based on the quantitative comparison report, and to conduct mine emergency rescue training based on the multi-terminal collaborative sand table interactive tactics.

[0047] The beneficial effects of adopting the above technical solution are as follows: The multi-level remote collaborative command training method and system for mine emergency rescue provided by this invention achieves full-domain situational awareness by constructing a digital twin model of dynamic disaster scenarios, solving the problem of data silos, and enabling real-time acquisition of cross-departmental disaster and resource status; by generating simulated different rescue plans and evaluating them to generate quantitative comparison reports, the implementation effect of the rescue plans is quantitatively evaluated, improving the feasibility of the plans; through efficient collaborative simulation, multi-terminal collaborative sand table interactive tactics are generated, and a three-level cloud architecture is used for remote synchronous simulation, so that tactical decision-making and execution are synchronized and unified, reducing the cost of plan drills and reproduction. Attached Figure Description

[0048] Figure 1 Flowchart of a multi-level, remote collaborative command training method for mine emergency rescue provided in Embodiment 1 of this invention;

[0049] Figure 2 A schematic diagram of the structure of the multi-level remote collaborative command and training system for mine emergency rescue provided in Embodiment 2 of the present invention. Detailed Implementation

[0050] The specific implementation methods of this application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0051] Example 1

[0052] A training method for multi-level, remote collaborative command in mine emergency rescue, such as Figure 1 As shown, it includes the following steps:

[0053] Step 1: Acquire dynamic disaster scenario data in mines and construct a digital twin model of the dynamic disaster scenario in mines;

[0054] Acquiring dynamic disaster scenario data in mines includes: 3D geological model of the mine, roadway topology map, equipment layout data, and roadway ventilation network parameters;

[0055] The specific method for constructing a digital twin model of a dynamic disaster scenario in a mine is as follows:

[0056] Based on the three-dimensional geological model of the mine, the topology map of the roadway, and the equipment layout data, a digital twin sand table that supports real-time rendering of disaster visualization effects is constructed.

[0057] The location of the fire source is determined, and the fire spread path and temperature distribution map are dynamically calculated based on the principle of fluid dynamics. Combined with the parameters of the roadway ventilation network, a fire spread model is constructed.

[0058] The location of toxic gas release is determined, and the diffusion range of toxic gas is simulated using the finite element method based on the gas concentration gradient, CO concentration gradient and ventilation parameters to construct a gas diffusion model.

[0059] Several fire source locations and several toxic gas release locations are set on the digital twin sand table. The fire spread model and gas diffusion model are assembled on the digital twin sand table. Rescue personnel, rescue equipment and virtual sensors are deployed on the digital twin sand table to obtain a digital twin model of the dynamic disaster scenario in the mine.

[0060] By setting different disaster intensities and environmental parameters, a three-dimensional model of disaster evolution is generated based on a digital twin model of a dynamic disaster scenario in a mine.

[0061] Disaster intensity parameters are used to quantify the initial state and development trend of a disaster, including: initial ignition temperature, ignition power, type of combustible material, initial fire spread rate, initial concentration of toxic gas release, gas release rate, initial gas diffusion pressure, and initial range of disaster impact.

[0062] Environmental parameters are the structural or dynamic environmental conditions of mine roadways that affect the evolution of disasters, including: roadway height, width, length, number of roadway corners and branching paths, roadway wind speed, air volume, ventilation fan operating status, ventilation network resistance coefficient, and humidity in the roadway;

[0063] On the digital twin sandbox, drag and drop to adjust the location of the fire source, set the initial disaster intensity, modify environmental parameters, generate a 3D model of disaster evolution, and drag and drop to deploy rescue personnel, equipment, and virtual sensors. Real-time mapping of resource status to the global view includes the number and allocation of rescue personnel, equipment, and virtual sensors.

[0064] Step 2: Generate tactical recommendations and rescue action options based on mine emergency rescue regulations and historical case database. Use a digital twin model of the dynamic mine disaster scenario to generate different rescue plans based on the tactical recommendations and rescue action options. Use a tactical evaluation algorithm to evaluate the rescue plans and generate a quantitative comparison report.

[0065] Tactical suggestions and rescue action options are generated based on mine emergency rescue regulations and a historical case database. The specific method is as follows:

[0066] Acquire historical cases of mine emergency rescue and build a historical case database;

[0067] The mine emergency rescue clauses are transformed into a structured decision tree, and key decision nodes are selected by combining historical case databases to generate tactical suggestions and rescue behavior options.

[0068] Different simulated rescue plans are evaluated using tactical evaluation algorithms to generate a quantitative comparison report. The specific method is as follows:

[0069] Based on time constraints, different rescue plans are generated using task Gantt charts. The first and second curves of key decision nodes are monitored in real time. The first curve is the time deviation curve generated by the difference between the actual arrival time of the rescue team at the key location and the planned time. The second curve is the completion curve of the key rescue tasks, which include fire fighting, gas emission and personnel search and rescue.

[0070] A warning is issued when the arrival time deviation of the rescue team and the completion plan of the rescue mission exceed the preset threshold;

[0071] Using a digital twin model of a dynamic mine disaster scenario, different rescue plans were simulated to obtain simulation results for different rescue plans;

[0072] The tactical evaluation algorithm is used to evaluate the simulation results of different rescue plans. The tactical evaluation algorithm adopts a multi-dimensional weighted comprehensive evaluation algorithm, including: time efficiency dimension, resource consumption dimension, rescue outcome dimension and risk occurrence dimension.

[0073] Time efficiency includes the deviation between the total time of the rescue mission and the preset minimum time;

[0074] Resource consumption includes: the total amount of rescue equipment, personnel, and materials required for the rescue mission;

[0075] The rescue results include: the survival rate of trapped personnel and the disaster control rate;

[0076] Risk occurrence includes: the incidence rate of new risk events;

[0077] The deviation between the total rescue mission time and the preset minimum time, the total input of rescue equipment, personnel, and materials, the survival rate of trapped personnel, the disaster control rate, and the incidence rate of new risk events are normalized to ensure that the values ​​corresponding to each dimension are consistent. Then, weights are assigned to each dimension, and the comprehensive score of the rescue plan is calculated by weighted summation, as shown in the following formula:

[0078]

[0079] Wherein, α, β, γ and δ are all weighting coefficients; in this embodiment, α=0.3, β=0.2, γ=0.4 and δ=0.1 are set.

[0080] Based on the comprehensive score of the rescue plan and the deviation between the actual value of each dimension and the preset threshold, a quantitative comparison report of different rescue plans is generated, including execution efficiency score, resource consumption statistics and deviation analysis, so as to realize the ranking of the advantages and disadvantages of different plans.

[0081] Step 3: Based on the quantitative comparison report, generate a multi-terminal collaborative sand table interactive tactic, and conduct mine emergency rescue training based on the multi-terminal collaborative sand table interactive tactic;

[0082] The multi-terminal collaborative sand table interaction tactic is based on a central cloud-regional edge cloud architecture, combined with a distributed architecture to generate a heterogeneous intelligent agent collaboration mechanism, and then combined with an LSTM time series prediction model to predict the evolution trend of the disaster, generating a three-level collaborative technology of "central command-regional execution-on-site terminal feedback".

[0083] The multi-terminal collaborative sandbox interaction tactic is a multi-level collaborative tactic, including: central cloud, regional edge cloud and field terminal;

[0084] The command center serves as the central cloud, where a main control engine is deployed to uniformly manage the global 3D disaster evolution model, resource status, and the time-series logic of rescue plans. Based on quantitative comparison reports, rescue tactical instructions are generated and sent to regional edge clouds. The central cloud also includes a role-based access control model to define multi-level command roles and operational permissions. The command center uses a distributed message queue to synchronize operational events of key decision nodes. An LSTM time-series prediction model is used based on data from virtual sensors in the digital twin sandbox to predict the disaster evolution trend within a preset timeframe and generate early warning information, which is then pushed to various field terminals.

[0085] The rescue site is used as a regional edge cloud to receive rescue tactical instructions sent by the central cloud and feed back on on-site execution data. It runs a three-dimensional model of disaster evolution and adjusts the rescue tactical instructions in real time based on real-time environmental data on site.

[0086] Rescue team members serve as on-site terminals, wearing lightweight client devices to receive real-time adjusted rescue tactical instructions, receive feedback on real-time location and on-site environmental data, and update environmental and personnel data in the three-dimensional disaster evolution model in real time.

[0087] The role-based access control model defines multi-level command roles and operational permissions, specifically as follows:

[0088] Establish a dynamic permission allocation mechanism, using a role-based access control permission model and dynamic permission allocation mechanism to assign different roles and permissions to on-site terminals at different stages of mine emergency rescue training. Define multi-level command roles and operation permissions, including: command level, rescue level, and logistics level, so that the operation permissions and data of personnel at different levels are synchronized in mine emergency rescue training, and realize multi-role remote collaborative operation of command level, rescue level, and logistics level.

[0089] Dynamic permission allocation mechanisms include:

[0090] In the initial phase of mine emergency rescue training, a role-based access control model was used to assign basic permissions according to preset roles, and tactical arrows and warning zone markers were drawn on the digital twin model and shared in real time.

[0091] During the simulation of mine rescue plans, permissions are escalated based on event triggers. When multiple users modify the status of the same device simultaneously, a role-based access control model is used to arbitrate the validity of the operation according to role priority.

[0092] Example 2:

[0093] A multi-level, remote collaborative command and training system for mine emergency rescue, such as Figure 2 As shown, it includes: a digital twin module, a report generation module, and a tactical generation module;

[0094] The digital twin module is used to integrate a disaster physics engine based on a 3D mine model to simulate dynamic changes in disaster conditions and construct a digital twin model of a dynamic disaster scenario in the mine.

[0095] The report generation module is used to generate tactical recommendations and rescue action options based on mine emergency rescue regulations and historical case databases. The digital twin model generates simulations of different rescue plans based on the tactical recommendations and rescue action options, uses tactical evaluation algorithms to evaluate the rescue plans, and generates a quantitative comparison report.

[0096] The tactical generation module is used to generate multi-terminal collaborative sand table interactive tactics based on the quantitative comparison report, and to conduct mine emergency rescue training based on the multi-terminal collaborative sand table interactive tactics.

[0097] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0098] The scope of protection of this application is not limited to the embodiments described above. Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from the scope and spirit of this disclosure. If such modifications and variations fall within the scope of this disclosure and its equivalents, then the intent of this disclosure also includes these modifications and variations.

Claims

1. A multi-level, remote collaborative command training method for mine emergency rescue, characterized in that, Includes the following steps: Acquire dynamic disaster scenario data in mines, and construct digital twin models and three-dimensional models of disaster evolution for dynamic disaster scenarios in mines; Based on mine emergency rescue regulations and historical case database, tactical suggestions and rescue action options are generated. Using a digital twin model of the dynamic mine disaster scenario, different rescue plans are generated based on the tactical suggestions and rescue action options. The rescue plans are evaluated using a tactical evaluation algorithm, and a quantitative comparison report is generated. Based on the quantitative comparison report, a multi-terminal collaborative sand table interactive tactic was generated, and mine emergency rescue training was conducted based on the multi-terminal collaborative sand table interactive tactic.

2. The multi-level, remote collaborative command training method for mine emergency rescue according to claim 1, characterized in that, Dynamic disaster scenario data in mines includes: a 3D geological model of the mine, a roadway topology map, equipment layout data, and roadway ventilation network parameters; The specific method for constructing a digital twin model of a dynamic disaster scenario in a mine is as follows: Based on the three-dimensional geological model of the mine, the topology map of the roadway, and the equipment layout data, a digital twin sand table that supports real-time rendering of disaster visualization effects is constructed. The location of the fire source is determined, and the fire spread path and temperature distribution map are dynamically calculated based on the principle of fluid dynamics. Combined with the parameters of the roadway ventilation network, a fire spread model is constructed. The location of toxic gas release is determined, and the diffusion range of toxic gas is simulated using the finite element method based on the gas concentration gradient, CO concentration gradient and ventilation parameters to construct a gas diffusion model. Several fire source locations and several toxic gas release locations are set on the digital twin sand table. The fire spread model and gas diffusion model are assembled on the digital twin sand table. Rescue personnel, rescue equipment and virtual sensors are deployed on the digital twin sand table to obtain a digital twin model of the dynamic disaster scenario in the mine. By setting different disaster intensities and environmental parameters, a three-dimensional model of disaster evolution is generated based on a digital twin model of a dynamic disaster scenario in a mine. Disaster intensity parameters are used to quantify the initial state and development trend of a disaster, including: initial ignition temperature, ignition power, type of combustible material, initial fire spread rate, initial concentration of toxic gas release, gas release rate, initial gas diffusion pressure, and initial range of disaster impact. Environmental parameters are the structural or dynamic environmental conditions of mine roadways that affect the evolution of disasters, including: roadway height, width, length, number of roadway corners and branching paths, roadway wind speed, air volume, ventilation fan operating status, ventilation network resistance coefficient, and humidity in the roadway; On the digital twin sandbox, drag and drop to adjust the location of the fire source, set the initial disaster intensity, modify environmental parameters, generate a 3D model of disaster evolution, and drag and drop to deploy rescue personnel, equipment, and virtual sensors. Real-time mapping of resource status to the global view includes the number and allocation of rescue personnel, equipment, and virtual sensors.

3. The multi-level remote collaborative command training method for mine emergency rescue according to claim 2, characterized in that, Tactical suggestions and rescue action options are generated based on mine emergency rescue regulations and a historical case database. The specific method is as follows: Acquire historical cases of mine emergency rescue and build a historical case database; The mine emergency rescue clauses are transformed into a structured decision tree, and key decision nodes are selected by combining historical case databases to generate tactical suggestions and rescue behavior options.

4. The multi-level remote collaborative command training method for mine emergency rescue according to claim 3, characterized in that, Different simulated rescue plans are evaluated using tactical evaluation algorithms to generate a quantitative comparison report. The specific method is as follows: Based on time constraints, different rescue plans are generated using task Gantt charts. The first and second curves of key decision nodes are monitored in real time. The first curve is the time deviation curve generated by the difference between the actual arrival time of the rescue team at the key location and the planned time. The second curve is the completion curve of the key rescue tasks, which include fire fighting, gas emission and personnel search and rescue. A warning is issued when the arrival time deviation of the rescue team and the completion plan of the rescue mission exceed the preset threshold; Using a digital twin model of a dynamic mine disaster scenario, different rescue plans were simulated to obtain simulation results for different rescue plans; The tactical evaluation algorithm is used to evaluate the simulation results of different rescue plans. The tactical evaluation algorithm adopts a multi-dimensional weighted comprehensive evaluation algorithm, including: time efficiency dimension, resource consumption dimension, rescue outcome dimension and risk occurrence dimension. Time efficiency includes the deviation between the total time of the rescue mission and the preset minimum time; Resource consumption includes: the total amount of rescue equipment, personnel, and materials required for the rescue mission; The rescue results include: the survival rate of trapped personnel and the disaster control rate; Risk occurrence includes: the incidence rate of new risk events; The deviation between the total time of the rescue mission and the preset minimum time, the total input of rescue equipment, personnel and materials, the survival rate of trapped people, the disaster control rate and the incidence of new risk events are normalized to make the corresponding values ​​of each dimension uniform. Then, weights are set for each dimension, and the comprehensive score of the rescue plan is calculated by weighted summation. A quantitative comparison report of different rescue plans is generated based on the comprehensive score of the rescue plan and the deviation between the actual value of each dimension and the preset threshold. The report includes execution efficiency score, resource consumption statistics and deviation analysis.

5. A multi-level, remote collaborative command training method for mine emergency rescue according to claim 4, characterized in that, The multi-terminal collaborative sandbox interaction tactic is a multi-level collaborative tactic, including: central cloud, regional edge cloud and field terminal.

6. A multi-level, remote collaborative command training method for mine emergency rescue according to claim 5, characterized in that, Mine emergency rescue training based on multi-terminal collaborative sand table interactive tactics, the specific method is as follows: The command center serves as the central cloud, where a main control engine is deployed to uniformly manage the global disaster evolution 3D model, resource status, and the time-series logic of rescue plans. Based on quantitative comparison reports, rescue tactical instructions are generated and sent to the regional edge cloud. The rescue site is used as a regional edge cloud to receive rescue tactical instructions sent by the central cloud and feed back on on-site execution data. It runs a three-dimensional model of disaster evolution and adjusts the rescue tactical instructions in real time based on real-time environmental data on site. Rescue team members serve as on-site terminals, wearing lightweight client equipment. Through these lightweight clients, they receive real-time adjusted rescue tactical instructions, receive feedback on their real-time location and on-site environmental data, and update the environmental and personnel data in the three-dimensional disaster evolution model in real time.

7. A multi-level, remote collaborative command training method for mine emergency rescue according to claim 6, characterized in that, The command center adopts a heterogeneous intelligent agent collaboration mechanism, which synchronizes the operation events of each key decision node through a distributed message queue; it uses an LSTM time series prediction model based on data from virtual sensors in a digital twin sandbox to predict the evolution trend of the disaster within a preset time and generate early warning information, which is then pushed to various field terminals.

8. A multi-level, remote collaborative command training method for mine emergency rescue according to claim 7, characterized in that, The central cloud also includes a role-based access control model for defining multi-level command roles and operational permissions; The role-based access control (RBAC) permission model defines multi-level command roles and operational permissions. Specifically, it establishes a dynamic permission allocation mechanism. Using the RBAC permission model and dynamic permission allocation mechanism, different roles and permissions are assigned to on-site terminals at different stages of mine emergency rescue training. Multi-level command roles and operational permissions are defined, including command, rescue, and logistics layers. This ensures that the operational permissions and data of personnel at different levels are synchronized during mine emergency rescue training, enabling multi-role remote collaborative operations at the command, rescue, and logistics levels.

9. A multi-level, remote collaborative command training method for mine emergency rescue according to claim 8, characterized in that, Dynamic permission allocation mechanisms include: In the initial phase of mine emergency rescue training, a role-based access control model was used to assign basic permissions according to preset roles, and tactical arrows and warning zone markers were drawn on the digital twin model and shared in real time. During the simulation of mine rescue plans, permissions are escalated based on event triggers. When multiple users modify the status of the same device simultaneously, a role-based access control model is used to arbitrate the validity of the operation according to role priority.

10. A multi-level remote collaborative command training system for mine emergency rescue, based on the method described in claim 1, for multi-level remote collaborative command training, characterized in that... include: Digital twin module, report generation module, and tactical generation module; The digital twin module is used to acquire dynamic disaster scenario data in mines and construct a digital twin model of the dynamic disaster scenario in mines; The report generation module is used to generate tactical recommendations and rescue action options based on mine emergency rescue regulations and historical case databases. The digital twin model generates simulations of different rescue plans based on the tactical recommendations and rescue action options, uses tactical evaluation algorithms to evaluate the rescue plans, and generates a quantitative comparison report. The tactical generation module is used to generate multi-terminal collaborative sand table interactive tactics based on the quantitative comparison report, and to conduct mine emergency rescue training based on the multi-terminal collaborative sand table interactive tactics.

Images