Multi-person collaborative practical training and emergency drilling system and method based on virtual reality

Abstract

The invention belongs to the technical field of virtual reality, and particularly relates to a multi-person collaborative practical training and emergency drilling system and method based on virtual reality. The invention provides a multi-person collaborative practical training and emergency drilling method based on virtual reality. The method comprises the following steps: constructing a virtual mine space according to a production mine mining engineering plan and a roadway profile map; constructing a digital twinborn body of the mining and transporting equipment; obtaining use mode selection information of the user and real-time state data of the user cooperatively operating the mining and transportation equipment physical model, wherein the mode selection information comprises a training mode and an assessment mode; according to the real-time state data, carrying out on-line updating on digital twinborn body data of the mining and transporting equipment; driving the corresponding mining and transporting equipment virtual three-dimensional model in the virtual mine space to execute a corresponding action, and adjusting the virtual mine space according to the executed action; the machine learning algorithm evaluates the cooperative operation state and the emergency operation state of the mining and transporting equipment. According to the invention, collaborative operation training of mining engineering can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of virtual reality, and in particular relates to a multi-person collaborative training and emergency drill system and method based on virtual reality. Background Art

[0002] At present, the mining talent training arranged by colleges and universities or mining companies mainly focuses on technical training for students, trainees and employees, and mainly includes emergency drills in fixed scenarios. It lacks the technical means for multi-job collaborative operation training and emergency drills in underground on-site environments. The training effect is not ideal, and therefore cannot fully meet the needs of enterprises and colleges and universities for employees and students' practice. Summary of the invention

[0003] In view of this, the embodiments of the present invention provide a system and method for multi-person collaborative training and emergency drills based on virtual reality to solve the technical problem that the current multi-type collaborative operation training and emergency drills are quite different from the on-site conditions.

[0004] The technical solution adopted by the present invention is:

[0005] In a first aspect, the present invention provides a multi-person collaborative training and emergency drill method based on virtual reality, the method comprising:

[0006] S1: Construct a virtual mine space based on the production mine excavation engineering plan and tunnel profile;

[0007] S2: Build the digital twin of the mining and transportation equipment based on the physical model of the mining and transportation equipment;

[0008] S3: Acquire the user's usage mode selection information and the real-time status data of the physical model of the mining and transportation equipment operated by the user in collaboration, wherein the mode selection information includes a training mode and an assessment mode;

[0009] S4: updating the digital twin data of the mining and transportation equipment online according to the real-time status data;

[0010] S5: Use the digital twin data to drive the corresponding virtual three-dimensional model of the mining and transportation equipment in the virtual mine space to perform corresponding actions, and adjust the virtual mine space according to the executed actions;

[0011] S6: The machine learning algorithm evaluates the collaborative operation status and emergency operation status of the mining and transportation equipment based on the mode selection information and real-time status data.

[0012] Second aspect, the present invention also provides a multi - person collaborative training and emergency drill method system based on virtual reality, which applies the teaching method described in the first aspect, and includes a physical model of mining, transportation and loading equipment, a cloud server, and a digital twin update module. Sensors are installed on the physical model of the mining, transportation and loading equipment. The digital twin update module is used to update the digital twin according to the data collected by the sensors, and the digital twin update module is communicatively connected to the cloud server.

[0013] Beneficial effects: The multi - person collaborative training and emergency drill method system and method based on virtual reality in the present invention utilize the mining engineering plan and roadway profile of a production mine to construct a virtual mine space, completely replicating complex environments such as underground roadways, equipment layouts, and geological structures, enabling trainees to experience the mine operation scenario as if they were on the spot. Through the virtual mine space, multiple users can cooperate and operate in the same scenario, simulating the processes of multi - type work cooperation in an actual mine, making up for the deficiency of the lack of collaborative operation training in traditional training. Through the digital twin of the mining, transportation and loading equipment, the real - time status data of the physical equipment is mapped into the virtual 3D model. When users operate the equipment in the virtual environment, the digital twin data is updated in real - time, ensuring that the actions of the virtual equipment are completely synchronized with the physical equipment. It can be used for virtual training and also provide rehearsal support for real equipment operation. The flexible training and assessment mode adapts to different needs. The system provides a training mode and an assessment mode: In the training mode, trainees can freely practice equipment operation, collaborative processes, and emergency responses. The assessment mode can automatically evaluate the performance of trainees and generate a quantitative assessment report. The two modes can be flexibly switched. Through machine learning algorithms, the system can evaluate the collaborative operation status and analyze the coordination of multi - user operations. When evaluating the emergency operation status, the virtual environment is used to replace the on - site operation, avoiding equipment damage or personal injury caused by operation errors of trainees in a real mine. Description of the Drawings

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings, and all of these are within the protection scope of the present invention.

[0015] Figure 1 It is a schematic flowchart of the multi - person collaborative training and emergency drill method based on virtual reality of the present invention;

[0016] Figure 2 It is a schematic diagram of the scene in the virtual mine in the training mode from the VR perspective of the present invention;

[0017] Figure 3 It is a schematic diagram of the scene in the virtual mine in the assessment mode from the VR perspective of the present invention;

[0018] Figure 4 Flow diagram of the method for evaluation according to the implementation status data in the present invention

[0019] Figure 5 Flow diagram of the method for training the support vector machine algorithm model in the present invention;

[0020] Figure 6 Flow diagram of the method for evaluation using the support vector machine algorithm in the present invention;

[0021] Figure 7 Flow diagram of the training mode in the present invention;

[0022] Figure 8 Interactive interface diagram for training item selection in the present invention;

[0023] Figure 9 Flow diagram of the assessment mode in the present invention;

[0024] Figure 10 Flow diagram of the exhibition viewing mode in the present invention;

[0025] Figure 11 Structural block diagram of the method for multi - person collaborative training and emergency drill based on virtual reality in the present invention. Detailed implementation manners

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. In the description of the present invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. This is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be construed as a limitation of the present invention. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, the elements defined by the statement "including..." do not exclude the existence of additional identical elements in the process, method, article or device including the said elements. If there is no conflict, the embodiments of the present invention and the various features in the embodiments can be combined with each other, and all are within the protection scope of the present invention.

[0027] Embodiment 1

[0028] As Figure 1 shown, the present invention provides a multi-person collaborative training and emergency drill method based on virtual reality. The method includes:

[0029] S1: Construct a virtual mine space according to the mining engineering plan view and roadway profile of a production mine;

[0030] The mining engineering plan view is a plan layout diagram showing the mining engineering of a mine, including roadways, working faces, etc. The roadway profile is a vertical profile diagram showing the mine roadways, demonstrating the shape, size, and geological structure of the roadways.

[0031] In this step, according to the mining engineering plan view and roadway profile of a production mine, a virtual mine space is constructed using 3D modeling technology. This space includes the terrain, roadways, equipment layout, etc. of the mine, and can truly reflect the actual environment of the mine. By constructing a virtual mine space, users can conduct operation training in a safe environment, which can not only realistically reproduce the real underground operation environment but also avoid the risks of on-site operation.

[0032] S2: Construct a digital twin of the mining, excavation, and transportation equipment based on the physical model of the mining, excavation, and transportation equipment;

[0033] The mining, excavation, and transportation equipment refers to the mechanical equipment used for mining, transporting, and processing resources such as ores and coal in mine excavation operations. These equipment play a crucial role in mine operations, can improve production efficiency, reduce manual labor intensity, and ensure operation safety. The mining, excavation, and transportation equipment generally includes three categories: mining equipment, transportation equipment, and auxiliary equipment.

[0034] Mining equipment is the mechanical equipment used for directly coal mining, usually used for tunneling, cutting, crushing, etc. For example, roadheaders and shearers. Among them, the shearer is used in longwall coal mining faces to cut coal seams through a rotating cutting drum. Transportation equipment, such as scraper conveyors, is used to transport the coal cut by the shearer to the roadway transfer point.

[0035] In mining engineering training, trainees use the physical model of the mining, excavation, and transportation equipment placed in the training site for training. The physical model of the mining, excavation, and transportation equipment is a model made according to the structural and functional characteristics of the actual mining, excavation, and transportation equipment, and can simulate the main functions of the actual mining, excavation, and transportation equipment. The digital twin is the digital representation of the physical model of the mining, excavation, and transportation equipment used by trainees in the virtual space, and can synchronize the status and data of the physical equipment in real time. In this embodiment, 3D modeling can be performed on various physical models of the mining, excavation, and transportation equipment as needed to obtain the corresponding virtual models of the mining, excavation, and transportation equipment. This virtual model can be used as the digital twin of the mining, excavation, and transportation equipment. Through the digital twin, users can operate the equipment in a virtual environment, observe the running status of the equipment in real time, and improve the operation proficiency.

[0036] S3: Obtain the user's usage mode selection information and the real-time status data of the user's collaborative operation of the physical model of the mining, excavation, and transportation equipment, where the mode selection information includes a training mode and an assessment mode;

[0037] The system obtains the usage mode (training mode or assessment mode) selected by the user and the real-time status data of the user's collaborative operation of the physical model of the mining, excavation, and transportation equipment. The training mode is used for learning and practice, and the assessment mode is used for evaluating the user's operation ability.

[0038] Among them, the training mode: In this mode, the user conducts learning and practice, and the system provides operation guidance and feedback, as Figure 2 shown,

[0039] The assessment mode: In this mode, the user conducts an operation assessment, and the system evaluates and grades according to the operation results, as Figure 3 shown.

[0040] Among them, the training mode includes teaching the centralized control one-key startup method, electro-hydraulic control operation method, shearer operation method, centralized control center operation method, typical fault troubleshooting method, emergency operation procedures and other operation methods, and can generate student learning record sheets. The exam mode includes single-post assessment and team assessment. During team assessment, a collaborative operation of a team within a set time is set. According to the operation methods of personnel at each position to handle problems, the overall accuracy rate of each position and the team is statistically judged to give scores, and student assessment score records can be formed.

[0041] Real-time status data is the real-time data generated by users when operating equipment, such as the location, speed, operating status, etc. of the equipment.

[0042] Through mode selection, the system can provide different operation scenarios according to the needs of users to meet different needs of training and assessment.

[0043] The physical model of the mining, excavation and transportation equipment in this embodiment is an actual teaching and training equipment made by imitating the structure and function of the mining, excavation and transportation equipment, and has basically the same structure and function as the actual mining, excavation and transportation equipment operating in a real mine.

[0044] S4: Online update the digital twin data of the mining, excavation and transportation equipment according to the real-time model parameters; this step uses the model parameters of the physical model of the mining, excavation and transportation equipment collected in real time to update the digital twin, so that the model parameters of the digital twin are consistent with those of the physical model, so that the state of the digital twin is consistent with the state of the physical model of the mining, excavation and transportation equipment. In this way, the state of the digital twin can accurately and timely reflect the operation of the physical model of the mining, excavation and transportation equipment by the trainees in various teaching scenarios. By real-time updating the digital twin data, it is ensured that the equipment state in the virtual environment is synchronized with the physical equipment, improving the accuracy and real-time performance of the operation.

[0045] S5: Use the digital twin data to drive the corresponding virtual three-dimensional model of the mining, excavation and transportation equipment in the virtual mine space to perform corresponding actions, and adjust the virtual mine space according to the executed actions;

[0046] Using the updated digital twin data, drive the virtual three-dimensional model of the mining, excavation and transportation equipment in the virtual mine space to perform corresponding actions, such as moving, rotating, excavating, etc. When the mining, excavation and transportation equipment performs actions in the real mine space, corresponding changes will also occur in the real mine space. For example, after the excavation equipment performs an excavation action, the wall thickness of the virtual mine space will be reduced accordingly. For this, this application also adjusts the virtual space according to the corresponding actions performed by the virtual three-dimensional model of the mining, excavation and transportation equipment. For example, after the virtual three-dimensional model of the excavation equipment performs an excavation action, adjust the wall thickness in the virtual space, so that the virtual space adapts to the training progress of the trainees.

[0047] In this embodiment, the virtual 3D model of the initial mining and haulage equipment can be imported into the virtual mine space generated in the previous steps in advance. Then, after the digital twin is updated, the data of the updated digital twin is used in a timely manner to update the virtual 3D model of the mining and haulage equipment in the virtual mine space, so that the state of the virtual 3D model of the mining and haulage equipment in the virtual mine space is synchronized with the state of the physical model of the mining and haulage equipment operated by the user. In this way, the user can experience the effects generated after operating the mining and haulage equipment in the virtual mine.

[0048] In this step, the virtual model is updated in real time through the sensor data of the mining and haulage physical model (such as the inclination angle of the hydraulic support and the transportation speed of the scraper conveyor), ensuring the synchronization of the virtual scene and the actions of the physical equipment. By driving the virtual equipment with digital twin data, the user can operate the equipment in real time in the virtual environment, enhancing the immersion and realism of the operation.

[0049] To ensure realism, it is necessary to improve the real-time update of the virtual 3D model. However, when the amount of updated data is large, there is a large amount of updated data to be processed, the data transmission time is long, and the update speed is slow. In this regard, for the update of the virtual 3D model of the hydraulic support, in this embodiment, the step S5: using digital twin data to drive the corresponding virtual 3D model of the mining and haulage equipment in the virtual mine space to perform corresponding actions further includes:

[0050] Obtain the local change set of the virtual 3D model when the hydraulic support's canopy board performs actions based on the current digital twin data and the previous digital twin data of the hydraulic support;

[0051] During the process of the hydraulic support's canopy board unfolding and retracting, parts such as the base, roof, and shield remain unchanged, while the length of a part of the canopy board will stretch and another part will rotate. Since the speed of the canopy board's length stretching and the angular velocity of rotation do not change suddenly, the speed of the canopy board's elongation and the angular velocity of rotation can be estimated based on the current digital twin data and the previous digital twin data. Thus, the possible states of the hydraulic support's canopy board at the arrival of the updated time node can be estimated. The set of these possible states is used as the local change set of the virtual 3D model when the hydraulic support's canopy board performs actions.

[0052] Send the shared iterative data of the parts related to the canopy board in the local change set to the server according to the local change collection;

[0053] Although there are various states of the hydraulic support rib protection plate within the local change set, when updating the virtual 3D model according to these states, some of the data used for updating is common to all these states. The update data common to each of the aforementioned overall entities is the shared iterative data. For example, during the telescoping process of the rib protection plate, the data of the telescoping part of the rib protection plate determined at the minimum elongation speed can be used as the shared iterative data and sent to the server first.

[0054] When reaching the preset update time node, use the shared iterative data to replace the corresponding data in the virtual 3D model of the hydraulic support;

[0055] If the shared iterative data has been received before the update time node arrives, it is necessary to wait for the update time node to arrive before using the data for update. If the shared iterative data is received after the update time node arrives, the update can be performed simultaneously when it is received.

[0056] Execute the shared iterative data update process, and the shared iterative data update process includes:

[0057] Update the current digital twin data;

[0058] In this step, continuously obtain the latest digital twin data

[0059] Update the local change set according to the updated digital twin data and the previous digital twin data;

[0060] In this step, obtain the latest local change set according to the new digital twin data

[0061] Obtain the newly added shared iterative data according to the updated local change set, and send the newly added shared iterative data to the server;

[0062] As the rib protection plate executes actions, the uncertainty of the final state of the rib protection plate becomes smaller and smaller, so new shared iterative data will continuously appear.

[0063] When the new shared iterative data is determined, it is sent to the server for update.

[0064] When reaching the new preset update time node, use the increased shared iterative data to replace the corresponding data in the virtual 3D model of the hydraulic support;

[0065] Judge whether to send redundant data according to the latest local change set and all the shared iterative data that has been sent;

[0066] Among them, redundant data refers to the data required for iterative replacement in all possible states of the remaining part of the virtual 3D model for receiving updated data. The more data has been updated and iterated, the less redundant data there is. When the amount of redundant data can meet the requirements of completing sending and receiving before the update time node arrives, all the redundant data can be sent to the server.

[0067] If so, send the redundant data to the server. The server extracts the corresponding data from the redundant data according to the final state of the canopy guard to replace the corresponding data in the virtual 3D model of the hydraulic support. If not, repeat the process of sharing and iterating data updates. When the final state of the canopy guard is determined, the corresponding part of the data corresponding to the final state of the canopy guard can be selected to replace the corresponding data in the virtual 3D model of the hydraulic support.

[0068] This step preprocesses the data used for local model replacement during the operation of the canopy guard of the hydraulic support. Before the update node arrives, the update module can determine a part of the shared data and transmit it to the server, which not only saves the server computing power but also improves the efficiency of updating the 3D virtual model.

[0069] S6: The machine learning algorithm evaluates the collaborative operation state and emergency operation state of the mining, excavation, and transportation equipment based on the mode selection information and real-time state data.

[0070] Among them, the collaborative operation state refers to the state when multiple users operate the equipment collaboratively, which can be the coordination of operations, the sequence of processes, efficiency, etc.

[0071] Emergency operation state: It refers to the state when multiple users operate the equipment collaboratively in an emergency scenario, which can reflect the emergency response ability.

[0072] The machine learning algorithm evaluates the collaborative operation state and emergency operation state of the mining, excavation, and transportation equipment based on the usage mode (training mode or assessment mode) selected by the user and the real-time state data. The evaluation results can be used to improve the operation process or provide feedback.

[0073] This embodiment constructs a virtual mine space based on the mining engineering plan and roadway profile diagram of a production mine to fully reproduce complex environments such as underground roadways, equipment layouts, and geological structures, enabling trainees to experience the mine operation scenario as if they were on the spot.

[0074] This embodiment supports multiple users (such as coal miners, support workers, safety officers, etc.) to cooperate and operate in a virtual environment, simulating the processes of multi-disciplinary cooperation in an actual mine (such as tunneling, coal mining, support, etc.), and making up for the defect of the lack of cooperative operation training in traditional training. Through the digital twins of mining, hauling and conveying equipment, the real-time status data of physical equipment (such as position, speed, operating parameters) are mapped into a virtual 3D model. When users operate the equipment in the virtual environment, the digital twin data are updated in real time to ensure that the actions of the virtual equipment are completely synchronized with the physical equipment, which can be used for virtual training and also provide rehearsal support for real equipment operation. The flexible training and assessment modes are adapted to different needs. Since the system provides a training mode and an assessment mode, in the training mode, trainees can freely practice equipment operation, cooperation processes and emergency responses, and the system provides real-time operation guidance and error prompts (such as equipment overlimit alarms). In the assessment mode, the system automatically evaluates the performance of trainees according to preset scoring criteria (such as operation standardization, cooperation efficiency, emergency response time) and generates a quantitative assessment report. The two modes can be flexibly switched to meet the full-process needs from basic training to skill certification.

[0075] Through machine learning algorithms, the system can evaluate the cooperative operation status, analyze the coordination of multi-user operations (such as equipment collision risks, task connection efficiency), and identify bottleneck problems in cooperative operations. Evaluate the emergency operation status: In simulated emergency scenarios such as gas leakage and mine pressure, analyze the response speed, disposal process standardization and teamwork ability of trainees. Based on historical operation data, provide data support for training course design and emergency plan improvement. The virtual environment replaces on-site operation, avoiding equipment damage or personal injuries caused by operation errors of trainees in a real mine. Through high-fidelity emergency drills (such as fire escape, gas outburst disposal), improve the emergency psychological quality and disposal ability of trainees in real scenarios and reduce accident risks in actual operations.

[0076] As Figure 4 shown, in this embodiment, step S6: The machine learning algorithm evaluates the cooperative operation status and emergency operation status of the mining, hauling and conveying equipment according to the mode selection information and real-time status data, including:

[0077] S61: Obtain the real-time status data of the physical model of the mining, hauling and conveying equipment during the process of trainees' cooperative operation of the physical model of the mining, hauling and conveying equipment;

[0078] In this embodiment, the obtained real-time model status data of the coal equipment physical model include but are not limited to data reflecting the height of the hydraulic support column, the angle of the hydraulic support rib protection plate, the position of the hydraulic support, the angle of the shearer rocker arm, the position of the shearer, the shearer traction speed, and the speed of the belt conveyor.

[0079] In this embodiment, sensors can be installed at the set positions of each physical model of the mining, excavation, and transportation equipment. These sensors are used to collect data reflecting the real-time state of the physical model so as to accurately obtain the real-time state of the physical model of the mining, excavation, and transportation equipment during the user training or assessment process. For example, for a hydraulic support, displacement sensors and pressure sensors can be installed on the columns, angle sensors can be installed at the hinge of the rib protection plate, displacement sensors can be installed on the front cantilever, and displacement sensors can be installed on the jacks, etc.; for a shearer, angle sensors can be installed on the rocker arm, and rotational speed sensors can be installed on the drum. Position sensors can also be installed on the scraper conveyor. The real-time state data refers to the parameters dynamically generated during the operation of the equipment, which can reflect the accuracy, efficiency, and safety of the current operation. By comprehensively collecting multi-dimensional data, it provides an objective basis for subsequent evaluation, avoids the limitations of relying on manual observation in traditional training, and ensures that the evaluation results truly reflect the actual operation level of the trainees.

[0080] S62: Obtain the support vector machine algorithm model trained with data;

[0081] Among them, the support vector machine (SVM) is a supervised learning algorithm. This algorithm realizes the classification of high-dimensional data by finding the optimal classification hyperplane, and is especially suitable for small-sample and non-linear data scenarios. Before using the support vector machine algorithm model for evaluation and analysis, in this step, the support vector machine algorithm model is first trained with the data collected by the sensors, and the parameters in the support vector machine algorithm model are optimized through data training.

[0082] S63: The support vector machine algorithm model evaluates the collaborative operation state and emergency operation of the mining, excavation, and transportation equipment according to the input real-time state data of the mining, excavation, and transportation equipment.

[0083] Among them, the evaluation of the collaborative operation state of the mining, excavation, and transportation equipment can include analyzing the multi-equipment collaborative efficiency (such as whether the material connection between the transport vehicle and the roadheader is smooth) and detecting collaborative conflicts (such as whether there is a risk of collision in the equipment movement trajectory).

[0084] Among them, the evaluation of the emergency operation state includes: judging the emergency response speed (such as whether the ventilation equipment starts in time after a gas alarm). Evaluating the operation standardization (whether the cutting arm returns to the safe position during emergency shutdown).

[0085] S63: The support vector machine algorithm model outputs an evaluation of the trainee's coal mining operation based on the input model parameters. After obtaining the data reflecting the model parameters detected by the sensors, these data are input into the trained machine learning algorithm model, and the machine learning algorithm outputs an evaluation result of the trainee's coal mining operation according to the input data. The model parameters can indirectly reflect the trainee's operation of the mining, excavation, and transportation equipment. However, the evaluation of whether the trainee's operation meets the standard operation requirements is affected by multiple model parameters, and different model parameters also affect each other. Therefore, there is a relatively complex non-linear relationship between the model parameters and the evaluation results. In this embodiment, the support vector machine algorithm in the machine learning algorithm is used to process the data collected by the sensors to obtain the evaluation result of the trainee's coal mining operation.

[0086] The SVM algorithm evaluates through data-driven, eliminating the subjectivity of traditional manual scoring and ensuring the fairness and consistency of the assessment results. The SVM's ability to process non-linear and high-dimensional data enables it to effectively handle the evaluation of the operation state with multi-variable coupling in the mine environment (such as the comprehensive impact of the various states of the shearer and the various states of the hydraulic support on equipment operation). The evaluation results are fed back to the training system, which can dynamically adjust the training difficulty (such as increasing the complexity of emergency scenarios) to achieve personalized training path design. By accurately identifying potential risks in collaborative operations (such as equipment collisions and emergency response delays), simulating the consequences of accidents in the virtual environment in advance, and strengthening the trainee's risk prediction ability.

[0087] As Figure 5 shown, in this embodiment, the S62: Obtaining the support vector machine algorithm model trained with data includes:

[0088] S621: Using sensors installed on the mining, excavation, and transportation equipment to obtain the state data of the mining, excavation, and transportation equipment during the user's operation of the mining, excavation, and transportation equipment;

[0089] S622: Extracting the equipment collaborative operation state data set and the emergency operation data set from the state data;

[0090] This embodiment can analyze and evaluate the training effect from two aspects: equipment collaborative operation status and emergency operation. For this, this embodiment extracts the equipment collaborative operation state data set for evaluating the equipment collaborative operation status and the emergency operation data set for evaluating the emergency operation from the collected state data.

[0091] S623: Obtaining the initial model of equipment collaborative operation state and the initial model of emergency operation, and both the initial model of equipment collaborative operation state and the initial model of emergency operation adopt the support vector machine algorithm model;

[0092] S624: Annotate the data in the device collaborative operation status dataset according to the evaluation result corresponding to the device collaborative operation status dataset;

[0093] When this step is specifically implemented, the operator can be allowed to perform collaborative operations on the mining, excavation, and transportation equipment, and then obtain the data related to the status of the mining, excavation, and transportation equipment detected by the sensors during the operation, and manually determine whether the operation meets the corresponding requirements according to the operation specifications for the collaborative operation of the mining, excavation, and transportation equipment. This judgment result is used as the evaluation result corresponding to the device collaborative operation status dataset.

[0094] S625: Annotate the data in the emergency operation dataset according to the evaluation result corresponding to the emergency operation dataset;

[0095] When this step is specifically implemented, the operator can be allowed to perform collaborative operations on the mining, excavation, and transportation equipment in case of emergency, and then obtain the data related to the emergency operation status detected by the sensors during the operation, and manually determine whether the operation meets the corresponding requirements according to the operation specifications for the emergency operation. This judgment result is used as the evaluation result corresponding to the emergency operation dataset.

[0096] S626: After training the initial model of the device collaborative operation status with the annotated device collaborative operation status data, obtain the device collaborative operation status evaluation algorithm model;

[0097] S627: After training the initial emergency operation model with the annotated emergency operation data, obtain the emergency operation evaluation algorithm model.

[0098] In this step, the support vector machine algorithm is trained with the annotated dataset to continuously optimize the parameters of the model until the evaluation effect of the machine learning algorithm model meets the requirements.

[0099] As Figure 6 shown, in this embodiment, S63: The support vector machine algorithm model evaluates the collaborative operation status and emergency operation status of the mining, excavation, and transportation equipment according to the input real-time status data of the mining, excavation, and transportation equipment, including:

[0100] S630: Use the sensors installed on the mining, excavation, and transportation equipment to obtain the real-time status data of the mining, excavation, and transportation equipment during the process of the user operating the mining, excavation, and transportation equipment;

[0101] S631: Extract the real-time device collaborative operation status dataset and the real-time emergency operation dataset from the real-time status data;

[0102] S632: Obtain the data category labels corresponding to each data in the device collaborative operation status dataset;

[0103] In this step, the data category labels include the positive class (represented by the value +1) and the negative class (represented by the value -1).

[0104] The method for obtaining the data category labels corresponding to each data in the device collaborative operation status dataset can adopt the threshold method, that is, set a threshold according to the safety operation specifications of the device, and directly judge the category through the sensor data. For example, if the data of the pressure sensor on the hydraulic support column exceeds the rated range, it is marked as the negative class; if it does not exceed the rated range, it is marked as the positive class. In addition, the sensor data can also be input into the deep learning algorithm model, and the deep learning algorithm model automatically outputs the data category labels corresponding to the sensor data.

[0105] S633: Input the data and the corresponding data category labels in the operation status dataset into the device collaborative operation status evaluation algorithm model;

[0106] S634: The device collaborative operation status evaluation algorithm model outputs the evaluation results of the corresponding excavation, transportation, and equipment collaborative operation status according to the input device collaborative operation status data, which specifically includes the following steps:

[0107] Obtain the support vectors corresponding to the device collaborative operation status data;

[0108] Calculate the kernel function values of the support vectors;

[0109] Multiply each kernel function value by the corresponding Lagrange multiplier and class label (multiply by +1 if the class label is the positive class, and multiply by -1 if the class label is the negative class), and then add up all the multiplied results;

[0110] Add a bias term to the added result to obtain the calculation result.

[0111] The evaluation results of the excavation, transportation, and equipment collaborative operation status can include whether the material transfer time between the shearer and the transportation equipment is accurate, and whether the ventilation is started synchronously when the shearer is cutting, etc.

[0112] S635: Obtain the data category labels corresponding to each data in the emergency operation dataset;

[0113] Similarly, the data category labels of the emergency operation data include the positive class (represented by the value +1) and the negative class (represented by the value -1).

[0114] The method for obtaining the data category labels corresponding to each data in the emergency operation data can adopt the threshold method, that is, set a threshold according to the safety operation specifications of the device, and directly judge the category through the sensor data. For example, if the support resistance of the hydraulic support does not reach the set value, it is marked as the negative class; if the support resistance of the hydraulic support reaches the set value, it is marked as the positive class.

[0115] In addition, the sensor data can also be input into a deep learning algorithm model, and the data category label corresponding to the sensor data can be automatically output by the deep learning algorithm model.

[0116] S636: Input the data and the corresponding data category labels in the emergency operation dataset into an emergency operation evaluation algorithm model;

[0117] S637: The emergency operation evaluation algorithm model outputs the evaluation result of the corresponding emergency operation according to the input emergency operation data, which specifically includes the following steps:

[0118] Obtain the support vector corresponding to the emergency operation data;

[0119] Calculate the kernel function value of the support vector;

[0120] Multiply each kernel function value by the corresponding Lagrange multiplier and class label (if the class label is a positive class, multiply by +1; if the class label is a negative class, multiply by -1), and then add up all the multiplied results;

[0121] Add a bias term to the added result to obtain the calculation result.

[0122] Among them, the evaluation result of the emergency operation can include: whether the emergency stop button is pressed immediately after the alarm is triggered. Whether the cutting arm returns to a safe position after the emergency stop. Whether the cutting head position avoids dangerous areas (such as interfering with the support).

[0123] Such as Figure 7 As shown, in this embodiment, when the obtained mode selection information is the training mode, the method further includes:

[0124] S71: Obtain the training items selected by the user;

[0125] The user can select whether to use the training mode or the assessment mode through the system. After the user makes a selection to the system, the system will generate mode selection information. If the generated mode selection information is the training mode, the system can also let the user select the current training item. Such as Figure 8 As shown, the user can select a specific training item through a virtual reality interaction interface (such as gesture selection, voice command, or menu click).

[0126] S72: Play the corresponding training video according to the training item;

[0127] The training video demonstrates and explains in detail the specific operation methods of the training item. For example, it explains operation methods such as the one-key start method of the centralized control, the electro-hydraulic control operation method, the shearer operation method, the centralized control center operation method, the typical fault troubleshooting method, and the emergency operation regulations. The trainees can select training items according to their own skill deficiencies, achieve training on demand, and improve learning efficiency.

[0128] S73: Generate a learning record sheet based on the collaborative operation status and emergency operation status of the mining, excavation, and transportation equipment.

[0129] During the training process, the system records the operation data of the trainees in real time (such as equipment parameters, collaborative action timings) and the evaluation results of the machine learning algorithm, and automatically generates a structured learning record sheet. The content of the record sheet includes: Summary of operation data: such as task completion time, equipment utilization rate, number of incorrect operations. Evaluation results: Collaborative efficiency score based on the SVM algorithm, emergency response level (such as "excellent", "needs improvement"). Improvement suggestions: Targeted suggestions generated based on error patterns (such as "shorten the loading and unloading interval time of the transport vehicle", "strengthen the operation training of roof support"). The record sheet supports export for trainees to review or for instructors to audit.

[0130] When the obtained mode selection information is the training mode, the method further includes:

[0131] Construct a group collaboration ability portrait based on the historical performance data of the group, including:

[0132] Collect data from multiple past trainings of the group, including the overall operation accuracy rate of the group, group task completion rate, equipment operation proficiency, collaborative time deviation, etc.;

[0133] Compare with historical excellent groups through the clustering algorithm to determine the assistance ability level of this group.

[0134] Input the group collaboration ability portrait and the spatial environmental status of the virtual mine space into the deep Q-network to output the optimal fault injection parameters, where the fault injection parameters include the fault type and the fault generation timing;

[0135] To improve the realism of the training, the environmental changes in the real mine space can be simulated in the virtual mine space, such as roof subsidence, rising gas concentration, etc. Different environmental changes are likely to cause different equipment failures. Different levels of group collaboration ability can handle different faults. This step uses the deep Q-network to generate the fault type and the fault generation timing that best match the current group ability and the spatial environmental status of the virtual mine space according to the group collaboration ability portrait and the spatial environmental status of the virtual mine space.

[0136] Inject the corresponding fault type into the virtual mine space when the fault generation timing condition is satisfied according to the fault injection parameters; for example, if the group's roof support collaboration ability is intermediate, a hydraulic support fault can be injected when the roof pressure is too high.

[0137] Update the virtual 3D models of each mining, excavation, and transportation equipment to the corresponding state of the fault type according to the fault type.

[0138] In different fault types, the mining and haulage equipment has different states. For example, when a hydraulic support fails, the height of the hydraulic support decreases and the supporting force decreases. This step updates the virtual 3D model to the state when the corresponding fault type occurs.

[0139] Determine the execution limit parameters of the virtual 3D model according to the fault type and the updated virtual 3D model;

[0140] After a fault occurs, the performance of the mining and haulage equipment is affected, so the execution actions are restricted, and the restricted situations of the execution actions are different under different fault types. The parameters that restrict the execution actions of the mining and haulage equipment are the execution limit parameters. For example, the hydraulic system pressure of the support is restricted to a certain pressure or below. The output power of the hydraulic support is restricted to a certain value or below.

[0141] Determine the states of the relevant components of the corresponding actual physical model according to the fault type and the updated virtual 3D model;

[0142] Update the actual physical model according to the states of the relevant components of the actual physical model;

[0143] This step reversely updates the virtual 3D model in the case of a fault to the actual physical model, so that the actual physical model is also in the corresponding state under the fault.

[0144] Generate a limit instruction for the actual physical model according to the fault type and the execution limit parameters of the virtual 3D model.

[0145] For example, a limit instruction can be generated according to the execution limit parameters to keep the hydraulic system pressure of the hydraulic support below a set value.

[0146] This embodiment can reasonably inject equipment faults according to the team ability and environmental conditions, which not only makes the training process of equipment fault handling closer to the actual situation, but also makes the fault injection adapt to the team ability, so that the team assistance ability can be quickly improved through targeted training.

[0147] As Figure 9 shown, when the obtained mode selection information is the assessment mode, the method further includes:

[0148] S74: Obtain the evaluation results of the collaborative operation state and emergency operation state of the mining and haulage equipment;

[0149] S75: Correlate each evaluation result with each user according to the physical model of the mining and haulage equipment operated by the user;

[0150] For example, if an evaluation result is related to the state of a physical model of a mining and haulage equipment, find the user who operates the physical model of the mining and haulage equipment as the user corresponding to the evaluation result.

[0151] S76: Generate single-post assessment score record data based on the evaluation results and the corresponding relationship of users;

[0152] In the assessment mode, the system, based on the Support Vector Machine (SVM) algorithm, real-time obtains the evaluation results of the collaborative operation status (such as collaborative efficiency score, number of conflicts) and the emergency operation status (such as response time, handling standardization score) of trainees when operating mining, excavation, and transportation equipment.

[0153] S77: Obtain the group information for team assessment;

[0154] In team assessment, users are divided into teams. The group information for team assessment in this step includes which trained users are in each team. Specifically, during implementation, the system can retrieve the group information for team assessment from a preset team database, including the list of team members, team types (such as mining teams, transportation teams), and assessment task assignments (such as collaborative tunneling tasks, joint emergency drills). The grouping information can also be manually configured by the administrator or automatically matched (such as according to the department and skill level of the trainees).

[0155] S78: Obtain the evaluation results corresponding to each group of users according to the group information for team assessment as the set of evaluation results corresponding to the team;

[0156] This step obtains the evaluation results corresponding to all users in each team.

[0157] S79: Generate team assessment score record data according to the set of evaluation results corresponding to the team.

[0158] This step statistically calculates the scores of all users in the same team to obtain the team assessment score and generates a record sheet for output. The system can also generate team assessment score record data based on the set of team evaluation results, including: Team comprehensive score: such as collaborative task completion rate, emergency drill passing rate. Member contribution analysis: such as key operation contributors, collaborative bottleneck positions. Team improvement suggestions: such as optimizing team division of labor, strengthening cross-post collaboration training.

[0159] Scientific and effective training requires a step-by-step approach. Although the current evaluation algorithm may automatically evaluate team performance, it cannot meet the requirement of step-by-step training. In this embodiment, the S6: The Support Vector Machine algorithm model evaluates the collaborative operation status and emergency operations of mining, excavation, and transportation equipment according to the input real-time status data of mining, excavation, and transportation equipment includes:

[0160] Obtain the evaluation indicators of team performance;

[0161] Among them, the evaluation indicators can be set according to needs. For example, the evaluation indicators can adopt:

[0162] Task completion rate: The percentage of task progress (e.g., the number of completed steps / the total number of steps).

[0163] Operation accuracy: The ratio of the number of incorrect operations to the total number of operations.

[0164] Collaboration efficiency: The timing consistency of device collaborative operations (the time synchronization of multi-device actions can be calculated through the DTW algorithm).

[0165] Collaboration efficiency: The time interval from the occurrence of an abnormal event to the team's response.

[0166] Communication effectiveness: The response delay of voice commands and the accuracy rate of command execution.

[0167] Each evaluation index can be quantified with corresponding scores.

[0168] Determine the team performance score according to the evaluation indexes of the team performance;

[0169] For example, four indexes such as task completion rate, operation accuracy, collaboration efficiency, and emergency response speed can be selected as the scoring basis. At this time, the team performance score P = w1 · task completion rate score + w2 · operation accuracy score + w3 · collaboration efficiency score + w4 · emergency response speed score, where w1 to w4 are the weights assigned to the task completion rate, operation accuracy, collaboration efficiency, and emergency response speed respectively.

[0170] Obtain the historical data of the team performance score and the optimal classification boundary of the support vector machine corresponding to the team performance score according to the classification comprehensive target value of the support vector machine algorithm model;

[0171] In this step, the team performance can be scored during the previous experiment process, and the optimal classification boundary of the support vector machine corresponding to each score can be found. For example, the optimal classification boundary can be selected by minimizing the classification comprehensive target value M, where M = r1 · classification accuracy rate + r2 · classification interval + r3 · classification error rate, and r1 to r3 are the weights assigned to the classification accuracy rate + r2 · classification interval + r3 · classification error rate respectively.

[0172] When the change rate of the sliding window of the team performance score exceeds the threshold, trigger the update of the classification boundary;

[0173] This step starts the update of the classification boundary when the team performance score changes rapidly. The threshold can be set according to experience.

[0174] When the update of the classification boundary is triggered, obtain the current team performance score according to the real-time status data of the mining, excavation, and transportation equipment;

[0175] Update the kernel function parameters of the support vector machine algorithm model according to the current team performance score and the corresponding relationship; in this step, the kernel function parameters corresponding to the optimal classification boundary corresponding to the current team score are predicted according to the foregoing corresponding relationship, such as the scale parameter in the kernel function.

[0176] Adjust the classification boundary of the support vector machine algorithm model according to the updated kernel function parameters.

[0177] After the kernel function parameters are updated, the classification boundary of the support vector machine algorithm model is also updated accordingly.

[0178] In this embodiment, by updating the classification boundary of the evaluation model in real time according to the team performance score during the assessment process, the evaluation model can adapt to different performance situations. It can relax the classification boundary when the score is low to reduce the classification error rate, and tighten the classification boundary when the score is high to strengthen the assessment of operation accuracy, making the difficulty of the assessment adapt to the team performance, thus conforming to the scientific training law from easy to difficult and step by step. In this step, the relationship between the team performance score and the optimal classification boundary is first determined by optimizing historical data and classification comprehensive objectives, and then the classification boundary is quickly updated according to the corresponding relationship during the assessment process, improving the update response speed while ensuring accuracy.

[0179] Such as Figure 10 shown, this embodiment also supports the exhibition viewing mode, which specifically includes the following steps:

[0180] S81: Detect the exhibition viewing mode request information;

[0181] The user can send the exhibition viewing mode request information to the system through the operation interface, and the system then detects the exhibition viewing mode request information.

[0182] S82: Determine the terminal display device for display according to the request information when the exhibition viewing mode request information is detected;

[0183] S83: Send the image information of the training process in the virtual mine to the terminal device for display.

[0184] The image information of the training process allows students and employees to display the operation processes of all collaborative personnel through the display screen, and view the error points in each process through the playback function, so as to carry out targeted practical training operations and exercise the collaborative operation skills of the team.

[0185] In this embodiment, the method further includes:

[0186] S91: Generate control instructions based on the actions performed by the virtual 3D model of the mining, excavation, and transportation equipment in the virtual mine space; in the virtual mine space, when a trainee or operator drives the virtual 3D model of the mining, excavation, and transportation equipment to perform actions (such as shearer cutting, support movement) through the digital twin, the system converts the virtual actions into control instructions that can be recognized by physical equipment.

[0187] S92: Send the control instructions to the corresponding mining, excavation, and transportation equipment in the actual mine;

[0188] S93: The mining, excavation, and transportation equipment performs coal mining operations according to the control instructions.

[0189] After receiving and parsing the control instructions, the mining, excavation, and transportation equipment (such as roadheaders, belt conveyors) in the actual mine drives the mechanical actuators to complete corresponding actions through the embedded control system. For example, the shearer automatically cuts the coal seam along the path planned in the virtual environment; during the execution process, the equipment sensors continuously collect status data (such as motor temperature, hydraulic pressure) and feedback it to the virtual mine space to form a closed-loop control. In this embodiment, the remote control instructions generated by the virtual mine space are used to drive the equipment, enabling key operations (such as shutting down equipment in dangerous areas, selecting escape routes) to be performed in extreme environments such as fires and gas leaks, maximizing personnel safety.

[0190] Embodiment 2

[0191] As Figure 11 shown, this embodiment provides a multi-person collaborative training and emergency drill method and system based on virtual reality, which applies the teaching method described in Embodiment 1. The system includes a physical model of mining, excavation, and transportation equipment, a cloud server, and a digital twin update module. Sensors are installed on the physical model of the mining, excavation, and transportation equipment. The digital twin update module is used to update the digital twin according to the data collected by the sensors, and the digital twin update module is communicatively connected to the cloud server.

[0192] It further includes VR glasses, which are communicatively connected to the cloud server. The VR glasses receive image data sent by the cloud server for displaying the virtual mine space. It also includes a simulation display terminal, which is communicatively connected to the cloud server. The display terminal is used to receive image information sent by the cloud server for displaying the training process in the virtual mine. During the learning and training process using this system, trainees view the virtual space generated by the cloud server through the worn VR glasses. The simulation walking platform is used to simulate the inclination state of the roadway in a steeply inclined coal seam, so that the trainee's body feeling is consistent with the visual effect of the virtual scene. In the spectator mode, the display terminal can display the operation process in the virtual mine during the training process to the user.

[0193] The above is a detailed introduction to a multi-person collaborative training and emergency drill method, system and method provided by an embodiment of the present invention based on virtual reality.

[0194] It should be clear that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, the detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications and additions, or change the order between steps after understanding the spirit of the present invention.

[0195] The functional blocks shown in the above structure block diagrams can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, and so on. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted through a data signal carried in a carrier wave on a transmission medium or a communication link. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical discs, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segment can be downloaded via a computer network such as the Internet, an intranet, and so on.

[0196] It should also be noted that in the exemplary embodiments mentioned in the present invention, some methods or systems are described based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, or different from the order in the embodiments, or several steps can be executed simultaneously.

[0197] As described above, the above is only the specific implementation manner of the present invention. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, modules and units can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A multi-person collaborative training and emergency drill method based on virtual reality, characterized in that: The method comprises: S1: Construct a virtual mine space based on the production mine excavation engineering plan and tunnel profile; S2: Build the digital twin of the mining and transportation equipment based on the physical model of the mining and transportation equipment; S3: Acquire the user's usage mode selection information and the real-time status data of the physical model of the mining and transportation equipment operated by the user in collaboration, wherein the mode selection information includes a training mode and an assessment mode; S4: updating the digital twin data of the mining and transportation equipment online according to the real-time status data; S5: Use the digital twin data to drive the corresponding virtual three-dimensional model of the mining and transportation equipment in the virtual mine space to perform corresponding actions, and adjust the virtual mine space according to the executed actions; S6: The machine learning algorithm evaluates the collaborative operation status and emergency operation status of the mining and transportation equipment based on the mode selection information and real-time status data.

2. The method for multi-person collaborative training and emergency drill based on virtual reality according to claim 1 is characterized in that: When the acquired mode selection information is the assessment mode, the method further includes: S74: Obtaining evaluation results of the coordinated operation status and emergency operation status of the mining and transportation equipment; S75: corresponding each evaluation result to each user according to the physical model of the mining and transportation equipment operated by the user; S76: Generate single-position assessment result record data according to the corresponding relationship between the assessment result and the user; S77: Obtain team assessment grouping information; S78: Obtaining the evaluation results corresponding to each group of users according to the team assessment grouping information as the evaluation result set corresponding to the team; S79: Generate team assessment performance record data based on the evaluation result set corresponding to the team.

3. The method for multi-person collaborative training and emergency drill based on virtual reality according to claim 1 is characterized in that: When the acquired mode selection information is a training mode, the method further includes: Build a team collaboration capability profile based on team historical performance data; The deep Q network outputs the optimal fault injection parameters according to the input team collaboration capability portrait and the spatial environment state of the virtual mine space. The fault injection parameters include the fault type and the timing of the fault occurrence. According to the fault injection parameters, when the conditions of the fault generation timing are met, the corresponding fault type is injected into the virtual mine space; According to the fault type, the virtual three-dimensional model of each mining and transportation equipment is updated to a corresponding state of the fault type; determining execution limit parameters of the virtual three-dimensional model according to the fault type and the updated virtual three-dimensional model; Determine the status of the corresponding components of the actual physical model according to the fault type and the updated virtual three-dimensional model; The actual physical model is updated according to the status of the relevant components of the actual physical model; The restriction instructions of the actual physical model are generated according to the fault type and the execution restriction parameters of the virtual three-dimensional model.

4. The method for multi-person collaborative training and emergency drill based on virtual reality according to claim 1 is characterized in that: The S6 includes: S61: Acquiring real-time status data of the physical model of the mining and transportation equipment during the process of the user operating the physical model of the mining and transportation equipment; S62: Obtain a support vector machine algorithm model trained with data; S63: The support vector machine algorithm model evaluates the collaborative operation status and emergency operation of the mining and transportation equipment according to the input real-time status data of the mining and transportation equipment.

5. The method for multi-person collaborative training and emergency drill based on virtual reality according to claim 4 is characterized in that: The S63 also includes: Obtain evaluation indicators of team performance; Determine the team performance score based on the team performance evaluation indicators; According to the classification comprehensive target value of the support vector machine algorithm model, the historical data of the team performance score and the optimal classification boundary of the support vector machine corresponding to the team performance score are obtained; Obtaining a correspondence between a team performance score and an optimal classification boundary based on the historical data and the optimal classification boundary; When the sliding window change rate of the team performance score exceeds the threshold, the classification boundary update is triggered; When the classification boundary update is triggered, the current team performance score is obtained based on the real-time status data of the mining and transportation equipment; Update the kernel function parameters of the support vector machine algorithm model according to the current team performance score and the corresponding relationship; The classification boundary of the support vector machine algorithm model is adjusted according to the updated kernel function parameters.

6. The method for multi-person collaborative training and emergency drill based on virtual reality according to claim 5 is characterized in that: The S62 includes: Using sensors installed on the mining and transportation equipment to obtain status data of the mining and transportation equipment during the user's operation of the mining and transportation equipment; Extracting a device collaborative operation status data set and an emergency operation data set from the status data; Acquire an initial model of equipment collaborative operation state and an initial model of emergency operation, wherein both the initial model of equipment collaborative operation state and the initial model of emergency operation adopt a support vector machine algorithm model; Annotate the data in the equipment collaborative operation status data set according to the evaluation result corresponding to the equipment collaborative operation status data set; Annotate the data in the emergency operation data set according to the evaluation results corresponding to the emergency operation data set; The equipment collaborative operation status evaluation algorithm model is obtained by training the equipment collaborative operation status initial model through the labeled equipment collaborative operation status data; The emergency operation evaluation algorithm model is obtained by training the initial emergency operation model with labeled emergency operation data.

7. The method for multi-person collaborative training and emergency drill based on virtual reality according to claim 5 is characterized in that: The S63 includes: Using sensors installed on the mining and transportation equipment to obtain real-time status data of the mining and transportation equipment during the user's operation of the mining and transportation equipment; Extracting a real-time equipment collaborative operation status data set and a real-time emergency operation data set from the real-time status data; Obtain the data category label corresponding to each data in the equipment collaborative operation status data set; Inputting the data in the operation status data set and the corresponding data category labels into the equipment collaborative operation status assessment algorithm model; The equipment collaborative operation status evaluation algorithm model outputs the evaluation result of the corresponding mining and transportation equipment collaborative operation status according to the input equipment collaborative operation status data; Obtain the data category labels corresponding to each data in the emergency operation data set; Inputting the data in the emergency operation data set and the corresponding data category labels into the emergency operation evaluation algorithm model; The emergency operation evaluation algorithm model outputs the corresponding emergency operation evaluation results according to the input emergency operation data.

8. The multi-person collaborative training and emergency drill method based on virtual reality according to any one of claims 1 to 7, characterized in that: The mining and transportation equipment includes a hydraulic support, and S5 includes: According to the current digital twin data of the hydraulic support and the previous digital twin data, a set of local changes of the virtual three-dimensional model when the hydraulic support guard plate performs an action is obtained; Before a preset update time node, the shared iteration data of the relevant parts of the guard plate in the local change collection are sent to the server according to the local change collection; When the preset update time node is reached, the shared iterative data is used to replace the corresponding part of the data in the virtual three-dimensional model of the hydraulic support; Execute a shared iterative data update process, the shared iterative data update process comprising: Update the current digital twin data; updating the local change set according to the updated digital twin data and the previous digital twin data; Acquire newly added shared iteration data according to the updated local change set, and send the newly added shared iteration data to the server; When a new preset update time node is reached, the increased shared iteration data is used to replace the corresponding part of the data in the virtual three-dimensional model of the hydraulic support; Determine whether to send redundant data based on the latest local change set and all shared iterative data that have been sent; If yes, the redundant data is sent to the server, and the server extracts corresponding data from the redundant data according to the final state of the side guard plate to replace the data of the corresponding part in the virtual three-dimensional model of the hydraulic support. If no, the shared iterative data updating process is repeated.

9. A multi-person collaborative training and emergency drill method system based on virtual reality, applying the teaching method described in any one of claims 1 to 8, characterized in that: It includes a physical model of mining and transportation equipment, a cloud server, and a digital twin update module. The physical model of mining and transportation equipment is equipped with sensors. The digital twin update module is used to update the digital twin according to the data collected by the sensors. The digital twin update module is communicatively connected to the cloud server.

10. The virtual reality-based multi-person collaborative training and emergency drill method system according to claim 9, characterized in that: It also includes VR glasses, which are communicatively connected to the cloud server, and the VR glasses receive image data sent by the cloud server for displaying the virtual mine space. It also includes a display terminal, which is communicatively connected to the cloud server, and the display is used to receive image information sent by the cloud server for displaying the training process in the virtual mine.

Images