Fire drill evaluation method fusing virtual reality and physiological feedback

Abstract

The application discloses a fire drill evaluation method fusing virtual reality and physiological feedback, relates to the technical field of fire safety management, and solves the technical problem that the evaluation effect is not accurate enough due to the fact that the physiological information of the drill personnel is not monitored and analyzed when the drill personnel is performing simulation drill; drill personnel data, simulation scene data and fire fighting process data are collected, the drill personnel is drilled through VR simulation, physiological baseline change data is collected in the drill process, ideal data simulated by AI is combined to score the drill result; fire fighting nodes are extracted in combination with the fire fighting process data, analysis is performed based on the physiological baseline change data, the physiological baseline change data corresponding to different fire fighting nodes is analyzed, link state data is acquired in combination with the review data, and the physiological stress influence of each node is accurately quantified; stress parameters and drill scores are fused, fire fighting node scores and split standard node scores are calculated, an optimization threshold is set to determine an optimizable node, and a complete evaluation result is integrated.

Description

Technical Field

[0001] This invention belongs to the field of fire safety management technology, specifically a fire drill evaluation method that integrates virtual reality and physiological feedback. Background Technology

[0002] In the field of fire safety management, fire emergency drills are an important means to improve fire fighting and rescue efficiency and respond to sudden fire incidents. Traditional fire drill methods typically involve the fire department or enterprise safety management department developing the plan and relevant personnel participating in the drill. However, fire emergency drill plans developed using traditional methods are often based on fixed patterns, resulting in inefficiency and wasted resources during the drill process.

[0003] The invention patent with publication number CN117993685B discloses an intelligent optimization method and system based on a fire emergency drill plan. This method involves structured data collection from the fire emergency drill plan to obtain data on the building structure, location distribution of fire-fighting facilities, and personnel distribution at the drill site. Based on this data, the method performs key area identification and analysis, and data collection and processing, to obtain the drill process data for key areas of the fire drill site. According to the data analysis, the drill process data of key areas in the fire drill site is used for drill evaluation and intelligent analysis for drill improvement. The fire emergency drill plan is intelligently optimized to obtain an optimized fire emergency drill plan. However, this plan did not take into account that even the drill participants, when faced with relatively dangerous scenarios, may have instinctive fear reactions that slow down decision-making or reduce coordination efficiency, thus affecting the overall drill effect. However, by monitoring and analyzing physiological information, it is possible to analyze the physiological feedback of the drill participants to the scenario, and to better evaluate the drill effect, thereby improving the evaluation effect.

[0004] This invention proposes a fire drill evaluation method that integrates virtual reality and physiological feedback, which addresses the aforementioned technical problems by monitoring and analyzing the physiological information of drill participants. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a fire drill evaluation method that integrates virtual reality and physiological feedback, in order to solve the technical problem that the evaluation effect is not accurate enough due to the lack of monitoring and analysis of the physiological information of the drill participants during the simulation drill.

[0006] To achieve the above objectives, a first aspect of the present invention provides a fire drill evaluation method integrating virtual reality and physiological feedback, comprising: Step 1: Collect simulated scenario data and fire protection process data from the database; Step 2: Conduct drills on the trainees using simulated scenario data to obtain simulated drill data, and collect real-time data on changes in the trainees' physiological baseline. Based on the simulated drill data, score the drill results to obtain a drill score. Step 3: Combine fire protection process data with simulated scenario data to obtain fire protection nodes, analyze fire protection nodes based on physiological baseline change data to obtain corresponding link status data, and set stress parameters for each fire protection node based on link status data; Step 4: Obtain exercise evaluation results based on stress parameters and exercise scores.

[0007] Preferably, the step of obtaining simulated exercise data by training personnel using simulated scenario data includes: Extract simulation scene data; The simulated scene data is input into the VR simulation rendering device, and the VR simulation rendering device is used to conduct fire simulation drills for the drill participants. The decision-making time, coordination efficiency and drill time of the drill participants in different stages of the fire simulation drill are recorded in real time. The ideal decision-making time, ideal coordination efficiency and ideal drill time corresponding to different stages are obtained through AI simulation. The exercise data is obtained by integrating the decision-making time, coordination efficiency, and exercise time of the participants; the ideal data is obtained by integrating the ideal decision-making time, ideal coordination efficiency, and ideal exercise time; and the simulated exercise data is obtained by integrating the exercise data and the ideal data.

[0008] Preferably, the step of scoring the exercise results based on simulated exercise data to obtain an exercise score includes: Extract simulation exercise data; Based on the decision-making time compliance rate, collaboration efficiency compliance rate, and exercise time compliance rate in the simulation exercise data, corresponding weight parameters are assigned to each; the exercise score is obtained by weighting and integrating the decision-making time, collaboration efficiency, and exercise time based on the weight parameters.

[0009] Preferably, the step of assigning corresponding weight parameters to the decision-making time compliance rate, the collaboration efficiency compliance rate, and the exercise time compliance rate in the simulation exercise data includes: Extract the simulation data and ideal data from the simulation exercise data; The ideal time weight is defined as the proportion of the ideal drill time of different stages in the ideal data to the total ideal drill time of all stages. The decision time weight is defined as the ratio of the decision time to the ideal drill time of the corresponding stage. The efficiency weight is defined as the ratio of the collaborative efficiency to the corresponding ideal collaborative efficiency. Ideal time weight, decision time weight, and efficiency weight are used as corresponding weight parameters.

[0010] Preferably, the step of analyzing the simulated scenario data in conjunction with the fire protection process to obtain the fire protection nodes includes: Extract fire protection process data and simulation scenario data; the fire protection process data includes disaster types and operational steps. The simulation scenario data is matched with the disaster type. The simulated disaster type of the current simulation scenario is obtained by feature comparison. The operation steps corresponding to the simulated disaster type are compared with the current simulation scenario data. The simulation scenario data is divided into different simulation scenarios, and the key scenarios corresponding to different simulation scenarios are marked as fire-fighting nodes.

[0011] Preferably, the step of analyzing fire-fighting nodes based on physiological baseline change data to obtain corresponding process status data includes: Extract physiological baseline change data and fire-fighting nodes; among which, physiological baseline change data includes real-time heart rate value, peak skin conductance response, respiratory rate and corresponding time; With time as the horizontal axis and real-time heart rate, peak skin conductance response, and respiratory rate as the vertical axis, heart rate-time curves, peak skin conductance response-time curves, and respiratory rate-time curves are constructed. Standard physiological data is obtained by collecting and statistically analyzing the constant values ​​corresponding to the heart rate-time curve, the peak electrical response-time curve, and the respiratory rate-time curve. The physiological baseline change data corresponding to the fire-fighting node is compared with the corresponding standard physiological data to obtain the stage status data.

[0012] Preferably, the step of comparing the physiological baseline change data corresponding to the fire-fighting node with the corresponding standard physiological data to obtain the link status data includes: Extract physiological baseline change data and standard physiological data; The differences between the heart rate, peak electrical response, and respiratory rate corresponding to the fire-fighting node and the standard physiological data are calculated to obtain the heart rate difference, electrical response difference, and respiratory rate difference. The heart rate difference, electrical response difference, and respiratory rate difference of all fire-fighting nodes are statistically analyzed. The heart rate difference, electrical response difference, and respiratory rate difference are obtained according to the normal distribution algorithm to obtain the corresponding heart rate difference threshold, electrical response difference threshold, and respiratory difference threshold, respectively, and then integrated to obtain the state difference threshold. The physiological baseline change data corresponding to the fire node are compared with the corresponding state difference threshold to obtain the heart rate status, electrical response status, and respiratory status. Scoring parameters are assigned to heart rate status, electrical response status and respiratory status respectively. The status of the link is obtained by judging based on the scoring parameters. The scoring parameters of the fire node and the status of the link are integrated to obtain the status data of the link. The scoring parameters are integers and the range is [0,3].

[0013] Preferably, the step of obtaining the stage status based on the scoring parameters includes: Extract scoring parameters; The total number of physiological states is obtained by statistically analyzing the scoring parameters corresponding to heart rate, electrical response, and respiratory states, and parameter thresholds are set according to the psychological impact of different fire scenarios in the database. The total number of physiological states is compared with the parameter threshold, and the state of each fire protection link is determined based on the comparison results.

[0014] Preferably, the setting of stress parameters for each fire-fighting node based on the link status data includes: Extract process status data; The status of each stage is sorted in ascending order and assigned values ​​from 1 to n. The scoring parameters of each fire-fighting stage are integrated with the corresponding assigned values ​​to obtain the stage score. The total exercise score is obtained by calculating the stage scores of all fire-fighting stages. The ratio of the stage score to the overall drill score is used as the stress parameter for the corresponding fire-fighting node.

[0015] Preferably, obtaining the exercise evaluation result based on stress parameters and exercise scores includes: Extracting stress parameters and exercise scores; The stress parameter is used as a weighted parameter for decision-making time and coordination efficiency as an additional influencing parameter. Combined with the exercise time, the exercise score is calculated by weighted fusion to obtain the fire node score for each link. The exercise score is divided according to the fire node to obtain the standard node score. Set an optimization threshold; determine whether the difference between the fire node score and the standard node score is greater than the optimization threshold; if yes, mark the corresponding fire node as an optimizable node; if no, mark the corresponding fire node as a non-optimizable node; the optimization threshold is set according to the fire scenario corresponding to different fire nodes. The optimizable and non-optimizable nodes are used as node evaluation states. The scores of the fire nodes are statistically analyzed to obtain the total exercise evaluation score. The total exercise evaluation score is then integrated with the node evaluation state corresponding to each fire node to obtain the exercise evaluation result.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention collects data on drill participants, simulated scenarios, and fire-fighting procedures, and conducts drills using VR simulation. During the drills, physiological baseline changes are collected, and the results are scored using ideal data from AI simulations. This makes the scores more objective and valuable, accurately reflecting the operational gaps among drill participants. Fire-fighting nodes are extracted from fire-fighting procedure data, and analysis is performed based on physiological baseline changes. This analysis examines the physiological baseline changes corresponding to different fire-fighting nodes and combines this with review data to obtain stage status data, accurately quantifying the physiological stress impact of each node. This solves the problem of traditional assessments failing to pinpoint weak points, providing support for subsequent accurate assessments. Furthermore, by integrating stress parameters with drill scores, fire-fighting node scores are calculated, standard node scores are broken down, and optimization thresholds are set to determine optimizable nodes. This comprehensive assessment result accurately quantifies the operational performance and physiological stress status of drill participants, improving the relevance and effectiveness of fire-fighting drills.

[0017] 2. This invention extracts fire-fighting nodes by combining fire-fighting process data and obtains stage status data based on physiological baseline change data analysis. It then sets stress parameters for each node. By extracting fire-fighting nodes from fire-fighting process data, it achieves a refined breakdown of drill stages, allowing assessments to extend beyond overall drill performance to each key operational node. Simultaneously, by comparing physiological baseline change data with standard physiological data, it accurately obtains stage status data corresponding to each fire-fighting node, and scientifically sets stress parameters. This quantifies the physiological stress impact of each node, clearly identifying key nodes with significant physiological stress impact on drill participants, and accurately pinpointing operational shortcomings and weak points in physiological regulation during the drill. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of the process of the present invention.

[0020] Figure 2 This is a schematic diagram illustrating the working principle of one embodiment of the present invention. Detailed Implementation

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

[0022] Please see Figure 1 The first aspect of this invention provides a fire drill evaluation method integrating virtual reality and physiological feedback, comprising: Step 1: Collect simulated scenario data and fire protection process data from the database; For example, in this embodiment, three people participate in a fire drill. This fire drill is a high-rise fire drill. The simulated scenario data of the high-rise fire includes parameters such as the location of the fire source, smoke concentration, fire spread speed, location of trapped personnel, and distribution of fire-fighting facilities. Fire-fighting process data is collected from a preset database. This fire-fighting process data is formulated based on national fire rescue standards and industry specifications. The corresponding operational steps for high-rise fires include fire identification, wearing fire-fighting equipment, fire source control, search and rescue of trapped personnel, on-site evacuation, and post-fire cleanup.

[0023] After the data collection is completed, the fire protection process data is classified and organized, and matched one by one with the disaster types in the simulated scenario data, which facilitates the extraction of subsequent fire protection nodes.

[0024] Step 2: Conduct drills on the participants using simulated scenario data to obtain simulated drill data, and collect real-time data on changes in the participants' physiological baseline. Based on the simulated drill data, score the drill results to obtain a drill score.

[0025] For example, the collected simulated scene data is input into the VR simulation rendering device. The participants wear VR headsets, physiological data acquisition devices (heart rate sensors, skin conductance sensors, and respiratory abdominal belts), and motion capture devices, ensuring that all devices are synchronized with the VR device. The sampling frequency of the physiological data acquisition device is set to 1000Hz, and the positioning accuracy of the motion capture device reaches ±0.1mm, capturing the participants' limb movements and operational trajectories in real time. Subsequently, the participants conduct fire drills in simulated scenarios. During the drills, the VR simulation rendering device presents the dynamic changes of the scene in real time, such as the spread of fire over time, the increase in smoke concentration, and changes in environmental parameters in a high-rise fire scenario. Simultaneously, multimodal feedback devices (thermal radiation suits and force feedback gloves) provide the participants with tactile stimulation, recreating a realistic fire scene and providing tactile feedback.

[0026] During the exercise, the decision-making time, coordination efficiency, and exercise time of the participants in different operational steps were recorded in real time. Decision-making time refers to the time from when participants encounter a sudden situation (such as discovering a fire source, a leak, or encountering trapped personnel) to making a clear operational decision. Coordination efficiency refers to indicators such as instruction response rate and operational coordination when multiple participants cooperate to complete an operational step. Exercise time refers to the time it takes for participants to complete a single operational step.

[0027] At the same time, AI technology is used to analyze and simulate fire scenarios, and machine algorithms are used to obtain the ideal decision-making time, ideal coordination efficiency, and ideal drill time for different stages.

[0028] It should be added that the ideal data obtained through AI technology is based on the analysis of a large number of similar past cases, analyzing the best data for the same scenario in similar cases, rather than the ideal data obtained by AI directly calculation.

[0029] The ideal time weight is defined as the proportion of the ideal drill time for different stages in the ideal data to the total ideal drill time for all stages. The decision time weight is defined as the ratio of the decision time to the ideal drill time for the corresponding stage. The efficiency weight is defined as the ratio of the coordination efficiency to the corresponding ideal coordination efficiency. In this embodiment, the ideal time for the fire source control stage is 15 minutes, and the ideal total drill time is 90 minutes, so the ideal time weight is 1 / 6. After calculating and obtaining the corresponding weight parameters, the decision time, coordination efficiency and drill time are normalized and then weighted and fused together to obtain the drill score.

[0030] It should be added that the weight parameters are assigned based on ideal data. Their significance represents the importance of the decision-making time, collaboration efficiency, and exercise time of each link to the overall score. Obtaining weight parameters through ideal data allows for a more rigorous scoring of the exercise.

[0031] Step 3: Combine fire protection process data with the analysis of simulated scenario data to obtain fire protection nodes. Analyze the fire protection nodes based on physiological baseline change data to obtain corresponding link status data. Set stress parameters for each fire protection node based on the link status data.

[0032] The simulated scenario data is matched with the disaster types in the fire protection process data. The simulated disaster type of the current simulated scenario is obtained through feature comparison. Specifically, the core features in the simulated scenario data (such as whether there is an open flame, smoke concentration, and type of leaked substance) are extracted and compared with the typical features of the two disaster types in the fire protection process data. For example, when there is an open flame, a continuously increasing smoke concentration, and no toxic gas detection signal in the simulated scenario data, the simulated disaster type is determined to be a high-rise fire. Once the simulated disaster type is determined to be a high-rise fire, the corresponding operation link is matched. This operation link represents different scenarios encountered in different stages of fire protection in a high-rise fire. The simulated scenarios corresponding to different operation links in the simulated scenario data are divided. For example, in a high-rise fire scenario, the scenario corresponding to "detecting an open flame" is divided into the fire identification stage scenario. The key scenarios corresponding to different simulated scenarios are marked as fire protection nodes.

[0033] It should be added that the key scenarios are specifically the scenarios in different stages of the simulated scenario when encountering danger or needing to perform operations, such as encountering critical situations such as a door frame falling or an open flame blocking the road.

[0034] Using time as the horizontal axis and real-time heart rate, peak skin conductance response, and respiratory rate as the vertical axis, heart rate-time curves, peak skin conductance response, and respiratory rate-time curves for each participant are constructed using data visualization tools (such as Matlab). The curves clearly show the changes in physiological indicators at each time point, facilitating intuitive observation of fluctuations in physiological state. Standard physiological data are obtained by collecting and statistically analyzing the constant values ​​corresponding to the heart rate-time curve, peak skin conductance response, and respiratory rate-time curves, respectively.

[0035] In another embodiment, when collecting constant values ​​from the curve graph, if there are no regular constant values, several sets of relatively close values ​​are selected, and their average value is selected as the constant value.

[0036] The differences between the heart rate, peak electrical response, and respiratory rate corresponding to the fire node and the standard physiological data are calculated to obtain the heart rate difference, electrical response difference, and respiratory rate difference. All the statistically obtained heart rate differences are fitted with a normal distribution to determine the mean and standard deviation. The range obtained based on the mean and standard deviation is used as the heart rate difference threshold. Similarly, the mean and standard deviation of the electrical response difference and the mean and standard deviation of the respiratory rate difference are fitted to determine the electrical response difference threshold and the respiratory rate difference threshold, respectively. The three are integrated into the state difference threshold.

[0037] Acquire heart rate status, electrical response status, and respiratory status. For example, when the heart rate difference of a certain fire node is within the heart rate difference threshold range, the heart rate status is judged as normal, and when it exceeds the threshold range, it is judged as abnormal. Acquire the heart rate status, electrical response status, and respiratory status of the corresponding fire node respectively.

[0038] It should be noted that heart rate, peak skin conductance, and respiratory rate can reflect the physiological reactions of trainees when facing fire scenarios. For example, an increased heart rate, a sudden increase in respiratory rate, and an increase in peak skin conductance can all reflect the danger of the fire scenario at that time. People have an instinctive fear response to danger, and physiological data can well represent the response to danger. When facing danger, people's decision-making and action time and efficiency will deteriorate.

[0039] Scoring parameters are assigned to heart rate, electrical response, and respiratory states respectively. These parameters are integers ranging from [0,3]. A normal state corresponds to 3 points, a mild abnormality to 2 points, a moderate abnormality to 1 point, and a severe abnormality to 0 points. The total number of physiological states is obtained by counting the scoring parameters for heart rate, electrical response, and respiratory states. Parameter thresholds are set based on the psychological impact of different fire scenarios stored in the database. For example, in a high-rise fire scenario, the threshold for fire identification is 7 points, and the threshold for fire source control is 6 points. The total number of physiological states at each fire node is compared with the corresponding parameter thresholds. Based on the comparison results, the stage status corresponding to each fire stage is determined. The scoring parameters and stage status of each fire node are integrated to obtain stage status data, which includes the scores of three physiological indicators, the total number of physiological states, and the stage status for each fire node.

[0040] All fire-fighting node statuses are sorted from best to worst and assigned values ​​from 1 to n (n is the total number of fire-fighting nodes; in this embodiment, n=6 for each scenario). A good status is assigned a value of 6, a fair status is assigned a value of 3, and a poor status is assigned a value of 1. Then, the scoring parameters of each fire-fighting node (the sum of heart rate, electrical response, and respiratory scores) are integrated with the corresponding assigned values ​​to obtain the node score. For example, if the sum of the three scores for a fire-fighting node is 8 points and the node status is assigned a value of 6, then the node score for that fire-fighting node is (8+6) / 2=7 points. The total drill score is obtained by calculating the node scores of all fire-fighting nodes. The total drill score is the average of the node scores of all fire-fighting nodes.

[0041] Step 4: Obtain exercise evaluation results based on stress parameters and exercise scores.

[0042] For example, the stress parameters of each fire node are used as weight parameters for the decision time and coordination efficiency of the corresponding operation link of that node, and are also used as additional influencing parameters. They are combined with the exercise time of that link for weighted fusion calculation to obtain the fire node score of each link. The exercise score is divided according to the fire node, and the exercise score is split according to the weight of the operation link corresponding to each fire node in the whole exercise process to obtain the standard node score of each fire node.

[0043] Optimization thresholds are set based on the specific fire scenarios corresponding to different fire nodes, taking into account the importance and difficulty of each operational step. For example, in a high-rise fire scenario, the optimization threshold for the fire source control step is set to 3 points, and the optimization threshold for the personnel search and rescue step is set to 2 points; in a chemical leak scenario, the optimization threshold for the leak sealing step is set to 4 points, and the optimization threshold for the environmental monitoring step is set to 1 point. Then, it is determined whether the difference between the fire node score and the standard node score for each fire node is greater than the optimization threshold: if the difference is greater than the optimization threshold, it indicates that the performance of the operational step corresponding to that fire node has not met the standard and there is room for optimization; the corresponding fire node is then marked as an optimizable node. If the difference is not greater than the optimization threshold, it indicates that the performance of the operational step corresponding to that fire node has met the standard and no optimization is needed; the corresponding fire node is then marked as a non-optimizable node. Finally, all optimizable and non-optimizable nodes are used as node evaluation states, and the average score of all fire-fighting nodes is calculated to obtain the total exercise evaluation score, which ranges from 0 to 100. At the same time, the total exercise evaluation score is integrated with the node evaluation state, link state data, and stress parameters corresponding to each fire-fighting node to obtain a complete exercise evaluation result.

[0044] It should be added that each optimization node is determined to be capable of being optimized through human training, thereby improving the decision-making time and coordination efficiency of the trainees when facing the same scenario, and thus achieving better training results.

[0045] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0046] Working principle of the invention: This invention collects data on drill participants, simulated scenarios, and fire-fighting procedures. It then conducts VR simulations to train participants, collecting physiological baseline change data during the drills. The results are scored based on ideal data from AI simulations. Fire-fighting procedures data is used to extract fire-fighting nodes, and physiological baseline change data is analyzed to determine the physiological baseline changes corresponding to different fire-fighting nodes. Furthermore, it combines trial data to obtain stage status data, accurately quantifying the physiological stress impact of each node. Finally, stress parameters are integrated with the drill scores to calculate fire-fighting node scores, break down standard node scores, set optimization thresholds to determine optimizable nodes, and integrate these to form a complete evaluation result.

[0047] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method of fire drill evaluation fusing virtual reality with physiological feedback, characterized in that, include: Step 1: Collect simulated scenario data and fire protection process data from the database; Step 2: Conduct drills on the trainees using simulated scenario data to obtain simulated drill data, and collect real-time data on changes in the trainees' physiological baseline. Based on the simulated drill data, score the drill results to obtain a drill score. Step 3: Combine fire protection process data with simulated scenario data to obtain fire protection nodes, analyze fire protection nodes based on physiological baseline change data to obtain corresponding link status data, and set stress parameters for each fire protection node based on link status data; Step 4: Obtain exercise evaluation results based on stress parameters and exercise scores.

2. The fire drill evaluation method of fusing virtual reality with physiological feedback according to claim 1, wherein, The simulated exercise data obtained by training personnel using simulated scenario data includes: Extract simulation scene data; The simulated scene data is input into the VR simulation rendering device, and the VR simulation rendering device is used to conduct fire simulation drills for the drill participants. The decision-making time, coordination efficiency and drill time of the drill participants in different stages of the fire simulation drill are recorded in real time. The ideal decision-making time, ideal coordination efficiency and ideal drill time corresponding to different stages are obtained through AI simulation. The exercise data is obtained by integrating the decision-making time, coordination efficiency, and exercise time of the participants; the ideal data is obtained by integrating the ideal decision-making time, ideal coordination efficiency, and ideal exercise time; and the simulated exercise data is obtained by integrating the exercise data and the ideal data.

3. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 1, characterized in that, The exercise score is obtained by scoring the exercise results based on the simulated exercise data, including: Extract simulation exercise data; Based on the decision-making time compliance rate, collaboration efficiency compliance rate, and exercise time compliance rate in the simulation exercise data, corresponding weight parameters are assigned to each; the exercise score is obtained by weighting and integrating the decision-making time, collaboration efficiency, and exercise time based on the weight parameters.

4. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 3, characterized in that, The steps involve assigning corresponding weight parameters to the decision-making time compliance rate, collaboration efficiency compliance rate, and exercise time compliance rate based on the simulation exercise data, including: Extract the simulation data and ideal data from the simulation exercise data; The ideal time weight is defined as the proportion of the ideal drill time of different stages in the ideal data to the total ideal drill time of all stages. The decision time weight is defined as the ratio of the decision time to the ideal drill time of the corresponding stage. The efficiency weight is defined as the ratio of the collaborative efficiency to the corresponding ideal collaborative efficiency. Ideal time weight, decision time weight, and efficiency weight are used as corresponding weight parameters.

5. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 1, characterized in that, The process of analyzing simulated scenario data in conjunction with fire protection procedures to obtain fire protection nodes includes: Extract fire protection process data and simulation scenario data; the fire protection process data includes disaster types and operational steps. The simulation scenario data is matched with the disaster type. The simulated disaster type of the current simulation scenario is obtained by feature comparison. The operation steps corresponding to the simulated disaster type are compared with the current simulation scenario data. The simulation scenario data is divided into different simulation scenarios, and the key scenarios corresponding to different simulation scenarios are marked as fire-fighting nodes.

6. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 1, characterized in that, The analysis of fire-fighting nodes based on physiological baseline change data to obtain corresponding process status data includes: Extract physiological baseline change data and fire-fighting nodes; among which, physiological baseline change data includes real-time heart rate value, peak skin conductance response, respiratory rate and corresponding time; With time as the horizontal axis and real-time heart rate, peak skin conductance response, and respiratory rate as the vertical axis, heart rate-time curves, peak skin conductance response-time curves, and respiratory rate-time curves are constructed. Standard physiological data is obtained by collecting and statistically analyzing the constant values ​​corresponding to the heart rate-time curve, the peak electrical response-time curve, and the respiratory rate-time curve. The physiological baseline change data corresponding to the fire-fighting node is compared with the corresponding standard physiological data to obtain the stage status data.

7. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 6, characterized in that, The step of comparing the physiological baseline change data corresponding to the fire-fighting node with the corresponding standard physiological data to obtain the link status data includes: Extract physiological baseline change data and standard physiological data; The differences between the heart rate, peak electrical response, and respiratory rate corresponding to the fire-fighting node and the standard physiological data are calculated to obtain the heart rate difference, electrical response difference, and respiratory rate difference. The heart rate difference, electrical response difference, and respiratory rate difference of all fire-fighting nodes are statistically analyzed. The heart rate difference, electrical response difference, and respiratory rate difference are obtained according to the normal distribution algorithm to obtain the corresponding heart rate difference threshold, electrical response difference threshold, and respiratory difference threshold, respectively, and then integrated to obtain the state difference threshold. The physiological baseline change data corresponding to the fire node are compared with the corresponding state difference threshold to obtain the heart rate status, electrical response status, and respiratory status. Scoring parameters are assigned to heart rate status, electrical response status and respiratory status respectively. The status of the link is obtained by judging based on the scoring parameters. The scoring parameters of the fire node and the status of the link are integrated to obtain the status data of the link. The scoring parameters are integers and the range is [0,3].

8. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 7, characterized in that, The process of determining the stage status based on scoring parameters includes: Extract scoring parameters; The total number of physiological states is obtained by statistically analyzing the scoring parameters corresponding to heart rate, electrical response, and respiratory states, and parameter thresholds are set according to the psychological impact of different fire scenarios in the database. The total number of physiological states is compared with the parameter threshold, and the state of each fire protection link is determined based on the comparison results.

9. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 1, characterized in that, The setting of stress parameters for each fire node based on the link status data includes: Extract process status data; The status of each stage is sorted in ascending order and assigned values ​​from 1 to n. The scoring parameters of each fire-fighting stage are integrated with the corresponding assigned values ​​to obtain the stage score. The total exercise score is obtained by calculating the stage scores of all fire-fighting stages. The ratio of the stage score to the overall drill score is used as the stress parameter for the corresponding fire-fighting node.

10. The fire drill evaluation method integrating virtual reality and physiological feedback according to claim 1, characterized in that, The process of obtaining exercise evaluation results based on stress parameters and exercise scores includes: Extracting stress parameters and exercise scores; The stress parameter is used as a weighted parameter for decision-making time and coordination efficiency as an additional influencing parameter. Combined with the exercise time, the exercise score is calculated by weighted fusion to obtain the fire node score for each link. The exercise score is divided according to the fire node to obtain the standard node score. Set an optimization threshold; determine whether the difference between the fire node score and the standard node score is greater than the optimization threshold; if yes, mark the corresponding fire node as an optimizable node; if no, mark the corresponding fire node as a non-optimizable node; the optimization threshold is set according to the fire scenario corresponding to different fire nodes. The optimizable and non-optimizable nodes are used as node evaluation states. The scores of the fire nodes are statistically analyzed to obtain the total exercise evaluation score. The total exercise evaluation score is then integrated with the node evaluation state corresponding to each fire node to obtain the exercise evaluation result.

Images