Multi-source data real-time monitoring system and method for rescue drill

Abstract

The invention relates to a multi-source data real-time monitoring system for rescue drilling, which belongs to the technical field of virtual simulation and emergency drilling and comprises a data acquisition module, a data transmission module, a data processing module and a display module. The data acquisition module acquires working parameters of a positive pressure oxygen respirator worn by a rescue worker, environmental parameters of a smoke and heat environment simulation device and physiological parameters of the worker in real time through various sensors; the data transmission module transmits data acquired by the data acquisition module to the data processing module through an ad hoc network and a local area network; the data processing module performs time alignment, format standardization and state fusion on the multi-source heterogeneous data to generate comprehensive situation information; and the display module displays all data, personnel states and alarm information in real time through a three-dimensional virtual simulation interface.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of virtual simulation and emergency drill, and relates to a multi-source data real-time monitoring system and method for rescue drill. BACKGROUND

[0002] In the field of emergency rescue drill, real-time collection of multi-source data (such as personnel physiological parameters, environmental parameters, and equipment states) in the drill process is a key link for improving the drill efficiency, and is of great significance for evaluating the drill effect and optimizing the rescue process.

[0003] Traditional monitoring methods often have problems such as data isolation, response lag, and inability to quantitatively evaluate the drill effect, and are difficult to meet the collaborative analysis needs in complex scenarios. The existing technology has the following problems: positive pressure oxygen respirator parameters, environmental parameters, and personnel physiological parameters are independently collected and displayed by different devices, and there is a lack of a unified platform, making it difficult for commanders to obtain a global perspective; traditional monitoring (such as cameras) has blind spots in smoke, high heat, or complex structures, and cannot obtain key internal parameters (such as cylinder pressure and personnel heartbeat); commanders need to integrate multiple information to judge the scene situation, and the reaction speed is slow, which may cause the opportunity to be missed in an emergency; the drill process data is not systematically recorded, and it is difficult to accurately and quantitatively evaluate the drill effect and individual performance.

[0004] Therefore, there is an urgent need for a solution that can integrate multi-source data and realize real-time comprehensive monitoring and analysis. SUMMARY

[0005] Therefore, the purpose of the present application is to provide a multi-source data real-time monitoring system and method for rescue drill.

[0006] To achieve the above purpose, the present application provides the following technical solutions:

[0007] On the one hand, the present application provides a multi-source data real-time monitoring system for rescue drill, comprising a data acquisition module, a data transmission module, a data processing module, and a display module.

[0008] The data acquisition module acquires, in real time, working parameters of a positive pressure oxygen respirator worn by a rescuer, environmental parameters of a smoke and heat simulation device, and physiological parameters of a person through a plurality of sensors;

[0009] The data transmission module transmits the data collected by the data acquisition module to the data processing module through ad hoc network and local area network;

[0010] The data processing module performs time alignment, format standardization, and state fusion on multi-source heterogeneous data to generate comprehensive situation information;

[0011] The display module displays all data, personnel status and alarm information in real time through a three-dimensional virtual simulation interface.

[0012] Further, the data acquisition module includes a microprocessor and various sensors, the various sensors send the collected data to the microprocessor for processing; the various sensors include a temperature sensor, a smoke concentration sensor, a motion capture camera, a personnel wearable device and a pressure sensor and a flow sensor arranged in the positive pressure oxygen respirator; the temperature sensor is used for collecting the ambient temperature in the smoke heat environment, the smoke concentration sensor is used for collecting the smoke concentration, and the motion capture camera is used for positioning personnel; the personnel wearable device includes a smart bracelet, an integrated photoelectric heart rate sensor and a blood oxygen saturation sensor; the pressure sensor is used for monitoring the gas cylinder pressure, and the flow sensor is used for monitoring the breathing frequency and instantaneous flow.

[0013] Further, the data transmission module includes an ad hoc network gateway, a switch and a router; the ad hoc network gateway forms a wireless mesh network, automatically finds an optimal path to receive the data transmitted by each mobile sensor to the microprocessor, and transmits the data to the data transmission module through the switch and the router.

[0014] Further, the data processing module processing process includes:

[0015] Receiving the original data packet from the transmission module, and parsing the various parameter values according to the predefined communication protocol;

[0016] Establishing a unified time coordinate system with the server clock as the reference, and eliminating the problem of time asynchronization;

[0017] The source end time of the data packet i is The server receiving time is The time reference provided for data fusion is t rxi ; using a delay compensation algorithm, estimating the approximate generation time of the data packet on the server time axis Where the average delay δ avg is dynamically estimated through an exponential weighted moving average model .

[0018] Converting the parsed parameter values into structured data of standard units, and comparing with the corresponding threshold values to determine whether the personnel are in a dangerous state;

[0019] Storing all original data and fused state data into a time series database to provide data support for real-time display and subsequent review.

[0020] Further, the structure of converting the parsed parameter value into standard unit structured data and comparing with the corresponding threshold value to determine whether the personnel is in danger state specifically includes the following rules:

[0021] Fusion rule 1: ambient temperature > 60℃, personnel heart rate > 140bpm, determine that the personnel is in a high risk state of heat stress failure;

[0022] Fusion rule 2: abnormal decrease of gas cylinder pressure, abnormal increase of personnel breathing rate, warning that the respirator may leak or the personnel is extremely nervous.

[0023] Further, the display module includes a command large screen, three-dimensional virtual simulation software, a VR all-in-one machine and a sound and light alarm device; a three-dimensional virtual environment consistent with the real training scene is constructed by using a Unity 3D engine, and corresponding elements in the virtual scene are driven by data by using the data processing module; the three-dimensional scene tracks and displays the virtual avatar of each personnel in real time through a motion capture camera; the personnel wears the VR all-in-one machine, and the physiological, equipment and environmental data of the personnel are displayed in real time in the VR scene in the form of a pop-up window; the smoke concentration in the training scene range is simulated by a color particle effect, and the ambient temperature distribution is represented by color gradient; the personnel realizes the intelligent warning and alarm functions through the warning pop-up box, flashing red light and sound and light alarm in the VR scene; when the data processing module determines an abnormal state, the command large screen and the three-dimensional virtual simulation software simultaneously issue corresponding warnings, the command large screen pops up a warning box, the virtual avatar of the corresponding personnel flashes red light, and the sound and light alarm in the scene prompts the commander and the personnel himself.

[0024] Further, the display module also has a training review and analysis function, which replays the entire three-dimensional process after the training ends, and calls the historical data of any team member for quantitative evaluation.

[0025] On the other hand, the present application provides a multi-source data real-time monitoring method for rescue training, which comprises the following steps:

[0026] S1: system hardware device deployment and initialization; in a large space training field simulating a disaster environment, motion capture cameras, smoke and heat simulation equipment and self-organizing network gateways are fixedly installed according to task requirements, and a set of integrated devices are worn by the training team members, the integrated devices including a positive pressure oxygen respirator with an oxygen respirator parameter acquisition module, a smart bracelet and a VR all-in-one machine; the positive pressure oxygen respirator is internally provided with a parameter acquisition device, an electronic alarm and a sensor, which acquire and transmit corresponding parameters in the training process of the team members, and remind the team members of matters needing attention when the parameters exceed the threshold value; the smart bracelet acquires physiological parameters of the team members such as heartbeat and blood oxygen saturation; the sensor modules on all integrated devices are connected with microprocessors through Bluetooth, and the microprocessors are connected with a three-dimensional virtual scene through a wireless gateway;

[0027] S2: After the exercise begins, each sensor operates at a preset frequency. The pressure sensor and flow sensor on the respirator continuously monitor the remaining pressure value of the gas cylinder and the breathing flow waveform of the personnel. The data is transmitted to the microprocessor via Bluetooth.

[0028] S3: The data receiving service on the server continuously listens to the network port and receives various data packets from different gateways and base stations; according to the predefined communication protocol, it parses the data packets and separates the data source ID, data type, data value and original data timestamp; the parsed data is sent to the data preprocessing module for timestamp unification, data filtering and format standardization.

[0029] S4: The preprocessed data is stored in the cache for system fusion and calling. The data fusion processing module performs periodic status judgments based on the system's preset rule base; combined with multi-source data, it performs higher-level comprehensive status judgments through logical combination.

[0030] S5: Based on the received positioning data and motion capture camera data, the position and posture of the corresponding personnel virtual avatar in the 3D scene are updated in real time. At the same time, the motion capture camera data and human skeleton calculation analysis are combined to determine the body shape and movements of the team members. Each team member can keep track of their latest physiological data, equipment data and overall status calculated by the fusion module in real time through the VR all-in-one device. When the data exceeds the safe range, the panel color changes from green to yellow. When a high-risk alarm is received from the data fusion feedback, a red warning window pops up in the center of the team member's VR field of vision and the command screen, and a voice alarm is broadcast. At the same time, the corresponding personnel virtual avatar in the scene begins to flash frequently.

[0031] S6: Commanders can perform spatiotemporal retrospection, data analysis, and event annotation through the back-end management terminal; after the exercise, commanders can retrieve exercise data to reproduce the entire exercise process; select any personnel to view the historical change curves of all parameters during the exercise and perform on-screen comparative analysis; mark key event points on the timeline and associate them with the data at that moment to generate a detailed exercise evaluation report.

[0032] The beneficial effects of this invention are as follows: It breaks down information silos, integrating and displaying all key information on a single platform. Team members can promptly understand each other's status, and the command center can clearly grasp various team member indicators, greatly improving exercise efficiency. Through multi-parameter fusion logic, it extracts deeper levels of danger from the data, achieving pre-emptive warnings rather than just post-event alerts. Three-dimensional virtual simulation makes abstract data intuitive and visible, reducing the cognitive load on command personnel and improving decision-making efficiency and level. Complete data recording provides precise data support for scientific review, optimization of exercise plans, and evaluation of team member performance, helping team members identify gaps in their knowledge and improve their capabilities during exercises.

[0033] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0034] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0035] Figure 1 A schematic diagram of the architecture of a multi-source data real-time monitoring system for rescue drills;

[0036] Figure 2 A schematic diagram showing the deployment of data acquisition equipment;

[0037] Figure 3 A diagram illustrating the hardware equipment worn by team members.

[0038] Attached label: 1-Command screen; 2-Back-end management terminal; 3-Smoke and heat simulation device; 4-Motion capture camera; 5-Environmental sensor; 6-Microprocessor; 7-Virtual reality control handle; 8-Smart wearable device; 9-VR all-in-one machine; 10-Positive pressure oxygen respirator. Detailed Implementation

[0039] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0040] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0041] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0042] Example 1:

[0043] like Figure 1 As shown, this embodiment provides a multi-source data real-time monitoring system for rescue drills, including a data acquisition module, a data transmission module, a data processing module, and a display module. The data acquisition module collects real-time operating parameters of the positive pressure oxygen respirator worn by rescue personnel, environmental parameters of the smoke and heat environment simulation device, and physiological parameters of the personnel through various sensors. The data transmission module reliably transmits data to the monitoring center via a self-organizing network and a local area network. The data processing module performs time alignment, format standardization, and status fusion on the multi-source heterogeneous data to generate comprehensive situational information. The display module displays all data, personnel status, and alarm information in real-time through a three-dimensional virtual simulation interface and supports drill debriefing.

[0044] The data acquisition module collects parameters of the positive pressure oxygen respirator, environmental parameters (smoke and heat simulation device), and personnel physiological parameters. Oxygen respirator parameters are collected through integrated pressure and flow sensors. The pressure sensor monitors cylinder pressure, and the flow sensor monitors respiratory rate and instantaneous flow rate. Environmental parameters include ambient temperature, smoke concentration, and personnel location during the exercise. Temperature and smoke concentration are collected by temperature and smoke concentration sensors, respectively, and personnel location is determined by motion capture cameras and positioning tags. Personnel physiological parameters typically include heart rate, respiration, and blood oxygen saturation, collected by wearable devices worn by rescue personnel, usually through smart bracelets, integrated photoelectric heart rate sensors, and blood oxygen saturation sensors. Each sensor converts physical or chemical quantities into analog electrical signals based on its physical characteristics, which are then converted into digital signals via analog-to-digital converters (ADCs). Each sensor transmits data to a microprocessor via Bluetooth, and the microprocessor wirelessly packages and transmits the data.

[0045] The data transmission module consists of an autonomous network gateway (located within the training space), switches, routers, and other equipment. It solves the problem of wireless signal coverage in large spaces by employing a sensor node-autonomous network gateway-local area network-server transmission mode. The autonomous network gateway forms a wireless mesh network, automatically finding the optimal path to receive data transmitted from various mobile sensors to the microprocessor. The microprocessor wirelessly accesses the local network of the training ground, ultimately transmitting the data stably to the server in the back-end monitoring center, ensuring data continuity and reliability in complex environments.

[0046] The data processing module includes a data receiving service, a data processing service, and a database. The data receiving service receives raw data packets from the transmission module and parses out various parameter values ​​according to a predefined communication protocol. A unified time coordinate system based on the server clock is established to eliminate time asynchrony issues caused by clock deviations from heterogeneous data sources and random network transmission delays, providing a foundation for subsequent spatiotemporal alignment and fusion of multi-source data. Let the source time of data packet i be... The server receiving time is Then the time reference it provides for data fusion is t. rxi To improve the timeliness and accuracy of data fusion, a delay compensation algorithm is further adopted to estimate the approximate generation time of data packets on the server's timeline. Where the average delay δ avg It can be achieved through the exponentially weighted moving average model Dynamic estimation is performed. The parsed values ​​of parameters such as pressure and heart rate are converted into structured data in standard units. Each parameter is compared with its corresponding threshold to determine whether a dangerous state exists. The threshold values ​​for different parameters are shown in Table 1.

[0047] Table 1

[0048]

[0049] Judgments are made based on simple logic such as "AND" and "OR". Fusion Rule 1: If the ambient temperature is >60℃ and the person's heart rate is >140 bpm, the person is determined to be at high risk of heat stress failure. Fusion Rule 2: A sharp drop in cylinder pressure and an abnormally high breathing rate indicate a possible respirator leak or extreme anxiety in the person. All raw data and fused status data are stored in a time-series database to provide data support for real-time display and subsequent review.

[0050] The display module includes a command screen, 3D virtual simulation software, and an audible and visual alarm system. Based on these devices, the system can achieve 3D visualization, intelligent early warning and alarm, and exercise review and analysis. The 3D visualization utilizes the Unity 3D engine to construct a 3D virtual environment consistent with the real exercise scenario, using data from the data processing module to drive corresponding elements in the virtual scene. The 3D scene uses motion capture cameras to track and display the virtual avatars of each person in real time, with their positions updated in real time by wireless communication (UWB) positioning data. Exercise participants in the VR scene see their own physiological, equipment, and environmental data displayed in real time via pop-up windows. Smoke concentration within the exercise scene is simulated using color particle effects, and environmental temperature distribution (heatmap) is represented by color gradients. Intelligent warning and alarm functions are implemented through warning pop-ups, flashing red lights, and audible and visual alarms in the VR scene. When the data processing module detects an abnormal state, the command screen and the 3D virtual scene in the display module will simultaneously issue corresponding warnings: a warning pop-up appears on the command screen, the corresponding participant's virtual avatar flashes red, and the on-site audible and visual alarm is triggered to alert the commander and the participant. After the drill, the entire 3D process can be replayed through the back-end management terminal, and historical data of any team member can be retrieved for quantitative evaluation (such as analyzing the stress response time of all team members during a fire rescue drill).

[0051] Example 2:

[0052] like Figures 2-3 As shown in the figure, this embodiment provides a method for real-time monitoring of multi-source data for rescue drills, including the following steps:

[0053] S1: System hardware device deployment and initialization, such as Figure 2 As shown, the hardware equipment involved in this invention consists of two main parts: environmental sensing equipment and personnel-worn equipment. In a large-space training ground simulating a disaster environment, motion capture cameras 4, smoke and heat simulation equipment 3, and self-organizing network gateways are fixedly installed according to mission requirements, and training team members wear a set of integrated equipment.

[0054] The integrated equipment worn by team members includes a positive pressure oxygen respirator 10 with a built-in oxygen respirator parameter acquisition module, a smart bracelet 8, and a VR all-in-one device 9. The positive pressure oxygen respirator 10 has a built-in parameter acquisition unit, electronic alarm, and sensors, which can collect and transmit corresponding parameters during the team member's exercise. When parameters exceed the threshold, it can remind the team member of precautions. The smart bracelet 8 can collect physiological parameters such as the team member's heart rate and blood oxygen saturation. The sensor modules on all team members' equipment are connected to the microprocessor 6 via Bluetooth. The microprocessor 6 is connected to the 3D virtual scene via a wireless gateway to ensure that the team member's actual actions, position, and other information are synchronized with the virtual scene throughout the exercise.

[0055] All training equipment is powered on, the software on the server is started, and the system performs a self-check. After confirming that all sensor nodes and gateways are connected normally, the 3D virtual scene is initialized, displaying the 3D model of the training scene and the initial positions of all team members. Commanders can monitor the training process in real time through the command screen 1.

[0056] S2: After the exercise begins, each sensor operates at a preset frequency. The pressure sensor and flow sensor on the respirator 10 continuously monitor the remaining pressure value of the gas cylinder and the breathing flow waveform of the personnel. The data is transmitted to the microprocessor 6 via Bluetooth. The data packet contains the device ID, pressure value, breathing rate, etc.

[0057] The smart bracelet worn by the person collects raw heart rate and blood oxygen saturation signals through photoelectric sensing technology. After calculation, the heart rate value (HR) and blood oxygen saturation value (SpO2) are obtained and wirelessly transmitted along with the device ID.

[0058] The environmental sensor 5 continuously collects ambient temperature and smoke concentration values ​​and transmits them wirelessly.

[0059] The motion capture camera 4 captures the training scene in real time and transmits the captured photos to the microprocessor 6. The microprocessor 6 wirelessly transmits the photo information to the back-end management terminal 2. The management terminal software calculates the precise three-dimensional coordinates of the team members and simultaneously identifies the team members' fine movements through human skeleton recognition, sending them directly to the server 2 via the local area network.

[0060] S3: The data receiving service on server 2 continuously listens to the network port, receiving various data packets from different gateways and base stations. According to a predefined communication protocol, it parses the data packets, separating the data source ID (used to determine which person or device is at which location), data type (whether it's pressure, heart rate, or temperature, etc.), data value, and the original timestamp.

[0061] The parsed data is sent to the data preprocessing module for timestamp unification, data filtering, and format standardization.

[0062] S4: The preprocessed data is stored in a cache for system fusion and processing. The data fusion processing module performs periodic state checks (once per second) based on the system's preset rule base. This invention preferably adopts a rule-based state fusion method.

[0063] By combining data from multiple sources, a higher-level comprehensive status judgment is made through logical combination.

[0064] Rule Example 1 (Integration of Environmental and Physiological Load):

[0065] If (environment.temperature>75℃And person.heart_rate>140bpm)OR (person.SpO2<90%)Then person.overall_status = "Extremely high heat load risk", the warning level is "high risk".

[0066] Rule Example 2 (Equipment and Behavior Integration):

[0067] If (breathing_apparatus.pressure decrease rate > 0.5 MPa / 10s and person.respiratory_rate > 30 breaths / minute)

[0068] Then breathing_apparatus.status = "Suspected leak or abnormally rapid breathing", with the warning level set to "Warning".

[0069] Rule Example 3:

[0070] If person.location is in "High-risk zone A" and person.overall_status! = "Normal"

[0071] Then, the highest level "personnel in distress" alert is triggered.

[0072] All generated status information and raw data will be synchronously written to the time-series database for permanent storage.

[0073] S5: Based on the received positioning data and motion capture camera data, the software updates the position and posture of the corresponding virtual avatar in the 3D scene in real time. Simultaneously, it combines motion capture camera data and human skeleton calculations to determine the team member's body shape and fine motor skills. Each team member can monitor their latest physiological data, equipment data, and overall status calculated by the fusion module in real time through the VR all-in-one device. When the data exceeds the safe range, the panel color changes from green to yellow.

[0074] The 3D scene, based on live footage, uses a particle system to render smoke concentration effects and shader technology to render heatmap effects, realistically recreating the exercise environment. When a high-risk alert is received from data fusion feedback, a red warning window pops up in the center of the team member's VR view and the command screen, accompanied by an audio alarm. Simultaneously, the corresponding personnel's virtual avatar in the scene begins to flash frequently, attracting the attention of the commander and the actual personnel.

[0075] S6: Commanders can perform spatiotemporal replay, data analysis, and event annotation through the backend management terminal. After the exercise, commanders can retrieve exercise data to reproduce the entire exercise process; they can select any personnel to view the historical change curves of all parameters during the exercise (pressure curve, heart rate curve, etc.) and perform on-screen comparative analysis; they can annotate key event points on the timeline and correlate them with the data at that moment to generate a detailed exercise evaluation report.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions 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 solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-source data real-time monitoring system for rescue drills, characterized in that: The system comprises a data acquisition module, a data transmission module, a data processing module and a display module. The data acquisition module acquires in real time working parameters of a positive pressure oxygen respirator worn by a rescue worker, environmental parameters of a smoke heat environment simulation device and physiological parameters of the worker through various sensors. The data transmission module transmits data acquired by the data acquisition module to the data processing module through ad hoc networks and local area networks. The data processing module performs time alignment, format standardization and state fusion on multi-source heterogeneous data to generate comprehensive situation information. The display module displays all data, worker states and alarm information in real time through a three-dimensional virtual simulation interface.

2. The multi-source data real-time monitoring system for rescue drills according to claim 1, characterized in that: The data acquisition module comprises a microprocessor and various sensors, which send acquired data to the microprocessor for processing. The various sensors include a temperature sensor, a smoke concentration sensor, a motion capture camera, a worker wearable device and a pressure sensor and a flow sensor arranged in the positive pressure oxygen respirator. The temperature sensor is used to acquire environmental temperature in a smoke heat environment, the smoke concentration sensor is used to acquire smoke concentration, and the motion capture camera is used to locate the worker. The worker wearable device comprises a smart bracelet, an integrated photoelectric heart rate sensor and a blood oxygen saturation sensor. The pressure sensor is used to monitor cylinder pressure, and the flow sensor is used to monitor breathing frequency and instantaneous flow.

3. The multi-source data real-time monitoring system for rescue drill of claim 1, wherein: The data transmission module comprises an ad hoc network gateway, a switch and a router. The ad hoc network gateway forms a wireless mesh network, automatically finds an optimal path to receive data transmitted by various mobile sensors to the microprocessor and transmits the data to the data transmission module through the switch and the router.

4. The multi-source data real-time monitoring system for rescue drill of claim 1, wherein: The data processing module processing process comprises: Receiving raw data packets from the transmission module, and parsing various parameter values according to a predefined communication protocol; Establishing a unified time coordinate system with a server clock as a reference to eliminate time asynchronization problems; Let the source time of data packet i be The server receives the time as The time reference provided for data fusion is t rxi ; using a delay compensation algorithm, estimate the approximate generation time of the data packet on the server time axis Where the average delay δ avg is dynamically estimated by an exponential weighted moving average model ; Converting the parsed parameter values into structured data in standard units, comparing the structured data with corresponding threshold values and determining whether the worker is in a dangerous state; Storing all raw data and fused state data in a time series database to provide data support for real-time display and subsequent review.

5. The multi-source data real-time monitoring system for rescue drills according to claim 1, characterized in that: The conversion of the parsed parameter values into structured data in standard units and the comparison of the structured data with corresponding threshold values to determine whether the worker is in a dangerous state specifically comprise the following rules: Fusion rule 1: environmental temperature > 60℃, worker heart rate > 140bpm, determining that the worker is in a high risk state of heat stress collapse; Fusion rule 2: abnormal decrease in cylinder pressure and abnormal increase in worker breathing frequency, warning that the respirator may leak or the worker is extremely nervous.

6. The multi-source data real-time monitoring system for rescue drills according to claim 1, characterized in that: The display module comprises a command large screen, three-dimensional virtual simulation software, a VR all-in-one machine and a sound and light alarm device; a three-dimensional virtual environment consistent with a real training scene is constructed by using a Unity 3D engine, and corresponding elements in the virtual scene are driven by data by using a data processing module; the three-dimensional scene tracks and displays virtual avatars of each person in real time through a motion capture camera; the person displays physiological, equipment and environmental data in real time in the VR scene in the form of a pop-up window by wearing the VR all-in-one machine; the smoke concentration in the training scene range is simulated by a color particle effect, and the environmental temperature distribution is represented by color gradient; the person realizes the intelligent warning and alarm functions through the warning pop-up box, flashing red light and sound and light alarm in the VR scene; when the data processing module judges an abnormal state, the command large screen and the three-dimensional virtual simulation software simultaneously issue corresponding warnings, the command large screen pops up a warning box, the virtual avatar of the corresponding person flashes red light, and the sound and light alarm in the scene prompts the commander and the person himself.

7. The multi-source data real-time monitoring system for rescue drills according to claim 1, characterized in that: The display module also has a training review and analysis function, which replays the entire three-dimensional process after the training ends, and calls up the historical data of any team member for quantitative evaluation.

8. A multi-source data real-time monitoring method for rescue drills, characterized in that: The method comprises the following steps: S1: system hardware device deployment and initialization; in a large space training field simulating a disaster environment, motion capture cameras, smoke and heat simulation equipment and self-organizing network gateways are fixedly installed according to task requirements, and a set of integrated devices are worn by the training team members, the integrated devices comprising a positive pressure oxygen respirator with a built-in oxygen respirator parameter acquisition module, a smart bracelet and a VR all-in-one machine; the positive pressure oxygen respirator is built-in with a parameter acquisition device, an electronic alarm and a sensor, acquires and transmits corresponding parameters in the training process of the team member, and reminds the team member of matters needing attention when the parameters exceed the threshold; the smart bracelet acquires heartbeat and blood oxygen saturation physiological parameters; the sensor modules on all integrated devices are connected with microprocessors through Bluetooth, and the microprocessors are connected with a three-dimensional virtual scene through a wireless gateway; S2: after the training starts, each sensor works at a preset frequency, the pressure sensor and the flow sensor on the respirator continuously monitor the residual pressure value of the gas cylinder and the breathing flow waveform of the person, and the data is transmitted to the microprocessor through Bluetooth; S3: a data receiving service on the server continuously listens to a network port, receives various data packets from different gateways and base stations; according to a pre-defined communication protocol specification, the data packets are parsed to separate data source IDs, data types, data values and data original timestamps; the parsed data is sent to a data preprocessing module for timestamp unification, data filtering and format standardization; S4: the preprocessed data is stored in a cache for system fusion and calling, and a data fusion processing module performs periodic state judgment according to a pre-set rule library; combined with multi-source data, a higher level of comprehensive state judgment is performed through logical combination; S5: According to the received positioning data and motion capture camera recording data, the position and posture of the corresponding personnel virtual avatar in the three-dimensional scene are updated in real time, and the team member's body shape and action are determined by combining the motion capture camera shooting data and human body skeleton calculation and analysis; each team member can master his / her latest physiological data, equipment data and overall state calculated by the fusion module through the VR integrated machine in real time, and the panel color changes from green to yellow when the data exceeds the safety range; when receiving the high-risk alarm from the data fusion feedback, a red warning window pops up in the team member's VR field of view and the central command large screen, and a voice alarm is broadcast; at the same time, the corresponding personnel virtual avatar in the scene starts to flash frequently; S6: The commander performs time-space backtracking, data analysis and event labeling through the backstage management terminal; after the exercise is completed, the commander calls up the exercise data to reproduce the entire exercise process; Select any personnel to view the historical change curve of all parameters during the exercise process and perform on-screen comparative analysis; key event points are marked on the time axis and associated with the data at that moment to generate a detailed exercise evaluation report.

Images