Portable fire blast risk dynamic identification and risk tracing method and terminal

Abstract

The invention provides a portable fire blast risk dynamic identification and danger tracing method and terminal, and the method comprises the steps: collecting positioning data in real time through a GNSS positioning module; a high-speed analog-to-digital conversion chip ADC is used for carrying out high-frequency and synchronous processing on the gas concentration collected in real time; collecting field environment parameters in real time; performing dynamic threshold intelligent judgment to obtain a leakage judgment result, and executing a grading early warning mechanism; recording an inspection track in real time, and carrying out space gas concentration gradient analysis to obtain a gas concentration gradient; calculating a built-in space distance; judging the leakage direction according to the gas concentration gradient data, and intelligently tracing the leakage source according to the leakage judgment result; and uploading the security data to the cloud. According to the invention, the limitation of static and single-threshold alarm of an instrument is solved, and the technical problems of difficulty in spatial continuous sensing and automatic tracing positioning of gas leakage abnormity in a complex inspection environment with dynamic changes of multi-component combustible gas and environmental parameters are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of energy infrastructure detection, in particular to a portable explosion risk dynamic identification and hazard tracing method and terminal. BACKGROUND

[0002] In recent years, with the dense layout of urban energy pipelines, gas stations, oil and gas storage and transportation, and urban underground comprehensive pipe corridors, combustible gases (such as natural gas) and oil and gas vapors (such as gasoline vapor) are widely distributed in urban space. In engineering practice, affected by factors such as equipment aging, corrosion, accidental impact, or maintenance negligence, these gases may leak and rapidly spread in a limited space to form a combustible mixture with air. In addition, due to complex conditions such as ventilation, environmental temperature and humidity, and air pressure, the leaking gas is prone to accumulate in the space to form an explosive gas cloud. The actual distribution of mixed gas, explosion limit, and risk interval change dynamically with time and space, and traditional static detection and single-point alarm technology cannot accurately control local risks and cannot timely find and locate potential leakage sources.

[0003] For example, the existing utility model patent application document with publication number CN222764076U, the existing portable combustible gas detector is mainly a small handheld device, which uses a fixed concentration threshold strategy to alarm single-component gas on site. Although it is small in size, easy to operate, and suitable for preliminary screening in general places, it is difficult to distinguish the synergistic risk effect of multi-component mixed gas and cannot integrate the influence of environmental parameters on the explosion limit, resulting in high false alarm and missed alarm rates. In addition, such devices can only provide point or short-term data, and cannot meet the needs of continuous space inspection and leakage tracing.

[0004] And the existing invention patent application document with publication number CN120126288A, a gas user intelligent safety inspection system based on the Internet of Things, such a system mainly relies on fixed multi-point sensing network and cloud-based large models, although it can realize continuous monitoring and intelligent risk analysis, the device is large in size, deployment is complicated, and it relies on fixed installation, which cannot be flexibly adapted to emergency inspection and on-site mobile tracing scenarios, and is not conducive to rapid response and accurate positioning of sudden leakage events.

[0005] Currently, existing portable combustible gas detection and alarm devices generally use a single component fixed concentration threshold as the risk discrimination standard, without considering dynamic changes in environmental temperature, humidity, pressure, and the synergistic risk effect of multi-component gases, resulting in frequent false alarms, missed alarms, and alarm delays in complex urban working conditions.

[0006] The prior art adopts fixed threshold value, single gas discrimination, and is difficult to adapt to real-time intelligent patrol and leakage tracing in complex scenes, has high false alarm and missed alarm rates, and cannot perform spatial accurate positioning. In order to reduce false alarms and improve intelligent discrimination capability, some schemes attempt to deploy a large-volume dynamic threshold model or rely on cloud computing, but such a model has large volume and high power consumption, is difficult to integrate into a portable terminal, and usually relies on a fixed sensing network, cannot support on-site mobile patrol and spatial tracing of leakage sources, and has limited practical application scenarios.

[0007] In a complex and variable on-site environment, how to realize spatial continuous dynamic patrol, intelligent risk discrimination and abnormal tracing positioning of multi-component gases and environmental parameters in a portable terminal with extremely small volume and low power consumption, and break through the engineering bottleneck that existing large model schemes cannot be deployed locally and have insufficient mobile patrol and tracing capability, is still a big problem. Therefore, it is urgent to develop a portable combustible risk intelligent patrol and tracing terminal which integrates multi-component gas dynamic patrol, abnormal discrimination and tracing positioning, and has lightweight local model and end-cloud collaboration capability.

[0008] With the wide application of combustible gases such as natural gas and gasoline in urban infrastructure, gas leakage and aggregation phenomena in complex spatial environments are increasingly frequent. Combustible gases leak in different areas and aggregate in the form of mixed gases in poorly ventilated areas such as building basements, pipe trenches and cable channels, and are extremely easy to cause explosions when encountering open flames, seriously threatening urban public safety. Due to the complexity of on-site gas composition and the drastic change of environmental parameters, traditional portable detection instruments can only realize static threshold alarm, are difficult to sense the dynamic distribution of gases in time and continuously, and cannot track the leakage source, have high false alarm and missed alarm rates, lack spatial tracing capability, and cannot meet the local intelligent prevention and control and accident tracing needs of high-risk areas.

[0009] In summary, the prior art has the limitations of instrument static and single threshold alarm, and has the technical problem that it is difficult to continuously sense and automatically trace and position the spatial gas leakage anomaly in a complex patrol environment with dynamic changes of multi-component combustible gases and environmental parameters. SUMMARY

[0010] The technical problem to be solved by the present application is how to solve the limitations of instrument static and single threshold alarm in the prior art, and the technical problem that it is difficult to continuously sense and automatically trace and position the spatial gas leakage anomaly in a complex patrol environment with dynamic changes of multi-component combustible gases and environmental parameters.

[0011] The present application solves the above technical problems by adopting the following technical solution: a portable combustible risk dynamic identification and hazard tracing method comprises: S1, in combination with a GNSS positioning module, real-time acquisition of positioning data; using a high-speed analog-to-digital conversion chip ADC for high-frequency, synchronous processing of real-time collected gas concentration; real-time acquisition of on-site environmental parameters; S2, dynamic threshold intelligent discrimination is carried out to obtain leakage discrimination results, and a hierarchical early warning mechanism is executed; wherein a lightweight multi-layer perception MLP neural network model is adopted, including: setting an input layer, a hidden layer and a double output branch; S3, real-time recording of inspection trajectory, spatial gas concentration gradient analysis, gas concentration gradient is obtained; built-in spatial distance calculation; according to the gas concentration gradient data, the leakage direction is judged, according to the leakage discrimination result, the leakage source intelligent tracing is carried out; according to the on-site environmental parameters, the gas concentration gradient value is corrected in real time; according to the gas concentration of all inspection trajectory points and the gas concentration gradient, the spatial risk thermal map is generated in real time, and the safety data is obtained by processing; S4, an end cloud linkage and multi-scene adaptive safety response mechanism is adopted, the safety data is uploaded to the cloud according to the field signal strength and the network environment adaptive switching communication mode.

[0012] The application develops a portable explosion risk intelligent inspection and tracing terminal which integrates multi-component gas dynamic inspection, abnormality discrimination and source positioning, and has the functions of lightweight local model and end cloud cooperation. The application breaks through the limitations of traditional static and single threshold alarm, and realizes the spatial continuous perception and automatic source positioning of gas leakage abnormality by the portable terminal in the complex inspection environment of dynamic changes of multi-component flammable gas and environmental parameters.

[0013] In a more specific technical solution, in S1, fixed frequency is used for data acquisition, and a series of continuous data points are recorded:

[0014] The main control unit is used for real-time filtering, denoising, data completion and automatic calibration of all sensor signals:

[0015] In the formula, M is the size of the sliding window.

[0016] In a more specific technical solution, in S2, the double output branch includes: a dynamic explosion limit regression branch and a risk classification branch.

[0017] The application is based on the local lightweight intelligent model and the end cloud cooperation mechanism, and carries out explosion risk intelligent inspection and danger tracing. The application can fuse multi-component gas and environmental parameter data, adaptively adjust the explosion limit, realize high-density spatial inspection, risk classification alarm and leakage source positioning.

[0018] In a more specific technical solution, in S3, the spatial gas concentration gradient analysis is performed using the following logic:

[0019] In the formula, is the difference between the gas concentrations of two consecutive points on the inspection path; That is, a certain gas concentration value is measured at the moment; That is, at the previous moment a certain gas concentration value is measured at the moment; is the spatial distance between the two points (obtained by GNSS module positioning); corresponds to the spatial coordinates of the measurement point at the current moment; corresponds to the spatial coordinates of the measurement point at the previous moment.

[0020] In a more specific technical solution, in S3, the built-in spatial distance calculation is performed using the following logic:

[0021] In the formula, R is 6371 km, and X and Y need to be converted to radians for calculation.

[0022] In a more specific technical solution, in the leakage direction judgment process of S3, the terminal judges the concentration gradient of different gases at consecutive points not less than 2 positions in real time, and when the following conditions are met, it is determined that the concentration gradient of the current gas is continuously increasing in a positive direction, and the concentration gradient of the remaining gas is in a stable state, and the guidance inspection information is issued:

[0023] When the following conditions are met, it is determined that the concentration of the current gas has an abnormal mutation, and an alarm and a traceability information are issued:

[0024] In a more specific technical solution, in S3, the gas concentration gradient value is corrected in real time according to the obtained environmental parameters:

[0025] ; wherein the correction factor comes from an empirical formula:

[0026] In the formula, is the standard environmental parameter; is an empirical correction coefficient, which can be determined through preliminary experiments and terminal model training.

[0027] In a more specific technical solution, in S3, the inspection data is uploaded to the cloud GIS platform in real time through the wireless communication module, and the GIS platform generates a spatial risk heat map in real time according to the gas concentration and gradient change information of all inspection track points:

[0028] In the formula, The risk heat value of a certain point in space; The concentration of various gases of the first The spatial distance between the to-be-analyzed position and the first The spatial distance between the to-be-analyzed position and the first The spatial distance between the to-be-analyzed position and the first The spatial influence radius parameter.

[0029] The application realizes local dynamic fusion discrimination of multi-component gas and environmental parameters through a portable lightweight multi-parameter intelligent discrimination terminal, can record spatial trajectory and gas concentration change in real time during continuous on-site inspection, and uses concentration distribution and gradient analysis to intelligently locate the leakage source. The terminal can complete hierarchical alarm and traceability function locally without relying on the cloud, significantly improving the risk identification accuracy and inspection traceability efficiency, meeting the intrinsic safety and emergency demand in high-risk scenarios.

[0030] In a more specific technical solution, when the inspection personnel enters the communication blind area, the inspection terminal continuously performs data and concentration gradient calculation and caching operation; when the inspection terminal reenters the network coverage area, the data transmission mechanism is automatically triggered to batch upload the cached data to the cloud platform.

[0031] In a more specific technical solution, the portable combustion risk dynamic identification and dangerous traceability terminal comprises: The GNSS positioning module is used to collect positioning data in real time; the high-speed analog-to-digital conversion chip ADC is used to perform high-frequency and synchronous processing on the real-time collected gas concentration; and the on-site environmental parameters are collected in real time. The leakage discrimination module is used for dynamic threshold intelligent discrimination to obtain a leakage discrimination result and execute a hierarchical early warning mechanism; wherein, a lightweight multi-layer perception MLP neural network model is adopted, including: setting an input layer, a hidden layer and a double-output branch, and the leakage discrimination module is connected with the GNSS positioning module.

[0032] The leakage tracing module is used to record the inspection track in real time, analyze the spatial gas concentration gradient, obtain the gas concentration gradient, calculate the built-in spatial distance, judge the leakage direction according to the gas concentration gradient data, intelligently trace the leakage source according to the leakage discrimination result, correct the gas concentration gradient value in real time according to the on-site environment parameters, and generate a spatial risk heat map in real time according to the gas concentration of all inspection track points and the gas concentration gradient, to obtain safety data, and the leakage tracing module is connected with the leakage discrimination module and the GNSS positioning module. The data uploading module is used to adopt end-cloud linkage and multi-scene adaptive safety response mechanism, adaptively switch the communication mode according to the on-site signal strength and network environment, upload the safety data to the cloud, and the data uploading module is connected with the leakage tracing module.

[0033] Compared with the prior art, the present application has the following advantages: The present application develops a portable explosion risk intelligent inspection and tracing terminal which integrates multi-component gas dynamic inspection, abnormality discrimination and source positioning, and has the functions of lightweight local model and end-cloud cooperation.

[0034] The present application can dynamically adjust the explosion limit by fusing multi-component gas and environmental parameter data, realize high-density spatial inspection, risk grading alarm and leakage source positioning.

[0035] The present application can dynamically fuse and discriminate multi-component gas and environmental parameters locally, record the spatial track and gas concentration change in real time during continuous on-site inspection, and intelligently locate the leakage source by using the concentration distribution and gradient analysis.

[0036] The present application solves the problems of instrument static and single threshold alarm in the prior art, and realizes spatial continuous perception and automatic source positioning of gas leakage in a complex inspection environment with dynamic changes of multi-component combustible gas and environmental parameters. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 It is a basic step schematic diagram of the portable explosion risk dynamic identification and hazard tracing method of embodiment 1 of the present application. Figure 2This is a schematic diagram of the terminal model network structure in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram illustrating the specific steps of the dynamic discrimination process of the terminal model in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram illustrating the specific steps of the terminal model discrimination process in Embodiment 1 of the present invention; Figure 5 This is a schematic diagram illustrating the measurement of various parameters in Embodiment 1 of the present invention; Figure 6 This is a schematic diagram of the portable terminal structure of Embodiment 1 of the present invention. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0039] Example 1 like Figure 1 and Figure 5 As shown, the portable dynamic identification and hazard tracing method for combustion and explosion risks provided by this invention includes the following basic steps: S1. Data input and preprocessing; To achieve dynamic perception and hazard tracing during inspections, this terminal employs an array-type gas sensor layout and a synchronous environmental parameter acquisition mechanism. Combined with a GNSS positioning module, it acquires various data in real time. The gas sensor array includes a catalytic combustion methane (CH4) sensor and a photoionization (PID) volatile organic compound (VOC) sensor, connected to the main control unit via standardized micro-connectors. A high-speed analog-to-digital converter (ADC) chip performs high-frequency, synchronous processing on the real-time gas concentration data. The terminal also integrates high-precision temperature, humidity, and air pressure sensors, connected to the main control unit via standard digital interfaces (such as I2C and SPI) to acquire on-site environmental parameters in real time.

[0040] In this embodiment, the terminal collects the above data at a fixed frequency (1 Hz) and records a series of continuous data points:

[0041] The main control unit performs real-time filtering, noise reduction, data completion, and automatic calibration on all sensor signals.

[0042] Wherein M is a sliding window size, and a typical value is 5-10 sampling points.

[0043] After generating reliable timing data flow, external disturbance interference is effectively suppressed by using digital filtering algorithm and real-time interpolation algorithm, so as to ensure data stability and accuracy in the dynamic inspection process, and provide high-quality data source for intelligent model discrimination.

[0044] S2, dynamic threshold intelligent discrimination and grading early warning mechanism; In the embodiment, in order to ensure the real-time and accuracy of the terminal in discriminating the field combustion risk, the application adopts a light multi-layer perception (MLP) neural network model.

[0045] The input neurons of the neural network are 5+n items of key component concentrations (methane-CH4, combustible volatile organic compounds-VOC) of the mixed gas of natural gas-oil gas-air, and environmental temperature, humidity, pressure, etc., the number of neurons in the three hidden layers is different, and is respectively set to 128, 64 and 32. A double-output branch is arranged after the hidden layer, including a dynamic explosion limit regression branch and a risk classification branch. The overall structure is as shown in Figure 2 .

[0046] As shown in Figure 3 , in the embodiment, the terminal model dynamic discrimination process includes: S21, gas and environmental parameter acquisition; S22, data noise reduction, abnormality rejection, etc.; S23, measurement and calculation to obtain . S24, data normalization and standardization; S25, inputting the processed data into the neural network; S26, outputting the dynamic change reading range / explosion limit interval; S27, . S28, if not, the system output is safe; S29, if yes, the softmax function normalization processing is performed; S210, the system outputs the corresponding danger level and corresponding point coordinates, and reports to the cloud; S211, different alarm signals are output according to the level.

[0047] The hierarchical alarm not only reminds the on-site personnel of safety in the first time, but also provides important data for subsequent "leak point finding" and analysis of the inspection route. Whenever the instrument detects an increased risk level or a significant change in gas concentration, the system automatically records the location, time, and specific risk level at which the event occurred, and uploads it to the cloud GIS system. Subsequent systems can calculate the possible leakage source and risk change path based on these key nodes.

[0048] S3, inspection trajectory recording and intelligent traceability mechanism of leakage source; In this embodiment, real-time inspection trajectory recording is achieved through a high-precision GNSS positioning chip, and the control chip and terminal discrimination model with built-in spatial distance calculation and gas concentration gradient analysis algorithm modules are used to achieve rapid intelligent traceability of the leakage source.

[0049] The traceability method is based on the real-time recorded gas concentration change trend, and the specific process is as follows: 1. Spatial gas concentration gradient calculation: During the inspection process, the terminal automatically measures the gas concentration and environmental parameter changes every 1 s, and the built-in intelligent algorithm automatically analyzes the spatial gradient change of the continuously collected gas concentration in real time:

[0050] In the formula, is the difference between the gas concentrations of two consecutive points on the inspection path; is the gas concentration value measured at the moment; is the gas concentration value measured at the moment; is the spatial distance between the two points (obtained by GNSS module positioning); is the spatial coordinate of the measurement point at the current moment; is the spatial coordinate of the measurement point at the previous moment.

[0051] The spatial distance is calculated by the latitude and longitude conversion formula:

[0052] In the formula, R is 6371 km, and X and Y need to be converted to radians for calculation.

[0053] The parameter diagram is shown in Figure 5 .

[0054] 2. Leakage direction judgment: The terminal judges the concentration gradient of different gases at multiple consecutive points in real time. When the concentration gradient of a certain gas continuously increases positively while the concentration gradients of other gases are relatively stable, that is,​​​​

[0055] Indicates that the corresponding gas is leaking and the inspector is currently approaching the source area of the leak. The terminal screen indicates the direction of concentration increase through arrows, guiding the inspector to quickly locate. Conversely, when the gradient of multiple consecutive points decreases, it indicates that the inspector is moving away from the source of the leak, and the terminal clearly reminds and suggests reverse inspection.

[0056] If the gas concentration abnormally mutates:

[0057] When the rate of change of gas concentration abnormally increases, and Indicates that the inspector may have just entered a high-risk area of abnormal gas accumulation, and the leak may be more serious or have spread. At this time, the system determines that there is an anomaly, automatically issues an alarm and suggests tracing back.

[0058] 3. Optimize the accuracy of traceability positioning in combination with environmental parameters: Consider the influence of environmental temperature, humidity, air pressure, etc. on gas diffusion, and the terminal corrects the gas concentration gradient value in real time, as follows:

[0059] Correction factor From the empirical formula:

[0060] In the formula, is the standard environmental parameter; is the empirical correction coefficient, which can be determined through preliminary experiments and terminal model training.

[0061] 4. Cloud-based GIS assisted accurate traceability: The inspection data is uploaded to the cloud GIS platform in real time through the wireless communication module, and the GIS platform generates a spatial risk heat map in real time according to the gas concentration and gradient change of all inspection track points. The heat map calculation formula is as follows:

[0062] In the formula, is the risk heat value of a certain point in space; is the concentration of various gases at the th measurement point; is the spatial distance between the position to be analyzed and the th measurement point; is the spatial influence radius parameter, which is usually determined according to the concentration of the leaking gas.

[0063] Through terminal uploading data, the cloud GIS platform can intuitively display high concentration areas and mark suspected leakage source points, providing basis for rapid decision-making and subsequent on-site disposal.

[0064] S4, end-to-cloud linkage and multi-scenario adaptive security response mechanism; The portable explosion risk dynamic threshold alarm terminal of the embodiment integrates multiple wireless communication modes (such as BLE, LoRa, NB-IoT, 4G / 5G, etc.), which can be adaptively switched according to the signal strength and network environment on site, to ensure that the key safety data is uploaded to the cloud in real time and reliably. All local discrimination results, dynamic thresholds, alarm events, and corresponding geographic location raw monitoring data are transmitted in encrypted form to ensure information security.

[0065] S5, communication blind area data closed loop mechanism.

[0066] In the embodiment, the terminal has a communication blind area data buffering mechanism. When the inspector enters a communication blind area, the terminal still calculates and records data and concentration gradient in real time locally. When the inspection terminal reenters the network coverage area, the data retransmission mechanism is automatically triggered to upload the buffered data to the cloud platform in batches, realizing the continuity of inspection data and the integrity of event traceability.

[0067] Embodiment 2 As shown in Figure 6 The terminal of the present application has a compact design, and the integrated housing (108) is made of high-strength explosion-proof engineering plastic or lightweight alloy material. The size of the body is preferably 150 mm x 65 mm x 28 mm, and the total mass is ≤300 g, which is convenient for single-handed holding, waist clamping, or shoulder strap suspension. The device shell reaches IP67 and above protection level, with the ability of dustproof, waterproof, corrosion-resistant, and drop-resistant. The device appearance 6 is shown.

[0068] Structurally, the terminal integrates several functional modules through partitioning, including: a multi-component gas sensor array (101), an environmental parameter acquisition module (102), a main control and local intelligent discrimination unit (103), a multi-mode wireless communication module (104), a local multi-level early warning output unit (105), a positioning and navigation unit-GNSS module (106), and an interactive display unit (107).

[0069] When the inspector holds or wears the terminal for real-time mobile inspection, the terminal can real-time collect the concentration changes of natural gas (methane) and oil and gas vapor (VOC) as well as environmental data such as temperature, humidity, and air pressure, and real-time judge the explosion risk level through internal algorithms. At the same time, the terminal also uses the built-in high-precision GNSS module to record the inspector's position coordinates, trajectory, and gas concentration change trend in real time, and displays the current coordinates, inspection trajectory, gas concentration gradient, and risk level on the OLED screen, to assist the inspector in quickly locating and tracing the leakage source.

[0070] In summary, the present application develops a portable explosion risk intelligent inspection and tracing terminal which integrates multi-component gas dynamic inspection, abnormality discrimination and source location, and has the functions of lightweight local model and end-cloud collaboration. The present application breaks through the limitations of traditional static and single threshold alarm, and realizes the spatial continuous perception and automatic source location of gas leakage abnormality in the complex inspection environment of dynamic changes of multi-component flammable gas and environmental parameters.

[0071] The present application is based on the local lightweight intelligent model and end-cloud collaboration mechanism, and performs explosion risk intelligent inspection and tracing. The present application can fuse multi-component gas and environmental parameter data, adaptively and dynamically adjust the explosion limit, and realize high-density spatial inspection, risk classification alarm and leakage source location.

[0072] The present application realizes the local dynamic fusion and discrimination of multi-component gas and environmental parameters through a portable lightweight multi-parameter intelligent terminal, can record the spatial trajectory and gas concentration change in real time during continuous on-site inspection, and can use the concentration distribution and gradient analysis to intelligently locate the leakage source. The terminal can complete the classification alarm and source location function locally without relying on the cloud, which significantly improves the risk identification accuracy and inspection and tracing efficiency, and meets the intrinsic safety and emergency demand in high-risk scenarios.

[0073] The present application solves the limitations of instrument static and single threshold alarm in the prior art, and realizes the spatial continuous perception and automatic source location of gas leakage abnormality in the complex inspection environment of dynamic changes of multi-component flammable gas and environmental parameters.

[0074] The above examples are only used to illustrate the technical solutions of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A portable method for dynamic identification of explosion risk and tracing of danger, characterized in that, The method comprises: S1, combining with the GNSS positioning module, real-time acquisition of positioning data; using a high-speed analog-to-digital conversion chip ADC for high-frequency, synchronous processing of real-time collected gas concentration; real-time acquisition of field environment parameters; S2, dynamic threshold intelligent discrimination is performed to obtain a leakage discrimination result, and a hierarchical early warning mechanism is executed; wherein a lightweight multi-layer perception MLP neural network model is adopted, including: setting an input layer, a hidden layer and a double-output branch; S3, real-time recording of inspection trajectory, spatial gas concentration gradient analysis, gas concentration gradient is obtained, built-in spatial distance calculation is performed, leakage direction judgment is performed according to the gas concentration gradient data, leakage source intelligent tracing is performed according to the leakage discrimination result, the gas concentration gradient value is corrected in real time according to the field environment parameters, and a spatial risk heat map is generated in real time according to all inspection trajectory point gas concentrations and the gas concentration gradient, and safety data is obtained by processing; S4, an end-cloud linkage and multi-scene adaptive safety response mechanism is adopted, the safety data is uploaded to the cloud according to the field signal strength and the network environment adaptive switching communication mode.

2. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, In S1, fixed frequency is used for data acquisition, and a series of continuous data points are recorded: Real-time filtering, denoising, data completion and automatic calibration of all sensor signals are performed by using a master control unit: In the formula, M is the size of the sliding window.

3. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, In S2, the double-output branch includes a dynamic explosion limit regression branch and a risk classification branch.

4. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, In S3, spatial gas concentration gradient analysis is performed using the following logic: In the formula, The difference in gas concentration between two consecutive points along the inspection path; Right now The concentration value of a certain gas is measured at any time. That is, at the previous moment The concentration value of a certain gas was measured at that time; This represents the spatial distance between the two corresponding points (obtained through GNSS module positioning). Corresponding to the current Spatial coordinates of the measurement point at any given time; Corresponding to the previous moment Spatial coordinates of the measurement point.

5. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, In S3, the built-in spatial distance calculation is performed using the following logic: In the formula, R is 6371 km, and X and Y need to be converted to radians for calculation.

6. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, In the leakage direction judgment process of S3, the terminal judges the concentration gradient of different gases of continuous position points not less than 2 in real time, and when the following conditions are met, it is determined that the concentration gradient of the current gas continuously increases in a positive direction, and the concentration gradient of the remaining gas is in a stable state, and guide inspection information is issued: When the following conditions are met, it is determined that the concentration of the current gas abnormally mutates, and an alarm and a suggestion tracing information are issued: 。 7. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, In S3, the gas concentration gradient value is corrected in real time according to the obtained and environmental parameters: wherein the correction factor From empirical formula: In the formula, is a standard environmental parameter; is an empirical correction coefficient, which can be determined through preliminary experiments and terminal model training.

8. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, In S3, the inspection data is uploaded to the cloud GIS platform in real time through the wireless communication module, and the GIS platform generates a spatial risk heat map in real time according to all inspection trajectory point gas concentrations and gradient change information: wherein is the risk thermodynamic value for a certain point in space; is the concentration of the various gases for the measurement point; is the spatial distance between the position to be analyzed and the first measurement point; is the spatial influence radius parameter.

9. The portable explosion risk dynamic identification and hazard tracing method according to claim 1, characterized in that, When the inspection personnel enters a communication blind area, the inspection terminal continuously performs data and concentration gradient calculation and caching operation; when the inspection terminal reenters the network coverage area, the data retransmission mechanism is automatically triggered, and the cached data is uploaded to the cloud platform in batches.

10. A portable terminal for dynamic identification of explosion risk and tracing of danger, characterized in that, The terminal comprises: A GNSS positioning module is used to acquire positioning data in real time; a high-speed analog-to-digital conversion chip ADC is used to perform high-frequency, synchronous processing of real-time collected gas concentration; and field environment parameters are acquired in real time. The leakage discrimination module is used to perform dynamic threshold intelligent discrimination to obtain a leakage discrimination result and execute a hierarchical early warning mechanism; wherein a lightweight multi-layer perception (MLP) neural network model is adopted, including: setting an input layer, a hidden layer and a double-output branch, the leakage discrimination module being connected with the GNSS positioning module; The leakage tracing module is used to record a patrol track in real time, perform spatial gas concentration gradient analysis to obtain a gas concentration gradient, perform built-in spatial distance calculation, judge a leakage direction according to the gas concentration gradient data, intelligently trace a leakage source according to the leakage discrimination result, correct a gas concentration gradient value in real time according to the field environment parameters, and generate a spatial risk heat map in real time according to all patrol track point gas concentrations and the gas concentration gradient to process safety data, the leakage tracing module being connected with the leakage discrimination module and the GNSS positioning module; The data uploading module is used to adopt end-cloud linkage and a multi-scene adaptive safety response mechanism, adaptively switch a communication mode according to field signal strength and a network environment, and upload the safety data to a cloud, the data uploading module being connected with the leakage tracing module.

Images