Portable fire smoke particle rapid collection and storage evaluation device

Abstract

The invention discloses a portable fire smoke particle rapid collection and storage evaluation device, and relates to the technical field of smoke particle data evaluation.According to the device, a fire scene space is divided into three-dimensional grids, and fire behavior marker feature values of all nodes are obtained; after the characteristic value exceeds an activation threshold value, generating a fire chemical gradient vector indicating the concentration increasing direction through spatial vector operation according to adjacent node data, and guiding the robot to move; positioning a local fire characteristic peak point through iterative optimization; verifying whether the point is a global optimal peak point or not through extended detection; and adapting to the dynamic change of the fire scene, and reversely tracking to a real origin point. According to the method, the conversion from passive sampling to active tracing is realized, finally, the chemical characteristic source with the highest evidence value in the fire scene can be autonomously and accurately positioned, and the accuracy and efficiency of fire investigation are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of smoke particle data evaluation technology, and more specifically, this application relates to a portable device for rapid collection, storage and evaluation of fire smoke particles. Background Technology

[0002] In fields such as fire accident investigation, fire safety monitoring, and environmental assessment, accurately collecting smoke particle samples from fire scenes and determining their spatial distribution characteristics is a crucial technical step in tracing fire sources, analyzing fire causes, and assessing pollutant diffusion. Traditional smoke particle collection methods mainly rely on the experience of investigators, involving manual or simple automated sampling at pre-set locations. This approach has significant limitations; it is essentially a passive and static sampling mode, ill-suited to the highly complex and dynamic nature of fire scene environments.

[0003] Existing technologies typically rely on pre-set fixed sampling paths or threshold triggering mechanisms based on a single sensor. This approach cannot detect spatial gradient changes in the chemical composition of smoke particles, thus failing to intelligently track the spread of the smoke plume or pinpoint the most evidence-worthy sources of chemical signatures. After a fire enters its decay phase, the concentration distribution of complete combustion markers may deviate from the initial ignition point, causing the collected samples to fail to accurately reflect the origin of the fire. Furthermore, traditional point sampling methods lack mechanisms to verify the representativeness and global optimality of sampling points, easily falling into the trap of being misled by localized high-concentration areas. The lack of systematic correlation between the collection process and the spatiotemporal information and environmental context of the samples results in insufficient evidentiary force and traceability of subsequent laboratory analysis results.

[0004] To address the aforementioned issues, there is an urgent need in this field for a technical solution capable of autonomous and intelligent dynamic tracing and sampling in complex fire environments. This solution needs to overcome the limitations of static sampling, shifting from passive collection to proactive source tracing. It should be able to adaptively adjust search strategies based on real-time environmental data, accurately locate the source of chemical evidence, and ensure the integrity of the sampling process and the reliability of the evidence chain.

[0005] In summary, existing technologies for collecting fire smoke particles have a fundamental flaw: they cannot autonomously and accurately locate the source of the chemical characteristics of fire smoke particles in complex and dynamic fire environments. Summary of the Invention

[0006] To address the aforementioned technical problems, a portable device for rapid collection, containment, and assessment of fire smoke particles is provided. This technical solution resolves the issues raised in the background section.

[0007] In a first aspect, embodiments of this application provide a portable rapid collection and containment assessment device for fire smoke particles, characterized in that the device includes a data acquisition module: used to divide the target fire space into several three-dimensional grid nodes and acquire the fire marker feature values ​​collected by the target robot at the three-dimensional grid nodes; a first judgment and positioning module: used to determine whether the fire marker feature values ​​are greater than a preset activation threshold, and if so, to perform the following operations: based on the fire marker feature values ​​of the three-dimensional grid node and adjacent three-dimensional grid nodes, to obtain the fire chemical gradient vector through spatial vector operation, and to control the target robot to move to a new three-dimensional grid node along the direction of the fire chemical gradient vector, starting from the three-dimensional grid node; The second judgment and positioning module is used to repeatedly execute the operations in the first judgment and positioning module until the following convergence condition is met: the fire marker feature value of the new three-dimensional grid node is greater than the fire marker feature values ​​of all its adjacent grid nodes and is recorded as the fire feature peak point; the third judgment and positioning module is used to record the outer adjacent grid nodes of all adjacent grid nodes of the fire feature peak point as secondary nodes, obtain and judge whether the fire marker feature value of the secondary node is greater than the fire marker feature value of the fire feature peak point, and if so, record the corresponding secondary node as the new fire feature peak point; the collection and sealing execution module is used to control the target robot to perform smoke particle sample collection and sealing operations at the fire feature peak point.

[0008] Secondly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned portable fire smoke particle rapid collection and containment assessment device.

[0009] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0010] 1. Instead of random or fixed-path sampling, it can autonomously initiate and execute a complete tracking procedure after detecting that the characteristic value of fire markers exceeds a threshold. By calculating the fire chemical gradient vector to guide the robot's navigation, it replaces the method that relies entirely on manual judgment or simple program control, improving the intelligence and automation of sampling and positioning.

[0011] 2. By iteratively optimizing to find the peak points of fire characteristics, and then using a third-level judgment and positioning module to perform secondary node judgment, a rigorous mathematical logic was established to ensure that the located points are the optimal points of chemical characteristics in the local area. This dual verification mechanism greatly avoids the risk of collecting low-value or misleading samples due to environmental fluctuations or local concentration anomalies, fundamentally guaranteeing the representativeness and evidentiary validity of the collected samples.

[0012] 3. By dividing the fire scene space into a three-dimensional grid and using spatial vector operations, it is possible to perceive and utilize the distribution gradient of chemical markers in three-dimensional space, rather than relying solely on planar or single-point information. This is more accurate than traditional two-dimensional or linear perception in reflecting the actual diffusion and distribution patterns of smoke in the fire scene, thereby realizing the active perception and utilization of the three-dimensional spatial characteristics of smoke particle distribution. Attached Figure Description

[0013] Figure 1 A schematic diagram of the portable fire smoke particle rapid collection and containment assessment device provided in the embodiments of this application;

[0014] Figure 2 A schematic diagram of the logic flow of the portable fire smoke particle rapid collection and sealing evaluation device provided in the embodiments of this application.

[0015] Figure 3 This is a schematic diagram of the logic flow for determining the potential origin provided in the embodiments of this application.

[0016] Figure 4 This is a schematic diagram of the logic flow for dynamic mesh optimization provided in an embodiment of this application. Detailed Implementation

[0017] This application provides a portable device for rapid collection, containment, and evaluation of fire smoke particles, which solves the fundamental defect in the prior art that cannot autonomously and accurately locate the source of the chemical characteristics of fire smoke particles in complex and dynamic fire environments.

[0018] In existing technologies, fire smoke particle collection mainly relies on fixed sampling paths or single sensor triggering mechanisms, failing to detect changes in chemical gradients. Traditional methods struggle to track smoke plume diffusion paths in dynamic fire environments, and samples deviate from the ignition point during the decay phase, leading to data distortion. Passive sampling lacks a global verification mechanism, is susceptible to interference from localized high-concentration areas, and suffers from insufficient spatiotemporal correlation of sample information, affecting the reliability of the evidence chain.

[0019] To address the aforementioned issues, a dynamic tracking mechanism needs to be established by understanding the spatial gradient distribution patterns of chemical characteristics in fire areas. This involves dividing the fire area into a three-dimensional grid, acquiring real-time fire marker feature values, and using vector operations to determine the gradient direction, guiding the robot towards high-concentration areas. Convergence conditions ensure local peak point localization, and secondary verification at peripheral nodes prevents the omission of higher-concentration areas. Finally, sample collection is performed at the optimal location.

[0020] Among them, the three-dimensional grid node refers to discretizing the fire space into regularly arranged coordinate points. Specifically, a spatial coordinate system can be constructed using lidar or visual SLAM technology, and each node stores the location coordinates and chemical characteristic value data.

[0021] Fire marker characteristic values ​​refer to quantitative indicators that reflect the concentration of combustion products. Specifically, they can be obtained by detecting the spectral intensity or ion current intensity of a specific chemical substance using a multispectral sensor or mass spectrometer.

[0022] Spatial vector operations refer to the calculation of direction vectors based on the differences in eigenvalues ​​of adjacent nodes. Specifically, a three-dimensional difference algorithm can be used to generate a gradient vector field.

[0023] Convergence criteria refer to the standards for determining when robot movement terminates. Specifically, it can be set as follows: the feature value of the current node is higher than that of all neighboring nodes, and no higher value is found in three consecutive iterations. Second-level nodes refer to the second-level neighboring nodes surrounding the peak point, which can be determined by traversing the spatial topology using a breadth-first search algorithm.

[0024] The above technical solutions enable precise location of smoke particle sources in fire scenes, enhancing the value of sample evidence. Dynamic gradient tracking adapts to changes in the fire environment, avoiding the mechanical limitations of fixed paths. A multi-level verification mechanism ensures global optimization of collection points, addressing the susceptibility of traditional methods to local interference. Automated sampling processes reduce human error, enhancing data reliability and the integrity of the evidence chain.

[0025] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0026] like Figure 1 The diagram shown is a structural schematic of a portable fire smoke particle rapid collection and sealing assessment device provided in this application embodiment, including: a collection module: used to divide the target fire space into several three-dimensional grid nodes and obtain the fire marker feature values ​​collected by the target robot at the three-dimensional grid nodes; a first judgment and positioning module: used to determine whether the fire marker feature value is greater than a preset activation threshold, and if so, to perform the following operations: based on the fire marker feature values ​​of the three-dimensional grid node and adjacent three-dimensional grid nodes, to obtain the fire chemical gradient vector through spatial vector operation, and to control the target robot to move to a new three-dimensional grid node along the direction of the fire chemical gradient vector, starting from the three-dimensional grid node; the second... The second judgment and positioning module is used to repeatedly execute the operations in the first judgment and positioning module until the following convergence condition is met: the fire marker feature value of the new 3D grid node is greater than the fire marker feature values ​​of all its adjacent grid nodes and is recorded as the fire feature peak point; the third judgment and positioning module is used to record the outer adjacent grid nodes of all adjacent grid nodes of the fire feature peak point as secondary nodes, obtain and judge whether the fire marker feature value of the secondary node is greater than the fire marker feature value of the fire feature peak point, and if so, record the corresponding secondary node as the new fire feature peak point; the collection and sealing execution module is used to control the target robot to perform smoke particle sample collection and sealing operations at the fire feature peak point.

[0027] Among them, the adjacent grid nodes include the current node's 6 orthogonal neighboring nodes and 12 oblique neighboring nodes in the three-dimensional space, for a total of 18 adjacent grid nodes.

[0028] like Figure 2 The diagram shown is a schematic flowchart of the portable fire smoke particle rapid collection and sealing evaluation device provided in the embodiment of this application.

[0029] Furthermore, it also includes a fire stage intelligent discrimination module: used to obtain the characteristic values ​​of fully burned markers and incompletely burned markers contained in the characteristic values ​​of fire markers at the peak point of the fire characteristic, and calculate the ratio between the two; if the ratio exceeds the preset decay stage threshold, it is determined that the fire is in the decay stage and a stage switching command is generated; the target robot responds to the stage switching command and performs the potential origin determination operation.

[0030] In this embodiment, the characteristic value of complete combustion marker refers to the concentration or intensity index of stable products generated when the combustion reaction is fully carried out. Specifically, it can be achieved by using sensor detection data of carbon dioxide, water vapor and aerosols generated by complete oxidation, which is used to characterize the intensity of the completed combustion reaction in the fire scene.

[0031] The characteristic value of incomplete combustion markers refers to the concentration or intensity of intermediate products generated when the combustion reaction is not complete. Specifically, it can be achieved by using spectral analysis data of carbon monoxide, hydrocarbons, and incompletely burned particulate matter to reflect the ongoing combustion reaction state in the fire scene.

[0032] The decay stage threshold refers to the critical ratio of the concentrations of completely combusted products to incompletely combusted products. This ratio can be determined through historical fire data analysis or experimental calibration and serves as the condition for triggering a change in fire stage. The potential origin determination operation refers to the process by which the target robot adjusts its detection strategy based on the characteristics of the decay stage. This can be achieved by switching core detection indicators and retrospectively tracing the chemical gradient, used to locate residual fire source traces after the fire has weakened.

[0033] This application can automatically adjust the detection strategy when the fire enters the decay stage, avoid sampling deviation caused by changes in the concentration distribution of combustion products, ensure that the collected smoke particle samples maintain a strong correlation with the fire source location, and provide more reliable evidence for the analysis of the cause of the fire.

[0034] Furthermore, the potential origin determination operation includes: switching the core fire marker feature value to be collected from the complete combustion marker feature value to the volatile organic compound (VOC) residue feature value; taking the current fire feature peak point as the new starting point, recalculating the fire chemical gradient vector based on the VOC residue feature value, and controlling the target robot to move and probe in the opposite direction of the fire chemical gradient vector; during the movement and probe, acquiring the temperature on the movement path in real time; if the temperature of a certain grid node is lower than the preset temperature threshold, determining that the area defined by the grid node according to the predefined radius is a cooling zone where the heat source has been extinguished, abandoning sampling in this area, and continuing to move and probe until the temperature of a certain grid node is higher than the preset temperature threshold and the VOC residue feature value is higher than the preset feature threshold, and recording the grid node as a potential origin point; performing a fine component scan on the potential origin point, if a decomposition marker of a specific fire source material is detected, confirming it as the optimal collection point in the decay stage, and triggering the collection and sealing execution module.

[0035] In this embodiment, as Figure 3 The diagram shown is a schematic flowchart of the potential origin determination operation provided in the embodiments of this application.

[0036] The logic of moving in the opposite direction of the gradient vector is that during the decay phase, the volatile organic compound residue diffuses outward from the origin point, and its concentration gradient points in the direction of diffusion, so the opposite direction points to the origin point.

[0037] The core fire marker characteristic value refers to the key chemical indicator used to determine the stage of fire development. Specifically, it can be achieved by using a multi-channel gas sensor array, which can adapt to different fire stages by dynamically switching detection modes.

[0038] The residual characteristic value of volatile organic compounds refers to the concentration index of organic matter produced by incomplete combustion in a fire.

[0039] The reverse direction of the fire chemical gradient vector refers to the path opposite to the direction of combustion product diffusion. Specifically, the gradient change trend can be calculated using a spatial vector synthesis algorithm to guide the robot to track in reverse.

[0040] Temperature threshold refers to the critical value for determining whether a heat source has been extinguished. Specifically, it can be achieved by using an infrared temperature measurement module combined with an ambient temperature compensation algorithm to exclude invalid sampling areas.

[0041] Fine composition scanning refers to high-precision analysis of the chemical composition of a target area, which can be achieved by using laser-induced breakdown spectroscopy technology to identify the fire source material by detecting specific elements or compounds.

[0042] Specifically, when a fire enters the decay phase, the concentration distribution of markers of complete combustion may deviate from the actual fire source location. At this point, the core detection target is switched to the residual characteristic values ​​of volatile organic compounds (VOCs), and a chemical gradient vector is reconstructed based on the current peak points of the fire's characteristics. The target robot moves in the opposite direction of the gradient, retrogradely tracing the diffusion path of incompletely burned materials. During this movement, temperature data is collected in real time and compared with preset thresholds. If the temperature in a certain area is below the threshold, it is identified as a cooling zone, and the robot automatically bypasses this area to continue detection. When the robot reaches a node where the temperature meets the threshold and the VOC characteristic value exceeds the threshold, a high-precision component scan is triggered. By detecting specific decomposition products, it confirms whether this is a residual area of ​​the actual fire source. This process effectively solves the problem of inaccurate fire source location during the decay phase by dynamically adjusting the detection strategy.

[0043] By dynamically switching eigenvalues ​​to ensure the matching of the detected object with the fire stage, using inverse gradient tracing to accurately reconstruct the fire source spread path, employing a temperature screening mechanism to avoid wasting resources in invalid areas, and using component scanning verification to ensure the reliability of the sampling points, this technical solution significantly improves the accuracy of sample collection during the decay stage, providing crucial evidence for fire source tracing.

[0044] Furthermore, the residual characteristic values ​​of volatile organic compounds include the characteristic peak intensities of benzene series compounds, aldehydes, or polycyclic aromatic hydrocarbons; the characteristic values ​​of complete combustion markers include the characteristic intensities of carbon dioxide, water vapor, and aerosols generated by complete combustion.

[0045] In this embodiment, the residual characteristic values ​​of volatile organic compounds include the characteristic peak intensities of benzene series compounds, aldehydes, or polycyclic aromatic hydrocarbons; the characteristic values ​​of complete combustion markers include the characteristic intensities of carbon dioxide, water vapor, and aerosols generated by complete combustion.

[0046] Among them, benzene series compounds refer to organic compounds containing benzene ring structures. Specifically, the intensity of their molecular ion peaks can be detected by gas chromatography-mass spectrometry, and the changes in their intensity can reflect the residual distribution of incompletely burned volatile organic compounds in a fire.

[0047] Aldehydes are organic compounds containing an aldehyde group, such as formaldehyde and acetaldehyde. Their specific absorption peak intensities can be detected by infrared spectroscopy or electrochemical sensors to trace the diffusion path of pyrolysis products.

[0048] Polycyclic aromatic hydrocarbons (PAHs) are hydrocarbons containing two or more benzene rings. Specifically, their signal intensity at characteristic emission wavelengths can be determined by fluorescence spectroscopy or high-performance liquid chromatography. Their presence can characterize residual pollutants after high-temperature combustion.

[0049] Carbon dioxide, as a product of complete combustion, can be detected using a non-dispersive infrared sensor to determine the intensity of its concentration characteristics, which can then be used to identify active combustion zones in a fire.

[0050] The characteristic intensity of water vapor can be measured by a humidity sensor or a dew point sensor, reflecting the amount of moisture released during combustion.

[0051] The characteristic intensity of aerosols can be determined using a laser scattering particle sensor or a beta-ray absorption method to characterize the concentration of solid particles generated by complete combustion.

[0052] After a fire enters the decay phase, the detection of residual volatile organic compound (VOC) characteristic values ​​is activated. The characteristic peak intensities of benzene series compounds, aldehydes, and polycyclic aromatic hydrocarbons are continuously monitored, for example, through online analysis at grid nodes using a target robot. Traditional methods only detect complete combustion products such as carbon dioxide and cannot capture the VOCs remaining in the decay phase. Existing technologies that rely solely on temperature or smoke concentration parameters easily miss crucial chemical evidence such as benzene series compounds. This solution establishes a multi-dimensional characteristic peak intensity system, enabling full-cycle monitoring of the combustion state and resolving the problem of missing fire source tracing data during the decay phase.

[0053] Furthermore, it also includes a decay phase tracing module: used to record the complete reverse tracing path from the peak point of fire characteristics to the potential origin point, and associate it with the finally collected samples to generate a tracing evidence chain for the fire decay process; the specific steps for generating the tracing evidence chain for the fire decay process are as follows: during the mobile detection process, the target robot is controlled to synchronously collect and package the data into a path data set at fixed time intervals; the path data set contains several path data units, each path data unit contains at least one type of specific three-dimensional grid node and the corresponding fire marker feature value; the specific three-dimensional grid node types include fire characteristic peak points, concentration inflection nodes, temperature exclusion nodes, and potential origin points; concentration inflection nodes represent the corresponding three-dimensional grid nodes where the rate of change of volatile organic compound residual feature values ​​exceeds a preset concentration threshold; temperature exclusion nodes represent the corresponding three-dimensional grid nodes where the temperature is lower than a preset temperature threshold; the digital fingerprint of the path data set is calculated, the digital fingerprint is bound with a timestamp and device identifier, a hash algorithm is used to generate a tracing evidence chain, and the tracing evidence chain is stored in the storage medium.

[0054] like Figure 2 The diagram shown is a schematic flowchart of the portable fire smoke particle rapid collection and sealing evaluation device provided in the embodiment of this application.

[0055] In this embodiment, the path data set is used to completely record the key node information on the reverse tracing path.

[0056] Concentration inflection nodes in a specific 3D mesh node type are mesh nodes where the rate of change of residual characteristic values ​​of volatile organic compounds exceeds a preset threshold. Specifically, this can be achieved by calculating the rate of change of characteristic values ​​of adjacent nodes in real time and comparing them with the threshold, which is used to identify regions of abrupt changes in chemical concentration.

[0057] Temperature exclusion nodes refer to grid nodes whose temperature is below a preset threshold. Specifically, they can be implemented using an infrared temperature sensor and a threshold comparator to exclude interference from cooled areas.

[0058] A digital fingerprint is a unique data identifier generated by a hash algorithm. Specifically, it can be implemented by using the SHA-256 algorithm to encrypt the path data set in one direction to ensure data integrity and immutability.

[0059] Tamper-proof storage media refers to storage devices with write protection functions. Specifically, this can be achieved using blockchain distributed storage nodes with physical write protection switches to prevent the evidence chain from being maliciously modified.

[0060] Specifically, during the source tracing process in the fire decay stage, the target robot automatically collects the coordinates of the three-dimensional grid nodes at its current location and the corresponding fire marker feature values ​​at fixed intervals to form path data units.

[0061] This system achieves full-process correlation between sampling paths and sample data during the fire decay phase. The identification of concentration inflection points accurately reflects the abrupt changes in chemical diffusion paths, while the marking of temperature exclusion points effectively filters out invalid detection areas. The combination of digital fingerprinting and tamper-proof storage ensures the legal validity of the evidence chain. This allows for the traceability of spatial context information during sample collection during laboratory analysis, significantly improving the accuracy of fire cause analysis and the admissibility of evidence.

[0062] Furthermore, based on the fire marker feature values ​​of the 3D grid node and its adjacent 3D grid nodes, the specific process of obtaining the fire chemical gradient vector through spatial vector operation is as follows: A local 3D coordinate system is established with the 3D grid node as the origin; the spatial coordinates and corresponding fire marker feature values ​​of the 3D grid node and its adjacent 3D grid nodes are obtained; the absolute value of the difference between the fire marker feature values ​​of the 3D grid node and any adjacent 3D grid node is calculated using the 3D grid node as the vector starting point and used as the corresponding vector magnitude; when the difference between the fire marker feature values ​​of the 3D grid node and any adjacent 3D grid node is positive, the direction from the adjacent 3D grid node to the 3D grid node is used as the corresponding vector direction; when the difference between the fire marker feature values ​​of the 3D grid node and any adjacent 3D grid node is negative, the direction from the 3D grid node to the adjacent 3D grid node is used as the corresponding vector direction; several feature gradient vectors are generated based on the vector starting point, the corresponding vector magnitudes, and the corresponding vector directions; all feature gradient vectors are synthesized into a 3D vector to obtain the fire chemical gradient vector.

[0063] In this embodiment, the local three-dimensional coordinate system refers to a coordinate system established with the current three-dimensional mesh node as the center, which is used to determine the spatial positional relationship between adjacent nodes. Specifically, it can be implemented using a Cartesian coordinate system or a polar coordinate system, providing a geometric reference for subsequent vector operations.

[0064] The absolute value of the difference in characteristic values ​​of fire markers refers to the quantified value of the difference in chemical concentration between the current node and its neighboring nodes. This can be obtained through direct measurement by sensors or by difference calculation, and is used to characterize the strength of the chemical gradient. The vector direction refers to the spatial orientation of the chemical concentration change trend, which can be determined by the sign relationship between the coordinate difference and concentration difference between neighboring nodes, and is used to indicate the path direction of robot movement.

[0065] Three-dimensional vector synthesis refers to the superposition of feature gradient vectors in multiple directions, which can be achieved using a vector addition algorithm to comprehensively reflect the overall trend of chemical concentration in three-dimensional space.

[0066] Ensure the robot moves along the true concentration gradient direction. A 3D vector synthesis mechanism eliminates the influence of local interference signals, avoiding invalid sampling in non-critical areas. The path planning method based on spatial vector operations significantly improves the localization efficiency of fire characteristic peak points, providing accurate navigation for subsequent sample collection and effectively solving the problem of insufficient accuracy in identifying the chemical gradient direction in a fire environment.

[0067] Furthermore, all feature gradient vectors are synthesized into a three-dimensional vector. The specific synthesis process is as follows: Let the three-dimensional grid node be P0, and the fire marker feature value of the three-dimensional grid node be V0; number the adjacent three-dimensional grid nodes of the three-dimensional grid node, i = 1, 2, ..., n, where i represents the number of the adjacent three-dimensional grid node and n represents the total number of adjacent three-dimensional grid nodes; then the formula for calculating the three-dimensional vector synthesis of the fire chemical gradient vector G is: in, This indicates a pointer from P0 to P. i The unit vector, V i This represents the fire marker characteristic value of the adjacent three-dimensional grid nodes of the given three-dimensional grid node.

[0068] In this embodiment, the distribution of chemical substance concentration in a fire environment exhibits spatial non-uniformity, and the differences in eigenvalues ​​between adjacent nodes directly affect the accuracy of gradient direction calculation. By establishing a mathematical formula to convert the concentration changes in each adjacent direction into vector components and superimposing them, the final synthesized fire chemical gradient vector can accurately reflect the maximum trend of chemical concentration change around the current node.

[0069] High-precision calculation of fire scene chemical gradient vectors is achieved through mathematical modeling, ensuring that the target robot moves along the path with the fastest chemical concentration change and avoiding path misjudgment caused by local concentration fluctuations. This calculation method significantly improves the search efficiency of fire feature peak points, enabling the robot to quickly converge to the core area with the most sampling value in complex fire scene environments, providing a reliable spatial positioning basis for subsequent sample collection.

[0070] Furthermore, it also includes a dynamic grid optimization module: used to dynamically adjust the density distribution of three-dimensional grid nodes based on the spatial distribution characteristics of fire marker feature values ​​collected in real time; the dynamic adjustment of the density distribution of three-dimensional grid nodes specifically includes: delineating a first region with the three-dimensional grid node as the origin and a preset first distance as the radius; when it is detected that the fire marker feature values ​​in the first region are all lower than the activation threshold, reducing the grid node density in the first region according to a predefined first rule; when it is detected that the absolute average difference of the fire marker feature values ​​of the adjacent three-dimensional grid nodes of the three-dimensional grid node exceeds a preset difference gradient threshold, delineating a second region with the three-dimensional grid node as the origin and a preset second distance as the radius, and increasing the grid node density in the second region according to a predefined second rule.

[0071] In this embodiment, Figure 4 This is a schematic diagram of the logic flow for dynamic mesh optimization provided in an embodiment of this application.

[0072] Dynamically adjusting the density distribution of three-dimensional grid nodes refers to adaptively optimizing the spatial density of grid nodes based on the real-time distribution changes of fire marker characteristic values. Specifically, this can be achieved using statistical analysis algorithms based on regional characteristic values. By reducing the number of redundant nodes in low characteristic value areas and increasing the node density in high gradient change areas, data acquisition efficiency and spatial resolution can be balanced.

[0073] The preset first distance refers to the radius parameter used to delineate areas with low eigenvalues. It can be dynamically set according to the size of the fire scene and the robot's mobility. For example, it can be set to 50% to 80% of the robot's maximum single movement distance to define the area where the grid density needs to be reduced.

[0074] The predefined first rule refers to the operational strategy for reducing the density of grid nodes. Specifically, it can adopt a proportional sparsification method, such as increasing the distance between adjacent nodes to 1.5-2 times the original distance in low eigenvalue regions, while maintaining the continuity of boundary nodes to avoid data gaps.

[0075] The preset difference gradient threshold is a conditional parameter that triggers an increase in grid density. Specifically, it can be a critical value of the rate of change of feature values ​​obtained by training with historical fire data. For example, it can be set to 1.2-1.5 times the absolute average difference of feature values ​​between adjacent nodes, and is used to identify areas where sampling accuracy needs to be improved.

[0076] The predefined second rule refers to the operational strategy for increasing the density of grid nodes. Specifically, it can be achieved by using interpolation densification methods, such as inserting new nodes in regions with high gradient changes and performing linear interpolation calculations based on the feature values ​​of adjacent nodes to form a higher resolution spatial grid.

[0077] Traditional fire sampling devices typically use a fixed-density grid division method, which cannot adjust the detection accuracy according to the dynamic changes in the fire situation, resulting in wasted resources in low-risk areas and insufficient sampling in high-gradient areas.

[0078] This application achieves real-time adaptive optimization of the fire detection grid structure, effectively balancing sampling accuracy and resource efficiency. Reducing redundant nodes in low eigenvalue regions shortens the robot's movement path, for example, reducing detection time by approximately 30%; increasing node density in high gradient change regions improves the accuracy of fire spread path identification, for example, reducing the location error of characteristic peak points to within 0.2 meters. This technique provides dynamically optimized spatial data support for the accurate collection of fire smoke particles.

[0079] Furthermore, it also includes a multi-machine collaborative verification module: when a single target robot determines the peak point of the fire feature, it dispatches at least one auxiliary robot to the peak point of the fire feature for independent verification; it compares the characteristic value data of the fire markers collected by the main and auxiliary robots and calculates the data consistency index; if the consistency index is lower than the preset confidence threshold, it initiates the re-detection process.

[0080] In this embodiment, the multi-machine collaborative verification module refers to a system component that uses multiple robots to cross-check data of the same target area. Specifically, it can be implemented using a distributed task scheduling algorithm and a data fusion algorithm. Its function is to eliminate the risk of misjudgment caused by measurement errors of a single device or environmental interference.

[0081] An auxiliary robot is a mobile detection device equipped with the same sensor configuration. Specifically, it can be implemented using a device of the same model as the main robot. It is used to independently collect data in the same spatial location to verify the repeatability of the results.

[0082] Data consistency index is a quantitative parameter that measures the degree of difference in measurement results from different devices. It can be calculated using Euclidean distance or correlation coefficient and is used to determine whether the data collected by the main and auxiliary robots are within an acceptable deviation range.

[0083] By employing a multi-machine independent measurement and cross-validation mechanism, abnormal data can be effectively identified and eliminated, avoiding the impact of single device failure or environmental noise on sampling results. The multi-robot collaborative verification mechanism ensures the reliability of data at key sampling points, providing high-quality basic data support for subsequent source tracing analysis, while reducing the cost of repeated sampling caused by misjudgment.

[0084] This application also provides a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the steps of a portable fire smoke particle rapid collection and sealing assessment device.

[0085] After the computer program is loaded into the processor's memory, it controls the target robot to dynamically collect the characteristic values ​​of fire markers and plan its path by calling the three-dimensional mesh partitioning algorithm, gradient vector calculation function and multi-machine collaborative communication interface. In the multi-machine collaborative verification scenario, the processor analyzes the independent verification data uploaded by the auxiliary robot and calls the consistency comparison algorithm to judge the credibility of the peak points of fire features. If the data difference exceeds the preset range, the re-detection process is triggered.

[0086] To ensure the compatibility and portability of computer programs on different hardware devices, and to achieve multi-dimensional verification of fire characteristic peak points through programmatic control, thereby improving the global optimality and data reliability of smoke particle sample collection.

[0087] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0089] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0090] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0091] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0092] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A portable device for rapid collection, containment, and assessment of fire smoke particles, characterized in that, include: Acquisition module: used to divide the target fire space into several three-dimensional grid nodes and obtain the feature values ​​of fire markers collected by the target robot at the three-dimensional grid nodes; The first judgment and positioning module is used to determine whether the characteristic value of the fire marker is greater than the preset activation threshold. If it is, the following operations are performed: Based on the characteristic value of the fire marker of the three-dimensional grid node and the adjacent three-dimensional grid nodes, the fire chemical gradient vector is obtained through spatial vector operation. Starting from the three-dimensional grid node, the target robot is controlled to move along the direction of the fire chemical gradient vector to the new three-dimensional grid node. The second judgment and positioning module is used to repeatedly execute the operation in the first judgment and positioning module until the following convergence condition is met: the fire marker feature value of the new three-dimensional grid node is greater than the fire marker feature values ​​of all its adjacent grid nodes and is recorded as the fire feature peak point. The third judgment and positioning module is used to record the outer adjacent grid nodes of all adjacent grid nodes of the fire feature peak point as secondary nodes, obtain and judge whether the fire feature value of the secondary node is greater than the fire feature value of the fire feature peak point. If so, the corresponding secondary node is recorded as the new fire feature peak point. Collection and Sealing Execution Module: Used to control the target robot to perform smoke particle sample collection and sealing operations at the peak point of fire characteristics.

2. The portable fire smoke particle rapid collection and containment assessment device according to claim 1, characterized in that, It also includes a fire stage intelligent identification module: The characteristic values ​​of the fire markers at the peak point of the fire are used to obtain the characteristic values ​​of the completely burned markers and the characteristic values ​​of the incompletely burned markers, and to calculate the ratio between the two. If the ratio exceeds the preset decay stage threshold, the fire is determined to be in the decay stage, and a stage switching instruction is generated. The target robot responds to the phase switching command and performs a potential origin determination operation.

3. The portable fire smoke particle rapid collection and containment assessment device according to claim 2, characterized in that, The process of performing the potential origin determination operation specifically includes: The core fire marker characteristic values ​​to be collected were switched from complete combustion marker characteristic values ​​to volatile organic compound residue characteristic values; Using the current peak point of the fire characteristics as a new starting point, recalculate the fire chemical gradient vector based on the residual characteristic value of volatile organic compounds, and control the target robot to move and detect in the opposite direction of the fire chemical gradient vector. During the mobile detection process, the temperature along the mobile path is acquired in real time. If the temperature of a certain grid node is lower than the preset temperature threshold, the area defined by the grid node according to the predefined radius is determined to be a cooling zone where the heat source has been extinguished. Sampling in this area is abandoned, and the mobile detection continues until the temperature of a certain grid node is higher than the preset temperature threshold and the residual characteristic value of volatile organic compounds is higher than the preset characteristic threshold. The grid node is then recorded as a potential origin point. A fine compositional scan is performed on potential origin points. If decomposition markers of specific ignition source materials are detected, the point is identified as the optimal collection point for the decay phase, and the collection and storage execution module is triggered.

4. The portable fire smoke particle rapid collection and containment assessment device according to claim 3, characterized in that, The residual characteristic values ​​of the volatile organic compounds include the characteristic peak intensities of benzene series compounds, aldehydes, or polycyclic aromatic hydrocarbons; the characteristic values ​​of the complete combustion markers include the characteristic intensities of carbon dioxide, water vapor, and aerosols generated by complete combustion.

5. The portable fire smoke particle rapid collection and containment assessment device according to claim 3, characterized in that, It also includes a recession phase tracing module: Used to record the complete reverse tracing path from the peak point of fire characteristics to the potential origin point, and to associate it with the finally collected samples to generate a chain of evidence for tracing the fire decay process; The specific steps for generating the chain of evidence for the fire decay process are as follows: During the mobile exploration process, the target robot is controlled to synchronously collect and package path data sets at fixed time intervals; The path data set contains several path data units, and each path data unit contains at least one type of specific three-dimensional mesh node and the corresponding fire marker feature value; The specific three-dimensional mesh node types include fire feature peak points, concentration inflection nodes, temperature exclusion nodes, and potential origin points; The concentration inflection nodes represent the corresponding three-dimensional mesh nodes where the rate of change of the residual characteristic value of volatile organic compounds exceeds a preset concentration threshold. The temperature exclusion node represents the corresponding three-dimensional mesh node whose temperature is lower than a preset temperature threshold; Calculate the digital fingerprint of the path data set, bind the digital fingerprint with a timestamp and a device identifier, and use a hash algorithm to generate a traceability evidence chain.

6. The portable fire smoke particle rapid collection and containment assessment device according to claim 1, characterized in that, The specific process of obtaining the fire chemical gradient vector through spatial vector operation based on the fire marker feature values ​​of the three-dimensional grid node and its adjacent three-dimensional grid nodes is as follows: Establish a local three-dimensional coordinate system with the three-dimensional mesh node as the origin; Obtain the spatial coordinates of the 3D grid node and its adjacent 3D grid nodes, as well as the corresponding fire marker feature values; Using the three-dimensional grid node as the starting point of the vector, the absolute value of the difference between the characteristic values ​​of the fire markers of the three-dimensional grid node and any of its adjacent three-dimensional grid nodes is calculated and used as the vector magnitude of the two. When the difference between the fire marker feature value of the three-dimensional grid node and any adjacent three-dimensional grid node is positive, the direction from the adjacent three-dimensional grid node to the three-dimensional grid node is taken as the corresponding vector direction between the two. When the difference between the fire marker feature value of the three-dimensional grid node and any adjacent three-dimensional grid node is negative, the direction from the three-dimensional grid node to the adjacent three-dimensional grid node is taken as the vector direction between the two. Several feature gradient vectors are generated based on the vector starting point, the vector magnitudes corresponding to the two vectors, and the vector directions corresponding to the two vectors. All feature gradient vectors are synthesized into a three-dimensional vector to obtain the fire chemical gradient vector.

7. The portable fire smoke particle rapid collection and containment assessment device according to claim 6, characterized in that, The process of synthesizing all feature gradient vectors into three-dimensional vectors is as follows: Let the three-dimensional mesh node be P0, and the fire marker feature value of the three-dimensional mesh node be V0; Number the adjacent 3D mesh nodes of the given 3D mesh node, i = 1, 2, ..., n, where i represents the number of the adjacent 3D mesh node and n represents the total number of adjacent 3D mesh node numbers; The formula for calculating the three-dimensional vector synthesis of the fire chemical gradient vector G is: in, This indicates a pointer from P0 to P. i The unit vector, V i This represents the fire marker characteristic value of the adjacent three-dimensional grid nodes of the given three-dimensional grid node.

8. The portable fire smoke particle rapid collection and containment assessment device according to claim 1, characterized in that, It also includes a dynamic mesh optimization module: Used to dynamically adjust the density distribution of three-dimensional grid nodes based on the spatial distribution characteristics of fire marker feature values ​​acquired in real time; The dynamic adjustment of the density distribution of the three-dimensional mesh nodes specifically includes: Using the three-dimensional grid node as the origin and a preset first distance as the radius, a first region is defined. When the characteristic values ​​of fire markers in the first region are all lower than the activation threshold, the grid node density in the first region is reduced according to a predefined first rule. When the absolute average difference of the fire marker feature values ​​of the adjacent three-dimensional grid nodes of the detected three-dimensional grid node exceeds the preset difference gradient threshold, a second region is delineated with the three-dimensional grid node as the origin and a preset second distance as the radius, and the grid node density is increased in the second region according to the predefined second rule.

9. The portable fire smoke particle rapid collection and containment assessment device according to claim 1, characterized in that, It also includes a multi-machine collaborative verification module: This is used to dispatch at least one auxiliary robot to the fire characteristic peak point for independent verification after a single target robot has determined the peak point of the fire characteristic; Compare the characteristic value data of fire markers collected by the main and auxiliary robots, and calculate the data consistency index; If the consistency index is lower than the preset confidence threshold, the re-probe process will be initiated.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the apparatus as described in any one of claims 1-9.

Images