Smoking fire evidence analysis method
By collecting, preprocessing and screening fire site data, conducting characteristic experiments and cigarette comparisons, building an analysis model, and using particle swarm optimization algorithm and machine learning technology, the data complexity and concealment problems of smoking fire investigation are solved, and efficient and accurate fire cause identification is achieved.
Patent Information
- Application Number
- CN202510704823.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-29
- Publication Date
- 2025-08-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing smoking fire investigation methods lack systematic data support and scientific analysis methods. The complexity of soot burning materials and the concealment of smoking fires make it difficult to identify the cause of fire.
By collecting fire site data, pre-processing and screening to extract soot burning material data, conducting characteristic experiments and cigarette comparisons, building a smoke burning material analysis model for smoking fires, and using particle swarm optimization algorithm and machine learning technology to analyze fire triggers.
It improves the accuracy and efficiency of smoke combustion substance analysis in smoking fires, realizes intelligent analysis of smoking fires, adapts to the analysis needs of different standards, and is universal.
Smart Images
Figure CN120493020A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of cigarette ash combustion analysis, in particular to a smoking fire evidence analysis method. Background Art
[0002] Smoking fires are fires caused by smoking, typically when unextinguished cigarette butts, matches, or other smoking-related items ignite combustible materials. These fires cause significant casualties and property damage worldwide, particularly in places like homes, hotels, and warehouses. The hidden and sudden nature of smoking fires makes their investigation and evidence collection challenging. To accurately identify the cause of fires and implement effective preventive measures, developing scientific and systematic methods for analyzing smoking fire evidence is crucial.
[0003] Traditional fire investigation methods rely primarily on the experience and subjective judgment of on-site investigators, lacking systematic data support and scientific analytical tools. With the advancement of fire science and soot combustion product analysis, a growing number of techniques have been introduced into the field of fire investigation, such as chemical composition analysis of combustion residues, image reconstruction of the fire scene, and comprehensive analysis of meteorological and building structural data. However, soot combustion product analysis for smoking fires still faces the following challenges: the complexity of soot combustion product data, the hidden nature of smoking fires, and the limitations of data analysis.
[0004] To address these issues, researchers have begun exploring data-driven and machine learning-based methods for analyzing fire ash combustion products in recent years. For example, by analyzing the chemical composition and microstructure of combustion residues, the characteristics of tobacco combustion can be identified; image data and building structure data can be used to reconstruct the fire scene to assist in determining the location of the fire source; and meteorological data and eyewitness data can be combined to verify the environmental conditions in which the fire occurred. These methods provide new ideas for analyzing smoke ash combustion products from smoking fires, but they still need to be further optimized and systematized. Summary of the Invention
[0005] The purpose of the present invention is to provide a method for analyzing smoking fire evidence.
[0006] To achieve the above object, the present invention is implemented according to the following technical solutions: The present invention comprises the following steps: Collecting fire scene investigation data and related data, and preprocessing the investigation data and related data; Screening and extracting the survey data according to the correlation to obtain soot combustion product data, and performing characteristic experiments on the soot combustion product data to obtain experimental data; Performing cigarette comparison on the experimental data to obtain comparison data, and combining the relevant data and the comparison data to obtain a cigarette ash combustion product analysis result; A smoking fire soot combustion product analysis model is constructed according to the soot combustion product analysis result, the data to be analyzed is input into the smoking fire soot combustion product analysis model, and the soot combustion product analysis result is output.
[0007] Furthermore, the method of screening and extracting the exploration data to obtain ash combustion data according to the correlation degree includes: Classify the survey data within the fire source range to obtain the classification and calculate the weight of the survey data: The number of x-th classification exploration data is , the ath exploration data is , the average value of the x-th classification exploration data is , exploration data The weight is , the number of categories is , exploration data The marginal probability is ; Calculate the importance of exploration data: The exploration data The importance of , the optimization coefficient is ; The exploration data with importance greater than 0.349 are regarded as important data, the exploration features of important data are extracted, and the correlation of important data is calculated: The important data The fire correlation degree is , the rth exploration feature is , exploration characteristics The proportion of fires occurring is , important data The number of exploration features in , the control factor is , the environmental combustion coefficient is L, important data The autoignition temperature is , the vth important data is , the step function is , important data The autoignition temperature is ; The important data with a correlation degree greater than 0.402 is output as soot combustion product data.
[0008] Furthermore, a method for performing a characteristic experiment on the soot combustion product data to obtain experimental data includes: A characteristic experiment is performed according to the soot combustion product data and the set substances, wherein the set substances include single soot combustion product data and combined soot combustion product data with different proportions; the characteristic experiment includes a physical characteristic experiment and a chemical characteristic experiment; and the characteristic experiment results are output as experimental data.
[0009] Furthermore, the method of performing cigarette comparison on the experimental data to obtain comparison data includes: Collect cigarette standard data under experimental conditions, input the experimental data and cigarette standard data into the comparative analysis model, and calculate the similarity between the cigarette standard data and the experimental data: The standard data of the cth cigarette is , the jth experimental data is , cigarette standard data and experimental data The similarity is , the kernel width is , the regulatory factor is , cigarette standard data and experimental data The Euclidean distance is , the number of cigarette standard data is , the number of experimental data is ; When the similarity is greater than 0.736, the cigarette standard data is used as the control group of the experimental data, and the cigarettes in the control group are used as the first comparison result; When the similarity is greater than 0.327 and less than 0.736, the experimental data is used as chaotic data, and the cigarette standard data is randomly combined to obtain the combined data. The matching degree between the chaotic data and the combined data is calculated: The cth chaotic data is , the jth chaotic data is , chaotic data and chaotic data The matching degree is , chaotic data and chaotic data The Euclidean distance is , the weight coefficient of Euclidean distance , the weight coefficient of similarity is , the optimization parameters are ; The combination data with the highest matching degree is used as the control group of chaotic data, and the combination cigarettes of the control group is used as the second comparison result; When the similarity is less than 0.327, the experimental data is irrelevant to cigarettes and the third comparison result is obtained; The first comparison result, the second comparison result, and the third comparison result are output as comparison data.
[0010] Furthermore, the method for obtaining smoke ash combustion product analysis results by combining the relevant data and the comparison data includes: Conduct fire simulation based on relevant data to obtain the probability of fire spread, and calculate the probability of fire initiation based on the comparison data: The environmental combustion coefficient is L, and the ath comparison data is , the vth comparison data is , comparison data The autoignition temperature is , comparison data The autoignition temperature is , comparison data The probability of fire starting is , comparison data The probability of spreading is , the correction factor for the tth iteration is ; Particle populations are introduced, and the variance between the predicted fire probability and the actual fire probability is used as the fitness function to calculate the particle fitness, and the particle position with the minimum fitness is taken as the optimal position. Use the minimum variance as the search strategy, search the control coefficient according to the search strategy, and calculate the initial position of the particle: The initial position of the sth particle in the kth dimension at the tth iteration is , the best position is , the step factor is , the maximum dimension is , the minimum dimension is , the number of dimensions is , the walk parameter of the kth dimension is ; Update the particle's position to get the updated position. The expression is: The updated position of the sth particle in the kth dimension in the t+1th iteration is , the optimal position of the tth iteration is , the random numbers from 0 to 1 are , the maximum number of iterations is , the current number of iterations is t, and the current dimension is k; Use the inertia weight to update the particle position to obtain the inertial position. The expression is: The inertial position of the sth particle in the kth dimension at the t+1th iteration is , the position of the particle in the k dimension at the tth iteration is , the correction factor for the tth iteration is ; Update the inertia weight, the expression is: The correction factor for the t+1th iteration is , a random number from 0 to 1 is ; Calculate the fitness value of the particle after the updated position. If it is less than the target position, replace the target position and continue iterating until the fitness value reaches the minimum. The fire probability is adjusted according to the correction factor of the minimum fitness value, and the comparison data with the highest fire probability is taken as the fire cause. When the fire cause is cigarettes, the ash combustion analysis result is cigarettes; when the fire cause is not cigarettes, the ash combustion analysis result is other.
[0011] Furthermore, a method for constructing a smoking fire soot combustion product analysis model based on the soot combustion product analysis results includes: The analysis model of smoke ash combustion products includes classification algorithm, feature extraction algorithm, recurrent neural network algorithm, and Bayesian optimization algorithm; The classification algorithm divides the data into cigarette fires and non-cigarette fires based on the results of the soot combustion product analysis; The feature extraction algorithm inputs cigarette fires into the autoencoder, uses the neural network to learn the low-dimensional representation of the data to obtain fire features, and outputs fire features with a cigarette fire frequency higher than 0.782 as dynamic cigarette fire features; The recurrent neural network algorithm captures the temporal dependencies in time series data, extracts dynamic cigarette fire features related to smoking fires, determines whether the fire was caused by smoking behavior based on the dynamic cigarette fire features, and obtains the analysis results of ash combustion products; The Bayesian optimization algorithm probabilistically models the hyperparameter space of the recurrent neural network by constructing a proxy model, and selects hyperparameters based on the acquisition function for iterative optimization to obtain the optimal hyperparameter configuration.
[0012] In a second aspect, an embodiment of the present application further provides an electronic device, including: A processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the method steps described in the first aspect.
[0013] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.
[0014] The beneficial effects of the present invention are: The present invention is a method for analyzing smoking fire evidence. Compared with the prior art, the present invention has the following technical effects: The present invention can improve the accuracy of smoking fire soot combustion product analysis through pretreatment, screening and extraction, characteristic experiments, cigarette comparison, acquisition of soot combustion product analysis results and model construction steps, thereby improving the precision of smoking fire soot combustion product analysis, optimizing smoking fire soot combustion product analysis, greatly saving resources, and improving work efficiency. It can realize intelligent analysis of smoking fire soot combustion products, and perform screening, extraction and data fusion of smoking fire soot combustion product analysis in real time. It is of great significance to smoking fire soot combustion product analysis, can adapt to smoking fire soot combustion product analysis with different standards and different smoking fire soot combustion product analysis needs, and has a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 A flowchart of the steps of the smoking fire evidence analysis method of the present invention; Figure 2 This is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION
[0016] The present invention will be further described below through specific examples. The illustrative examples and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.
[0017] The smoking fire evidence analysis method of the present invention comprises the following steps: like Figure 1 As shown, in this embodiment, the following steps are included: Collecting fire scene investigation data and related data, and preprocessing the investigation data and related data; In actual assessments, the survey data includes fire location, fire occurrence time, fire starting point information, possible ignition sources, combustible materials, burn marks, fire spread path, electrical equipment and wiring, burn marks, smoke direction, ash, burn residue, ventilation conditions, oxygen concentration, and smoke composition; the relevant data includes meteorological data, building structure data, image data, eyewitness testimony data, and surrounding environment; The study focused on a residential fire in XXX. The fire occurred in the living room, with visible signs of burns on the sofa and carpet. Smoke spread from the living room to the bedroom. Large amounts of ashes were found on the sofa and carpet, along with unburned cigarette butts and match debris. The fire occurred in dry weather with low wind speeds. There was no fire door between the living room and bedroom. Surveillance footage showed someone smoking in the living room before the fire, and the resident admitted to smoking in the living room. Screening and extracting the survey data according to the correlation to obtain soot combustion product data, and performing characteristic experiments on the soot combustion product data to obtain experimental data; In the actual evaluation, cardboard fragments were selected as the ash combustion data. The experimental data included chemical characteristic test results and physical characteristic test results. The chemical characteristic results: nicotine and tar components were detected based on the cardboard fragments. The physical characteristic results: the combustion temperature was determined to be 300°C based on the combustion residue. Performing cigarette comparison on the experimental data to obtain comparison data, and combining the relevant data and the comparison data to obtain a cigarette ash combustion product analysis result; In the actual evaluation, cigarette standard data was collected and the similarity between the experimental data and the cigarette standard data was calculated. The similarity was 0.85, which was determined as the first comparison result; A fire simulation was conducted based on relevant data, and the fire ignition probability was calculated to be 0.92. The fire ignition probability was adjusted using the particle swarm optimization algorithm, and it was ultimately determined that the fire was caused by smoking. A smoking fire soot combustion product analysis model is constructed according to the soot combustion product analysis result, the data to be analyzed is input into the smoking fire soot combustion product analysis model, and the soot combustion product analysis result is output.
[0018] In this embodiment, the method of screening and extracting the survey data according to the correlation to obtain the soot combustion data includes: Classify the survey data within the fire source range to obtain the classification and calculate the weight of the survey data: The number of x-th classification exploration data is , the ath exploration data is , the average value of the x-th classification exploration data is , exploration data The weight is , the number of categories is , exploration data The marginal probability is ; Calculate the importance of exploration data: The exploration data The importance of , the optimization coefficient is ; The exploration data with importance greater than 0.349 are regarded as important data, the exploration features of important data are extracted, and the correlation of important data is calculated: The important data The fire correlation degree is , the rth exploration feature is , exploration characteristics The proportion of fires occurring is , important data The number of exploration features in , the control factor is , the environmental combustion coefficient is L, important data The autoignition temperature is , the vth important data is , the step function is , important data The autoignition temperature is ; The important data with a correlation degree greater than 0.402 is output as soot combustion product data.
[0019] In this embodiment, the method of performing a characteristic experiment on the soot combustion product data to obtain experimental data includes: A characteristic experiment is performed according to the soot combustion product data and the set substances, wherein the set substances include single soot combustion product data and combined soot combustion product data with different proportions; the characteristic experiment includes a physical characteristic experiment and a chemical characteristic experiment; and the characteristic experiment results are output as experimental data.
[0020] In this embodiment, the method of performing cigarette comparison on the experimental data to obtain comparison data includes: Collect cigarette standard data under experimental conditions, input the experimental data and cigarette standard data into the comparative analysis model, and calculate the similarity between the cigarette standard data and the experimental data: The standard data of the cth cigarette is , the jth experimental data is , cigarette standard data and experimental data The similarity is , the kernel width is , the regulatory factor is , cigarette standard data and experimental data The Euclidean distance is , the number of cigarette standard data is , the number of experimental data is ; When the similarity is greater than 0.736, the cigarette standard data is used as the control group of the experimental data, and the cigarettes in the control group are used as the first comparison result; When the similarity is greater than 0.327 and less than 0.736, the experimental data is used as chaotic data, and the cigarette standard data is randomly combined to obtain combined data. The matching degree between the chaotic data and the combined data is calculated: The cth chaotic data is , the jth chaotic data is , chaotic data and chaotic data The matching degree is , chaotic data and chaotic data The Euclidean distance is , the weight coefficient of Euclidean distance , the weight coefficient of similarity is , the optimization parameters are ; The combination data with the highest matching degree is used as the control group of chaotic data, and the combination cigarettes of the control group is used as the second comparison result; When the similarity is less than 0.327, the experimental data is irrelevant to cigarettes and the third comparison result is obtained; The first comparison result, the second comparison result, and the third comparison result are output as comparison data.
[0021] In this embodiment, the method of combining the relevant data and the comparison data to obtain the soot combustion product analysis results includes: Conduct fire simulation based on relevant data to obtain the probability of fire spread, and calculate the probability of fire initiation based on the comparison data: The environmental combustion coefficient is L, and the ath comparison data is , the vth comparison data is , comparison data The autoignition temperature is , comparison data The autoignition temperature is , comparison data The probability of fire starting is , comparison data The probability of spreading is , the correction factor for the tth iteration is ; Particle populations are introduced, and the variance between the predicted fire probability and the actual fire probability is used as the fitness function to calculate the particle fitness, and the particle position with the minimum fitness is taken as the optimal position. Use the minimum variance as the search strategy, search the control coefficient according to the search strategy, and calculate the initial position of the particle: The initial position of the sth particle in the kth dimension at the tth iteration is , the best position is , the step factor is , the maximum dimension is , the minimum dimension is , the number of dimensions is , the walk parameter of the kth dimension is ; Update the particle's position to get the updated position. The expression is: The updated position of the sth particle in the kth dimension in the t+1th iteration is , the optimal position of the tth iteration is , the random numbers from 0 to 1 are , the maximum number of iterations is , the current number of iterations is t, and the current dimension is k; Use the inertia weight to update the particle position to obtain the inertial position. The expression is: The inertial position of the sth particle in the kth dimension at the t+1th iteration is , the position of the particle in the k dimension at the tth iteration is , the correction factor for the tth iteration is ; Update the inertia weight, the expression is: The correction factor for the t+1th iteration is , a random number from 0 to 1 is ; Calculate the fitness value of the particle after the updated position. If it is less than the target position, replace the target position and continue iterating until the fitness value reaches the minimum. The fire probability is adjusted according to the correction factor of the minimum fitness value, and the comparison data with the highest fire probability is taken as the fire cause. When the fire cause is cigarettes, the ash combustion analysis result is cigarettes; when the fire cause is not cigarettes, the ash combustion analysis result is other.
[0022] In this embodiment, the method for constructing a smoking fire soot combustion product analysis model based on the soot combustion product analysis results includes: The analysis model of smoke ash combustion products includes classification algorithm, feature extraction algorithm, recurrent neural network algorithm, and Bayesian optimization algorithm; The classification algorithm divides the data into cigarette fires and non-cigarette fires based on the results of the soot combustion product analysis; The feature extraction algorithm inputs cigarette fires into the autoencoder, uses the neural network to learn the low-dimensional representation of the data to obtain fire features, and outputs fire features with a cigarette fire frequency higher than 0.782 as dynamic cigarette fire features; The recurrent neural network algorithm captures the temporal dependencies in time series data, extracts dynamic cigarette fire features related to smoking fires, determines whether the fire was caused by smoking behavior based on the dynamic cigarette fire features, and obtains the analysis results of ash combustion products; The Bayesian optimization algorithm constructs a proxy model to probabilistically model the hyperparameter space of the recurrent neural network, and selects hyperparameters based on the acquisition function for iterative optimization to obtain the optimal hyperparameter configuration; In the actual evaluation, the hyperparameter chosen is the learning rate.
[0023] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2 At the hardware level, the electronic device includes a processor and, optionally, an internal bus, a network interface, and memory. The memory may include internal memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for its services.
[0024] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.
[0025] The memory is used to store programs. Specifically, the program may include program code, which includes computer operating instructions. The memory may include internal memory and non-volatile memory, and provides instructions and data to the processor.
[0026] The processor reads the corresponding computer program from the non-volatile memory into the internal memory and then runs it, forming a smoking fire ash and combustion product analysis device at the logical level. The processor executes the program stored in the memory and is specifically configured to perform any of the aforementioned smoking fire evidence analysis methods.
[0027] The above application Figure 1 The smoking and fire evidence analysis method disclosed in the illustrated embodiments can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the method described above can be performed by hardware integrated logic circuits within the processor or by software instructions. The processor described above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of this application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly executed by a hardware decoding processor or by a combination of hardware and software modules within the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.
[0028] The electronic device may also perform Figure 1 Method for analyzing smoking fire evidence and implement Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.
[0029] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple application programs, they execute any of the aforementioned smoking and fire evidence analysis methods.
[0030] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0031] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0032] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0033] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0034] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.
[0035] Memory may include non-permanent storage in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
[0036] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology to store information. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.
[0037] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.
[0038] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0039] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. A method for analyzing smoking fire evidence, characterized in that: The following steps are involved: Collecting fire scene investigation data and related data, and preprocessing the investigation data and related data; Screening and extracting the survey data according to the correlation to obtain soot combustion product data, and performing characteristic experiments on the soot combustion product data to obtain experimental data; Performing cigarette comparison on the experimental data to obtain comparison data, and combining the relevant data and the comparison data to obtain a cigarette ash combustion product analysis result; A smoking fire soot combustion product analysis model is constructed according to the soot combustion product analysis result, the data to be analyzed is input into the smoking fire soot combustion product analysis model, and the soot combustion product analysis result is output.
2. The smoking fire evidence analysis method according to claim 1, characterized in that: The method for screening and extracting the survey data according to the correlation to obtain smoke ash combustion data includes: Classify the survey data within the fire source range to obtain the classification and calculate the weight of the survey data: The number of x-th classification exploration data is , the ath exploration data is , the average value of the x-th classification exploration data is , exploration data The weight of , the number of categories is , exploration data The marginal probability is ; Calculate the importance of exploration data: The exploration data The importance of , the optimization coefficient is ; The exploration data with importance greater than 0.349 are regarded as important data, the exploration features of important data are extracted, and the correlation of important data is calculated: The important data The fire correlation degree is , the rth exploration feature is , exploration characteristics The proportion of fires occurring is , important data The number of exploration features in , the control factor is , the environmental combustion coefficient is L, important data The autoignition temperature is , the vth important data is , the step function is , important data The autoignition temperature is ; The important data with a correlation degree greater than 0.402 is output as soot combustion product data.
3. The smoking fire evidence analysis method according to claim 1, characterized in that: The method of performing a characteristic experiment on the soot combustion product data to obtain experimental data includes: A characteristic experiment is performed according to the soot combustion product data and the set substances, wherein the set substances include single soot combustion product data and combined soot combustion product data with different proportions; the characteristic experiment includes a physical characteristic experiment and a chemical characteristic experiment; and the characteristic experiment results are output as experimental data.
4. The smoking fire evidence analysis method according to claim 1, characterized in that: The method for performing cigarette comparison on the experimental data to obtain comparison data comprises: Collect cigarette standard data under experimental conditions, input the experimental data and cigarette standard data into the comparative analysis model, and calculate the similarity between the cigarette standard data and the experimental data: The standard data of the cth cigarette is , the jth experimental data is , cigarette standard data and experimental data The similarity is , the kernel width is , the regulatory factor is , cigarette standard data and experimental data The Euclidean distance is , the number of cigarette standard data is , the number of experimental data is ; When the similarity is greater than 0.736, the cigarette standard data is used as the control group of the experimental data, and the cigarettes in the control group are used as the first comparison result; When the similarity is greater than 0.327 and less than 0.736, the experimental data is used as chaotic data, and the cigarette standard data is randomly combined to obtain combined data. The matching degree between the chaotic data and the combined data is calculated: The cth chaotic data is , the jth chaotic data is , chaotic data and chaotic data The matching degree is , chaotic data and chaotic data The Euclidean distance is , the weight coefficient of Euclidean distance , the weight coefficient of similarity is , the optimization parameters are ; The combination data with the highest matching degree is used as the control group of chaotic data, and the combination cigarettes of the control group is used as the second comparison result; When the similarity is less than 0.327, the experimental data is irrelevant to cigarettes and the third comparison result is obtained; The first comparison result, the second comparison result, and the third comparison result are output as comparison data.
5. The smoking fire evidence analysis method according to claim 1, characterized in that: The method for obtaining smoke ash combustion product analysis results by combining the relevant data and the comparison data comprises: Conduct fire simulation based on relevant data to obtain the probability of fire spread, and calculate the probability of fire initiation based on the comparison data: The environmental combustion coefficient is L, and the ath comparison data is , the vth comparison data is , comparison data The autoignition temperature is , comparison data The autoignition temperature is , comparison data The probability of fire starting is , comparison data The probability of spreading is , the correction factor for the tth iteration is ; Particle populations are introduced, and the variance between the predicted fire probability and the actual fire probability is used as the fitness function to calculate the particle fitness, and the particle position with the minimum fitness is taken as the optimal position. Use the minimum variance as the search strategy, search the control coefficient according to the search strategy, and calculate the initial position of the particle: The initial position of the sth particle in the kth dimension at the tth iteration is , the best position is , the step factor is , the maximum dimension is , the minimum dimension is , the number of dimensions is , the walk parameter of the kth dimension is ; Update the particle's position to get the updated position. The expression is: The updated position of the sth particle in the kth dimension in the t+1th iteration is , the optimal position of the tth iteration is , the random numbers from 0 to 1 are , the maximum number of iterations is , the current number of iterations is t, and the current dimension is k; Use the inertia weight to update the particle position to obtain the inertial position. The expression is: The inertial position of the sth particle in the kth dimension at the t+1th iteration is , the position of the particle in the k dimension at the tth iteration is , the correction factor for the tth iteration is ; Update the inertia weight, the expression is: The correction factor for the t+1th iteration is , a random number from 0 to 1 is ; Calculate the fitness value of the particle after the updated position. If it is less than the target position, replace the target position and continue iterating until the fitness value reaches the minimum. The fire probability is adjusted according to the correction factor of the minimum fitness value, and the comparison data with the highest fire probability is taken as the fire cause. When the fire cause is cigarettes, the ash combustion analysis result is cigarettes; when the fire cause is not cigarettes, the ash combustion analysis result is other.
6. The smoking fire evidence analysis method according to claim 1, characterized in that: The method for constructing a smoking fire soot combustion product analysis model based on the soot combustion product analysis results includes: The analysis model of smoke ash combustion products includes classification algorithm, feature extraction algorithm, recurrent neural network algorithm, and Bayesian optimization algorithm; The classification algorithm divides the data into cigarette fires and non-cigarette fires based on the results of the soot combustion product analysis; The feature extraction algorithm inputs cigarette fires into the autoencoder, uses the neural network to learn the low-dimensional representation of the data to obtain fire features, and outputs fire features with a cigarette fire frequency higher than 0.782 as dynamic cigarette fire features; The recurrent neural network algorithm captures the temporal dependencies in time series data, extracts dynamic cigarette fire features related to smoking fires, determines whether the fire was caused by smoking behavior based on the dynamic cigarette fire features, and obtains the analysis results of ash combustion products; The Bayesian optimization algorithm probabilistically models the hyperparameter space of the recurrent neural network by constructing a proxy model, and selects hyperparameters based on the acquisition function for iterative optimization to obtain the optimal hyperparameter configuration.
7. An electronic device comprising: processor; as well as A memory arranged to store computer-executable instructions, wherein when the instructions are executed, the processor performs the method according to any one of claims 1 to 6.
8. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute the method according to any one of claims 1 to 6.