Forest fire prevention accurate positioning system based on laser radar aerosol detection
By fusing lidar data with a canopy-modulated plume diffusion model and combining it with a particle swarm optimization algorithm, the shortcomings of traditional fire source location methods in complex environments have been overcome, and high-precision location of early-stage forest fire sources has been achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHENGDU JIUZHOU TIANKE INFORMATION TECHNOLOGY CO LTD
- Filing Date
- 2026-03-02
- Publication Date
- 2026-05-29
AI Technical Summary
Traditional fire source location methods are insufficient in complex terrain and high canopy environments, making it difficult to accurately identify the ignition location, emission intensity, and ignition time. They are particularly unstable and unsuitable in scenarios with multiple smoke plumes or airflow disturbances.
A precise forest fire prevention and positioning system based on lidar aerosol detection was constructed. By integrating forest canopy height models with real-time three-dimensional aerosol distribution data, a canopy-modulated plume diffusion model was established. Particle swarm optimization algorithm was used to identify the fire source location in reverse, and forward diffusion simulation was combined with observation data matching.
It improves the accuracy of plume concentration evolution analysis in complex terrain and high canopy environments, and realizes high-precision automatic identification of fire source location, intensity and ignition time. It is suitable for forest areas with limited visibility and smoke sources that are obscured by tree canopies.
Smart Images
Figure CN122110141A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fire prevention positioning technology, and in particular to a precise forest fire prevention positioning system based on lidar aerosol detection. Background Technology
[0002] In forest fire monitoring and emergency response, rapid and accurate identification of fire source location is crucial for early intervention and resource allocation. However, traditional fire source location methods mostly rely on visible light remote sensing images, infrared thermal imaging, or video surveillance data. Their identification capabilities are often limited by canopy obstruction, smoke blockage, and line-of-sight limitations, making it difficult to effectively capture the weak signals generated by smoke plumes in the early stages of forest fires. At the same time, most existing smoke plume diffusion models are based on simplified wind field and regular terrain assumptions, failing to fully consider the nonlinear modulation effect of forest canopy on wind field disturbance and material diffusion, resulting in significant errors in simulation results in complex terrain and high-canopy environments.
[0003] Furthermore, existing fire source inversion methods typically employ empirical back-inference or image fitting, lacking a reverse simulation mechanism coupled with atmospheric diffusion physics processes. This makes it difficult to jointly identify the ignition location, emission intensity, and ignition time with the support of multi-source data. In particular, when faced with scenarios involving multiple smoke plumes or significant airflow disturbances, the inversion stability and adaptability of existing methods are poor, failing to meet the needs for highly dynamic, three-dimensional fire source inversion in forest fire scenarios. Summary of the Invention
[0004] This invention provides a precise positioning system for forest fire prevention based on lidar aerosol detection.
[0005] The forest fire prevention precision positioning system based on lidar aerosol detection includes the following modules: Data acquisition and fusion module: acquires forest canopy height model data of the monitoring area and simultaneously acquires real-time three-dimensional aerosol distribution data obtained by lidar scanning of the monitoring area, and fuses the two data. Canopy-modulated plume diffusion model construction module: Using the underlying surface boundary conditions characterized by the forest canopy height model, combined with real-time near-surface wind speed and direction, a canopy-modulated plume diffusion model describing the diffusion of plume material near the canopy interface is constructed. Fire source location module: The real-time three-dimensional aerosol distribution data is used as input to the canopy modulated plume diffusion model. Through reverse simulation calculation, the fire source location most likely to generate the aerosol distribution is deduced and determined, and the precise fire source coordinates are output.
[0006] Optionally, the data acquisition and fusion module specifically includes: Forest canopy height model data acquisition unit: By calling pre-stored historical scanning data of airborne lidar and satellite remote sensing images, and calculating the difference between the digital elevation model and the digital surface model, a high spatial resolution forest canopy height model of the monitored area is generated; Real-time three-dimensional aerosol distribution data synchronous acquisition unit: By deploying multiple ground-based micropulse lidar stations in the monitoring area, the atmosphere is scanned, and aerosol backscattering signal intensity data marked with timestamps and three-dimensional geographic coordinates is acquired and transmitted back in real time to generate real-time three-dimensional aerosol distribution data. Spatiotemporal alignment and fusion processing unit: unifies the real-time three-dimensional aerosol distribution data and the forest canopy height model data to the same geographic coordinate system and the same spatial resolution; performs terrain and canopy height correction and fusion on the aerosol distribution data to generate a fused three-dimensional spatial concentration field.
[0007] Optionally, the real-time three-dimensional aerosol distribution data synchronization acquisition unit specifically includes: Aerosol backscattering signal acquisition: By deploying multiple ground-based micropulse lidar stations in the monitoring area, vertical and horizontal scanning of the target atmospheric space is performed to acquire aerosol backscattering signal intensity data marked with timestamps and three-dimensional geographic coordinate information in real time, and the signal data is transmitted back to the data processing terminal; Three-dimensional aerosol distribution data generation: The aerosol backscattering signal intensity data is inverted and processed to convert the backscattering signal intensity into aerosol concentration information with spatial coordinates, thereby generating real-time three-dimensional aerosol distribution data of the monitoring area.
[0008] Optionally, the spatiotemporal alignment and fusion processing unit specifically includes: Spatiotemporal benchmark unification and data alignment: The real-time 3D aerosol distribution data and the forest canopy height model data are uniformly projected to the same geographic coordinate system and resampled to the same spatial resolution; based on the timestamps carried by the real-time 3D aerosol distribution data, the time periods corresponding to the real-time meteorological data are matched and aligned to form a spatiotemporally consistent dataset; Correction and fusion processing under canopy constraints: The real-time three-dimensional aerosol distribution data is subjected to terrain correction and canopy height correction, and the corrected aerosol distribution data is fused to generate a fused three-dimensional spatial concentration field.
[0009] Optionally, the canopy-modulated plume diffusion model construction module specifically includes: Canopy undersurface boundary condition generation unit: converts the forest canopy height model data into a three-dimensional digital surface model and extracts the undulating interface at the top of the canopy; Meteorological-driven parameterization unit: Receives real-time near-surface wind speed and direction data, and calculates canopy roughness and zero-plane displacement height parameters based on the vegetation spatial structure represented by the forest canopy height model, thereby correcting the wind field profile; Canopy Modulation Diffusion Model Construction Unit: Couples the physical underlying surface boundary with the modified wind field profile to construct a canopy modulation plume diffusion model that can describe the turbulent diffusion behavior of plumes on rough underlying surfaces and near the canopy-atmosphere interface.
[0010] Optionally, the canopy-modulated plume diffusion model introduces a canopy drag term and sets a turbulent diffusion coefficient related to canopy height, so that the horizontal and vertical diffusion processes of the plume are dynamically modulated by the forest canopy structure.
[0011] Optionally, the canopy undersurface boundary condition generation unit specifically includes: 3D surface model construction: The forest canopy height model data is converted into a 3D digital surface model to construct a spatial interface that reflects the undulating shape of the top of the canopy; Physical boundary condition extraction and definition: The canopy top undulation interface is extracted from the three-dimensional surface model and defined as the physical underlying surface boundary of the plume diffusion simulation, which characterizes the obstruction and drag effect of the forest canopy on airflow and material transport.
[0012] Optionally, the fire source location module specifically includes: Initial Hypothesis and Forward Simulation Unit: Hypothetical fire sources are set within the monitoring area, and initial emission intensity and time parameters are assigned to each hypothetical point; based on the canopy modulated plume diffusion model and the corresponding real-time meteorological conditions, forward diffusion simulation is performed on each hypothetical fire source to generate a predicted three-dimensional aerosol concentration distribution field; Observation-simulation matching evaluation unit: performs spatiotemporal matching and comparison between the predicted concentration distribution field and real-time three-dimensional aerosol distribution observation data; calculates the difference between the predicted distribution and the observed distribution under each hypothetical fire source scenario through a predefined cost function; Reverse Iterative Optimization Unit: Calls the reverse simulation optimization algorithm to automatically and iteratively adjust the spatial location, emission intensity, and ignition time parameters of the assumed fire source with the goal of minimizing the difference. Optimal fire source determination and output unit: The hypothetical fire source parameter that minimizes the difference between the predicted distribution and the observed distribution when the convergence condition is met is determined as the most likely actual fire source.
[0013] Optionally, the predefined cost function is obtained by weighted accumulation of the difference between the predicted concentration and the observed concentration at each spatial location and time point within the monitoring area. The weight is used to reflect the importance of different heights, time periods, or data reliability to the matching results. The smaller the value of the cost function, the higher the degree of matching between the simulated results of the hypothetical fire source and the actual observation results.
[0014] Optionally, the reverse iterative optimization unit uses a particle swarm optimization algorithm to perform reverse optimization on the assumed fire source parameters, taking the spatial location, emission intensity, and ignition time of the fire source as the position parameters of the particles, and the difference as the fitness evaluation index of the particles; by performing forward diffusion simulation and matching degree evaluation on the fire source parameters corresponding to each particle, the particle parameters are updated, so that the difference gradually decreases until the preset convergence condition is met.
[0015] The beneficial effects of this invention are: This invention integrates the physical modulation effect of forest canopy structure on the plume diffusion process, constructing a canopy-modulated plume diffusion model with realistic underlying surface boundaries, canopy roughness correction, and highly correlated diffusion coefficients. This enhances the spatial resolution capability of plume concentration evolution in complex terrain and high-canopy vegetation environments. Compared to traditional diffusion models based on simplified wind fields or unstructured surfaces, it exhibits higher accuracy in simulating near-surface aerosol distribution and is suitable for plume identification and concentration extrapolation during the early, low-intensity stages of wildfires.
[0016] This invention constructs a reverse fire source identification mechanism based on particle swarm optimization (PSO) algorithm. It combines hypothetical fire source parameters with forward diffusion simulation, achieving high-precision automatic identification of fire source location, intensity, and ignition time by minimizing the simulation-observation matching degree. This invention can deduce the most probable ignition point even with limited observation data, making it particularly suitable for forest areas where conventional monitoring methods are ineffective due to limited visibility and smoke obscured by tree canopies. It provides an efficient and intelligent reverse-location method for forest fire prevention, possessing strong engineering practicality and algorithmic scalability. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a schematic diagram of the system modules according to an embodiment of the present invention; Figure 2 This is a schematic diagram of fire source location according to an embodiment of the present invention. Detailed Implementation
[0019] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art may employ other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.
[0020] like Figures 1-2 As shown, the forest fire prevention precision positioning system based on lidar aerosol detection includes the following modules: Data acquisition and fusion module: acquires forest canopy height model data of the monitoring area and simultaneously acquires real-time three-dimensional aerosol distribution data obtained by lidar scanning of the monitoring area, and fuses the two data. The data acquisition and fusion module specifically includes: Forest canopy height model data acquisition unit: By calling historically stored airborne lidar scan data and high-resolution satellite remote sensing imagery, it extracts the digital elevation model and digital surface model of the monitored area, and then generates the forest canopy height model through difference calculation. , represented as: ; in, For any point within the monitoring area Forest canopy height, characterizing the differences in vegetation height across different areas of a forest region, is used to construct canopy diffusion boundary conditions. This includes a digital surface model of the land surface and vegetation, measured in meters, ranging from 10 to 80 meters, generated using lidar point clouds. The digital elevation model for bare terrain, with values ranging from 0 to 40 meters, is obtained by filtering and extracting bare points from lidar, reflecting the terrain undulations; the canopy height model has a high spatial resolution of 1-10 meters and covers the entire area.
[0021] Real-time 3D aerosol distribution data synchronous acquisition unit: By deploying multiple ground-based micropulse lidar stations within the monitoring area, scanning at different heights and horizontally, and acquiring calibrated, timestamped data... With three-dimensional geographic coordinates Backscattering signal intensity The real-time aerosol mass concentration field was calculated. , represented as: ; in, This represents the aerosol concentration, in units of... , The backscattered signal intensity is expressed in units of 1. The range of values is Characterizing the scattering ability of aerosols at different altitudes to a laser beam is fundamental for concentration inversion. This is the laser radar wavelength, taken as 532nm. Different wavelengths have different sensitivities to particle size; 532nm is more sensitive to fine particulate matter and is suitable for forest fire monitoring. Atmospheric pressure, temperature, and relative humidity corresponding to time are used for atmospheric correction, with atmospheric pressure ranging from 85 to 110 kPa. Mie scattering correction is used to address the impact on signal propagation and inversion accuracy. Temperature ranges from -30 to 50°C, and relative humidity ranges from 0 to 100%. The concentration inversion function is based on the Mie scattering inversion principle. The scattering signal is converted into mass concentration and then fitted using an optical parameter library.
[0022] Spatiotemporal alignment and fusion processing unit: This unit processes the aforementioned real-time three-dimensional aerosol concentration field. With canopy height model Projected to a unified geographic coordinate system and resampled to a grid with the same spatial resolution; using the timestamp of each scan in the LiDAR data. Based on the baseline, corresponding meteorological data at the same moment are matched to complete the alignment process of meteorological and aerosol time periods, and the three-dimensional volume constraint region is defined. Represented as: ; The original aerosol concentration field is Interpolation mapping is performed within the region, based on the canopy modulation coefficient. After correcting the vertical concentration distribution, the fused three-dimensional aerosol field is obtained. , represented as: ; in, To integrate the corrected aerosol concentration field, The canopy modulation coefficient has a value range of 1. The damping effect of vegetation on the vertical diffusion of aerosols was simulated, with smaller values indicating more restricted vertical transport. This was obtained through CFD simulation.
[0023] Canopy-modulated plume diffusion model construction module: Using the underlying surface boundary conditions represented by the forest canopy height model and combined with real-time near-surface wind speed and direction, a canopy-modulated plume diffusion model describing the diffusion of plume material near the canopy interface is constructed. The module for constructing the canopy-modulated plume diffusion model specifically includes: Canopy underlying surface boundary condition generation unit: This unit generates boundary conditions for forest canopy height model data. Convert to 3D canopy interface function Defined as the physical underlying surface boundary of the plume diffusion model, this boundary is used to characterize the irregular undulating terrain at the top of the forest canopy and plays the following physical roles in diffusion modeling, including: It creates shear and turbulent disturbances in the low-level wind field; It creates resistance and drag on the exchange of aerosol particles between the canopy and the free atmosphere; The boundary conditions of the underlying surface are expressed as follows: ; Used to constrain the plume diffusion boundary conditions at the bottom of the simulation domain. It is a three-dimensional spatial set of the boundary of the canopy surface, used to constrain the bottom boundary of the simulation domain and limit the initiation and contact surface of plume diffusion.
[0024] Meteorological-driven parameterization unit: To accurately drive the near-surface plume diffusion model, canopy roughness correction is required for surface wind speed and direction, considering the influence of vegetation structure on the vertical wind speed profile. Specifically, this includes: (1) Canopy roughness length: ; (2) Zero-plane displacement height: ; in, This is the roughness length, characterizing the degree to which the rough structure of the earth's surface affects wind; the unit is meters, and the value ranges from 0.5 to 5 meters. The zero-plane displacement height represents the location of the effective wind speed calculation point within the canopy, with a value ranging from 3 to 40 meters. This is the roughness coefficient, ranging from 0.05 to 0.15. Differences in vegetation type affect wind resistance. The zero plane displacement coefficient, with a value range of 0.6-0.8, represents the relative height position of the wind speed profile starting point within the canopy. (3) Wind speed profile correction: The logarithmic wind speed profile formula is used for correction, expressed as: ; in, For height Wind speed at that location, in m / s. Friction velocity, ranging from 0.1 to 1.0, characterizes the frictional effect of wind on rough surfaces and is a fundamental parameter for wind field modeling. This represents the von Kármán constant, with a value of 0.4. Canopy Modulation Diffusion Model Construction Unit: Constructs a physical model of plume diffusion that couples the underlying surface boundary and the modified wind field. The canopy modulation plume diffusion model introduces the dynamic modulation effect of the canopy on plume behavior, which is manifested in the following two core mechanisms, specifically including: (1) Canopy drag introduction: In the canopy-modulated plume diffusion model, a canopy drag term is added to the momentum equation. , represented as: ; in, The canopy drag coefficient, ranging from 0.1 to 0.3, characterizes the intensity of momentum exchange between wind and blades and can be obtained through wind tunnel experiments. The leaf area density of local vegetation is used to characterize the density of the contact surface between plants and wind, and to simulate the momentum exchange and energy dissipation between wind and vegetation. Its value ranges from 0.1 to 5 m² / m³. The term represents the canopy drag force, used to simulate wind speed decay and energy dissipation. (2) Parameterization of turbulent diffusion coefficient: To describe the mixing intensity of aerosols in the canopy region, a highly correlated turbulent diffusion coefficient is introduced. , represented as: ; in, is the vertical turbulent diffusion coefficient, describing the variation of the plume's diffusion capacity with height, and its unit is . , This is the maximum diffusion coefficient in free atmosphere, ranging from 10 to 100 m² / s. It depends on the intensity of free atmospheric turbulence and affects the mixing and diffusion rate of the plume. To control the power exponent of diffusion attenuation, the value is set to 1-2, which controls the attenuation rate of the diffusion coefficient near the top of the canopy and reflects the barrier effect of the top of the canopy.
[0025] Fire source location module: It takes real-time three-dimensional aerosol distribution data as input to the canopy-modulated smoke aerosol diffusion model, and through inverse simulation calculation, it deduces and determines the fire source location most likely to generate aerosol distribution, and outputs accurate fire source coordinates. The fire source location module specifically includes: Initial assumptions and forward simulation units: Multiple hypothetical fire sources are set within the monitoring area, each point... Assigned the following parameter set, represented as: ; in, These are the coordinates of the hypothetical fire source, in meters, representing the spatial distribution of potential ignition points and serving as the starting point for forward simulation. The emission intensity of the assumed ignition source is expressed in units of... The value ranges from 1 to 100 kg / s, simulating the source intensity of the smoke plume from a fire source. The assumed ignition time is in seconds. The current monitoring time is shifted forward by 0-3600 seconds to consider the possibility that the plume started to spread at different historical moments, which would affect the spatial structure of the current concentration. For each set of hypothetical parameters As input, the canopy-modulated plume diffusion model is invoked. Under given real-time weather conditions Next, a forward diffusion simulation is performed to generate the predicted three-dimensional concentration distribution field, represented as: ; in, For the first The predicted concentration field of a hypothetical fire source. To and The corresponding meteorological input set includes real-time wind speed, wind direction, temperature, and humidity; Observation-simulation matching evaluation unit: This unit evaluates each predicted concentration field. The observed concentration field provided by the fusion module A comparison is made using a pre-defined cost function. The degree of difference is calculated and expressed as: ; Cost function The weighted Euclidean distance, measuring the global difference between predictions and observations, serves as the inverse optimization objective and is expressed as: ; in, For the first The predicted three-dimensional aerosol concentration field corresponding to each hypothetical fire source is given in μg / m³. The observed aerosol concentration field is obtained by the real-time fusion module, with values ranging from 0 to 1000 μg / m³. This is a spatial weighting coefficient, which can be weighted according to altitude, time, or data confidence level. This is a measure of the matching error between the predicted field and the observed field; the smaller the value, the better the match. Inverse Iterative Optimization Unit: Calls the inverse optimization algorithm to minimize the difference. For the objective function, iteratively update the hypothetical fire source parameters. , represented as: ; Each iteration includes the following process: Use current parameters Perform forward simulation; Generate prediction field ; Calculate the difference between the observed and actual fields. ; like If the condition is met, the iteration terminates; otherwise, it continues.
[0026] in, This represents the total difference between the predicted field and the observed field for the nth hypothetical fire source. The smaller the value, the closer it is to the observation. The convergence threshold, ranging from 0.01 to 5.0, controls the iteration stopping condition and avoids overfitting or computational redundancy. For the first The fire source parameter set in the next iteration. For the first Parameter update amount in each iteration For the predicted concentration field generated in the current iteration, This represents the matching error value in the current iteration.
[0027] The inverse optimization algorithm is a particle swarm optimization algorithm, which specifically includes: (1) Initial settings: First, multiple hypothetical fire sources are randomly generated within the monitoring area. Each fire source includes five parameters: spatial location. Emission intensity Time of ignition These hypotheses suggest that the fire source is like a particle, serving as the starting point for the search.
[0028] (2) Forward simulation: For each hypothetical fire source, a forward simulation is performed using the plume diffusion model to obtain the predicted aerosol concentration distribution map.
[0029] (3) Error calculation: The simulation results are compared with the real three-dimensional aerosol data obtained by lidar observation, and the error value of each hypothetical fire source is calculated by the difference function.
[0030] (4) Particle update: Based on the current error results, combined with the historical performance of each particle and the global performance, the parameters of each particle are adjusted and the next round of simulation is carried out.
[0031] (5) Iterative optimization: Repeat the forward simulation, error evaluation and parameter update, gradually converge to a solution with minimum error. Stop when the error threshold or the maximum number of iterations is reached.
[0032] (6) Output results: The hypothetical fire source with the smallest error is considered the real fire source, and its precise coordinates, intensity and estimated ignition time are output.
[0033] Optimal fire source determination and output unit: When the iteration meets the convergence condition, determine the current parameter set. The optimal fire source solution is represented as: ; The final output includes: optimal fire source coordinates Estimate emission intensity Start time and matching confidence index , range .
[0034] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.
[0035] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A precise forest fire prevention positioning system based on lidar aerosol detection, characterized in that, Includes the following modules: Data acquisition and fusion module: acquires forest canopy height model data of the monitoring area and simultaneously acquires real-time three-dimensional aerosol distribution data obtained by lidar scanning of the monitoring area, and fuses the two data. Canopy-modulated plume diffusion model construction module: Using the underlying surface boundary conditions characterized by the forest canopy height model, combined with real-time near-surface wind speed and direction, a canopy-modulated plume diffusion model describing the diffusion of plume material near the canopy interface is constructed. Fire source location module: The real-time three-dimensional aerosol distribution data is used as input to the canopy-modulated smoke aerosol diffusion model. Through reverse simulation calculation, the most likely fire source location to generate the aerosol distribution is deduced and determined, and the precise fire source coordinates are output.
2. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 1, characterized in that, The data acquisition and fusion module specifically includes: Forest canopy height model data acquisition unit: By calling pre-stored historical scanning data of airborne lidar and satellite remote sensing images, and calculating the difference between the digital elevation model and the digital surface model, a high spatial resolution forest canopy height model of the monitored area is generated; Real-time three-dimensional aerosol distribution data synchronous acquisition unit: By deploying multiple ground-based micropulse lidar stations in the monitoring area, the atmosphere is scanned, and aerosol backscattering signal intensity data marked with timestamps and three-dimensional geographic coordinates is acquired and transmitted back in real time to generate real-time three-dimensional aerosol distribution data. Spatiotemporal alignment and fusion processing unit: unifies the real-time three-dimensional aerosol distribution data and the forest canopy height model data to the same geographic coordinate system and the same spatial resolution; performs terrain and canopy height correction and fusion on the aerosol distribution data to generate a fused three-dimensional spatial concentration field.
3. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 2, characterized in that, The real-time three-dimensional aerosol distribution data synchronous acquisition unit specifically includes: Aerosol backscattering signal acquisition: By deploying multiple ground-based micropulse lidar stations in the monitoring area, vertical and horizontal scanning of the target atmospheric space is performed to acquire aerosol backscattering signal intensity data marked with timestamps and three-dimensional geographic coordinate information in real time, and the signal data is transmitted back to the data processing terminal; Three-dimensional aerosol distribution data generation: The aerosol backscattering signal intensity data is inverted and processed to convert the backscattering signal intensity into aerosol concentration information with spatial coordinates, thereby generating real-time three-dimensional aerosol distribution data of the monitoring area.
4. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 2, characterized in that, The spatiotemporal alignment and fusion processing unit specifically includes: Spatiotemporal benchmark unification and data alignment: The real-time 3D aerosol distribution data and the forest canopy height model data are uniformly projected to the same geographic coordinate system and resampled to the same spatial resolution; based on the timestamps carried by the real-time 3D aerosol distribution data, the time periods corresponding to the real-time meteorological data are matched and aligned to form a spatiotemporally consistent dataset; Correction and fusion processing under canopy constraints: The real-time three-dimensional aerosol distribution data is subjected to terrain correction and canopy height correction, and the corrected aerosol distribution data is fused to generate a fused three-dimensional spatial concentration field.
5. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 1, characterized in that, The canopy-modulated plume diffusion model construction module specifically includes: Canopy undersurface boundary condition generation unit: converts the forest canopy height model data into a three-dimensional digital surface model and extracts the undulating interface at the top of the canopy; Meteorological-driven parameterization unit: Receives real-time near-surface wind speed and direction data, and calculates canopy roughness and zero-plane displacement height parameters based on the vegetation spatial structure represented by the forest canopy height model, thereby correcting the wind field profile; Canopy Modulation Diffusion Model Construction Unit: Couples the physical underlying surface boundary with the modified wind field profile to construct a canopy modulation plume diffusion model that can describe the turbulent diffusion behavior of plumes on rough underlying surfaces and near the canopy-atmosphere interface.
6. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 5, characterized in that, The canopy-modulated plume diffusion model introduces a canopy drag term and sets a turbulent diffusion coefficient related to canopy height, so that the horizontal and vertical diffusion processes of the plume are dynamically modulated by the forest canopy structure.
7. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 5, characterized in that, The canopy underlying surface boundary condition generation unit specifically includes: 3D surface model construction: The forest canopy height model data is converted into a 3D digital surface model to construct a spatial interface that reflects the undulating shape of the top of the canopy; Physical boundary condition extraction and definition: The canopy top undulation interface is extracted from the three-dimensional surface model and defined as the physical underlying surface boundary of the plume diffusion simulation, which characterizes the obstruction and drag effect of the forest canopy on airflow and material transport.
8. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 1, characterized in that, The fire source location module specifically includes: Initial Hypothesis and Forward Simulation Unit: Hypothetical fire sources are set within the monitoring area, and initial emission intensity and time parameters are assigned to each hypothetical point; based on the canopy modulated plume diffusion model and the corresponding real-time meteorological conditions, forward diffusion simulation is performed on each hypothetical fire source to generate a predicted three-dimensional aerosol concentration distribution field; Observation-simulation matching evaluation unit: performs spatiotemporal matching and comparison between the predicted concentration distribution field and real-time three-dimensional aerosol distribution observation data; calculates the difference between the predicted distribution and the observed distribution under each hypothetical fire source scenario through a predefined cost function; Reverse Iterative Optimization Unit: Calls the reverse simulation optimization algorithm to automatically and iteratively adjust the spatial location, emission intensity, and ignition time parameters of the assumed fire source with the goal of minimizing the difference. Optimal fire source determination and output unit: The hypothetical fire source parameter that minimizes the difference between the predicted distribution and the observed distribution when the convergence condition is met is determined as the most likely actual fire source.
9. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 8, characterized in that, The predefined cost function is calculated by weighted accumulation of the difference between the predicted concentration and the observed concentration at each spatial location and time point within the monitoring area. The weights are used to reflect the importance of different altitudes, time periods, or data reliability to the matching results. The smaller the cost function value, the higher the degree of matching between the simulated results of the hypothetical fire source and the actual observation results.
10. The forest fire prevention precise positioning system based on lidar aerosol detection according to claim 8, characterized in that, The reverse iterative optimization unit uses the particle swarm optimization algorithm to perform reverse optimization on the assumed fire source parameters. The spatial location, emission intensity, and ignition time of the fire source are used as the position parameters of the particles, and the difference is used as the fitness evaluation index of the particles. By performing forward diffusion simulation and matching degree evaluation on the fire source parameters corresponding to each particle, the particle parameters are updated so that the difference is gradually reduced until the preset convergence condition is met.