Fire scene unmanned aerial vehicle risk assessment method and device, electronic equipment and storage medium

Abstract

The application provides a fire scene unmanned aerial vehicle risk assessment method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring fire scene measured data of a fire scene; constructing a fire scene dynamic evolution model according to the fire scene measured data; acquiring unmanned aerial vehicle parameters, and determining disaster-causing factors corresponding to a grid in the fire scene model according to the unmanned aerial vehicle parameters and the fire scene dynamic evolution model; for each grid, determining a risk level corresponding to the grid based on the disaster-causing factors; and obtaining a fire scene risk assessment result of the unmanned aerial vehicle by comprehensively considering the risk levels corresponding to all the grids. The evolution process of the risk indexes in the fire scene is dynamically quantified by constructing the fire scene dynamic evolution model, so that the risk conditions of the specific grid in the fire scene at different times can be determined, and the overall risk of the unmanned aerial vehicle in the fire scene can be assessed, thereby providing dynamic risk assessment for the safe flight of the unmanned aerial vehicle.

Description

Technical Field

[0001] This invention relates to the field of forest fire prevention and control and low-altitude drone technology, and in particular to a method, device, electronic equipment and storage medium for risk assessment of fire sites using drones. Background Technology

[0002] Forest fires, as one of the world's major natural disasters, pose a significant threat to natural ecosystems, socio-economic development, and human lives due to their high complexity, suddenness, and destructiveness. In recent years, drones, with their advantages of high flexibility and rapid response, have become one of the core pieces of equipment for forest fire emergency rescue. However, accidents involving drones going out of control and being damaged due to complex fire environments are frequent, and the issue of safe drone flight operations in extreme environments is becoming increasingly prominent. Currently, there is a lack of quantitative risk assessment and analysis methods to support drone flights over fire sites, making it difficult to cope with the challenges posed by the complex and ever-changing extreme environments of forest fires. Summary of the Invention

[0003] The main objective of this invention is to propose a method, device, electronic device, and storage medium for assessing the risks of drones in fire situations, aiming to solve the problem of difficulty in quantifying the risks of drones in complex fire environments in the prior art.

[0004] To achieve the above objectives, the present invention provides a method for risk assessment of fireground drones, the method comprising the following steps: Obtain actual fire scene measurement data; A dynamic evolution model of the fire scene is constructed based on the measured fire scene data. The parameters of the UAV are obtained, and the disaster-causing factors corresponding to the grid in the fire field model are determined based on the UAV parameters and the fire field dynamic evolution model. The fire field model is a model constructed based on the space where the fire field is located, and the grid is a discrete grid that constitutes the fire field model. For each grid, the risk level corresponding to the grid is determined based on the disaster-causing factor; The fire risk assessment result of the UAV is obtained by combining the risk levels corresponding to each grid.

[0005] Optionally, constructing a dynamic evolution model of the fire scene based on the measured fire scene data includes: Obtain wind field data from the measured fire field data, and construct a wind field projection model based on the wind field data; Obtain fire data from the measured fire data, and construct a fire simulation model based on the fire data; Obtain smoke data from the measured fire scene data, and construct a smoke diffusion simulation model based on the smoke data; The dynamic evolution model of the fire field is obtained by combining the wind field simulation model, the fire field simulation model, and the smoke diffusion simulation model.

[0006] Optionally, the disaster-causing factors include environmental disaster-causing factors, and determining the disaster-causing factors corresponding to the grid in the fire scene model based on the UAV parameters and the fire scene dynamic evolution model includes: In the fire dynamic evolution model, the wind turbulence index corresponding to the grid at the target time is determined, and the corresponding turbulence hazard factor is determined based on the wind turbulence index; In the fire dynamic evolution model, the heat release index of the grid at the target time and the distance between the UAV and the grid are determined, and the corresponding thermal radiation factor is determined based on the heat release index and the distance. In the fire dynamic evolution model, the particle concentration data of the grid at the target time and the particle resistance of the UAV are determined, and the corresponding smoke factor is determined based on the particle concentration data and the particle resistance. The environmental disaster factor corresponding to the grid is obtained by combining the turbulence hazard factor, the thermal radiation factor, and the smoke factor.

[0007] Optionally, the disaster-causing factors include the vulnerability factor of the UAV, and the step of determining the disaster-causing factors corresponding to the grid in the fire model based on the UAV parameters and the fire dynamic evolution model includes: In the dynamic evolution model of the fire site, the dominant turbulence frequency in the grid is determined, and the flight component parameters of the UAV are obtained; The aerodynamic stability vulnerability of the UAV is determined based on the dominant turbulence frequency and the parameters of the flight components. The simulated motor temperature of the UAV is determined in the dynamic evolution model of the fire scene, and the operating temperature of the UAV motor is obtained. The thermal vulnerability of the UAV is determined based on the predicted motor temperature and the motor operating temperature. The smoke concentration of the grid is determined in the fire dynamic evolution model, and the optical detection parameters of the UAV are obtained; The sensor vulnerability of the UAV is determined based on the smoke concentration and the optical detection parameters; The vulnerability factor of the UAV bearing corresponding to the grid is obtained by combining the aerodynamic stability vulnerability, thermal vulnerability and sensor vulnerability.

[0008] Optionally, the disaster-causing factor includes the drone-borne disaster exposure factor, and the step of determining the disaster-causing factor corresponding to the grid in the fire scene model based on the drone parameters and the fire scene dynamic evolution model includes: The expected arrival time of the UAV at the grid is determined in the dynamic evolution model of the fire site; The trajectory correction coefficient is determined based on the historical flight data of the UAV; The drone disaster exposure factor corresponding to the grid is determined based on the expected arrival time and the trajectory correction coefficient.

[0009] Optionally, determining the risk level corresponding to each grid cell based on the hazard-causing factor includes: For each of the disaster-causing factors, obtain the expert weight and historical data weight corresponding to the disaster-causing factor; The target weight corresponding to the disaster-causing factor is obtained by combining the expert weights and the historical data weights. The risk index corresponding to the grid is obtained by weighting each of the disaster-causing factors based on the target weights. Determine the risk level corresponding to the risk index.

[0010] Optionally, determining the risk level corresponding to the risk index includes: Determine the risk range in which the risk index falls; The risk level corresponding to the risk interval is taken as the risk level corresponding to the risk index.

[0011] To achieve the above objectives, the present invention also provides a fireground drone risk assessment device, the fireground drone risk assessment device comprising: The first acquisition module is used to acquire measured data of the fire scene; The first construction module is used to construct a dynamic evolution model of the fire scene based on the measured fire scene data; The second acquisition module is used to acquire UAV parameters and determine the disaster-causing factors corresponding to the grid in the fire field model based on the UAV parameters and the fire field dynamic evolution model. The fire field model is a model constructed based on the space where the fire field is located, and the grid is a discrete grid that constitutes the fire field model. The first determining module is used to determine the risk level corresponding to each grid based on the disaster-causing factor. The first comprehensive module is used to integrate the risk levels corresponding to each grid to obtain the fire risk assessment result of the UAV.

[0012] To achieve the above objectives, the present invention also provides an electronic device, the electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the fireground drone risk assessment method as described above.

[0013] To achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the fireground drone risk assessment method as described above.

[0014] This invention proposes a method, device, electronic equipment, and storage medium for fire scene drone risk assessment. The method involves acquiring measured fire scene data; constructing a dynamic fire scene evolution model based on the measured data; acquiring drone parameters; and determining the hazard factors corresponding to grids in the fire scene model based on the drone parameters and the dynamic fire scene evolution model. The fire scene model is a model constructed based on the spatial location of the fire scene, and the grids are discrete grids constituting the fire scene model. For each grid, the risk level corresponding to that grid is determined based on the hazard factors. The fire scene risk assessment result for the drone is obtained by combining the risk levels corresponding to each grid. By constructing a dynamic fire scene evolution model, the evolution process of risk indicators in the fire scene is dynamically quantified, enabling the determination of the risk status of specific grids in the fire scene at different times. This allows for the assessment of the overall risk of the drone within the fire scene, providing dynamic risk assessment for the safe flight of the drone. Attached Figure Description

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the first embodiment of the fire scene drone risk assessment method of the present invention; Figure 2 This is a schematic diagram illustrating the overall implementation principle of the fire scene drone risk assessment method of the present invention; Figure 3 This is a detailed flowchart of the fire scene drone risk assessment method of the present invention; Figure 4 This is a schematic diagram of the module structure of the electronic device of the present invention. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0019] This invention provides a method for risk assessment of fire scene drones, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the fire scene drone risk assessment method of the present invention. The method includes the following steps: Step S10: Obtain actual fire scene measurement data; The fire site is the location where a fire occurs; the fire site can also refer to the location where a complex forest fire starts.

[0020] The measured data at the fire site are data obtained by detecting the specific conditions of the fire site. The specific types of measured data at the fire site can be set according to actual needs. For example, the measured data at the fire site can include fire site topography, wind speed, wind direction, combustion status, smoke conditions, etc. Understandably, specific types of measured data at the fire site can be collected based on corresponding acquisition devices. For example, fire site topography can be obtained by obtaining the topographic map of the corresponding location from map software, or it can be determined by analyzing images of the fire site. Wind speed and wind direction can be detected by wind detection devices set up at the fire site.

[0021] Step S20: Construct a dynamic evolution model of the fire scene based on the measured fire scene data; The dynamic evolution model of a fire is a model that extrapolates specific environmental changes in a fire based on measured fire data.

[0022] The measured data at the fire site indicates the actual environmental state of the fire site. By constructing a dynamic evolution model of the fire site based on the measured data, it is possible to analyze the changing trend of the fire site and provide a data foundation for the risk assessment of drones.

[0023] A dynamic evolution model of a fire can be constructed by setting variations between simulation time steps.

[0024] Step S30: Obtain UAV parameters, and determine the disaster-causing factors corresponding to the grid in the fire field model based on the UAV parameters and the fire field dynamic evolution model, wherein the fire field model is a model constructed based on the space where the fire field is located, and the grid is a discrete grid that constitutes the fire field model. The parameters of a drone are those of the drone itself; the specific types of drone parameters can be set based on actual needs, such as the drone's rated operating parameters, drone location, and drone working parameters.

[0025] It is understandable that different drones have different requirements for the operating environment. For example, some drones are more heat-resistant, while others can work in poorer visual environments. Therefore, the drone parameters themselves reflect the drone's risk tolerance to a certain extent. Thus, in this embodiment, the disaster-causing factors are determined by combining the drone parameters, so that the disaster-causing factors can match the actual capabilities of the drone.

[0026] The dynamic evolution model of the fire field indicates the environmental conditions of the fire field, i.e. the risks to drones; while the drone parameters indicate the drone's tolerance to environmental risks. Therefore, by combining the dynamic evolution model of the fire field and the drone parameters, the disaster-causing factors that indicate the risks to drones can be identified.

[0027] Disaster-causing factors reflect the risks that drones face in a fire. Disaster-causing factors are specifically set for the specific types of risks that drones pose in a fire. For example, disaster-causing factors may include environmental disaster-causing factors, drone exposure factors, and drone vulnerability factors. Environmental disaster-causing factors indicate the degree of risk of the environment itself, drone exposure factors indicate the degree of interaction between the drone and the environment, and drone vulnerability factors indicate the degree of risk resistance of the drone itself.

[0028] In this embodiment, a fire scene model corresponding to the fire scene is first constructed based on the actual spatial structure of the fire scene, such as the fire scene airspace. The fire scene model is then divided into a set of regular, closely adjacent three-dimensional grids, with each grid serving as the smallest risk determination unit for risk assessment.

[0029] Understandably, fire scene models consist of multiple grids, each with states corresponding to different times. Therefore, it is necessary to align the data of the fire scene dynamic evolution model in time and space to ensure data consistency and the accuracy of grid data. For example, spatial alignment can establish spatial relationships between multiple sources of indicators through grid indexing, binding related data to the same grid cell to ensure a one-to-one correspondence between spatial locations. Temporal alignment can be based on the extrapolated time step Δt, synchronizing the time series of all indicators to construct a time series. For each time step t, all grid data corresponding to that time step can be extracted to ensure time synchronization of multiple sources of indicators and eliminate risk misjudgment caused by time differences.

[0030] Step S40: For each grid, determine the risk level corresponding to the grid based on the disaster-causing factor; Once the hazard factors for each grid are determined, the risk level of that grid can be calculated. It can be understood that the hazard factors indicate the specific risk situation of the grid. Therefore, the risk level of the grid can be obtained by combining all the hazard factors corresponding to that grid.

[0031] Step S50: The fire risk assessment result of the UAV is obtained by combining the risk levels corresponding to each grid.

[0032] Once the risk levels of all grid cells are determined, the risk status of all grid cells in the fire scene model can be obtained, thus enabling us to understand the risk status of drones at specific locations and ultimately obtain the overall fire scene risk assessment results.

[0033] When outputting specific fire risk assessment results, a dynamic risk map can be constructed based on a gridded 3D fire scene model. (See [link / reference]). Figure 2 For example, specific grids can be color-coded according to their corresponding risk levels, such as green for low risk, yellow for medium risk, and red for low risk wind turbines. This can generate a map of available flight risks in forest fire areas, providing dynamic risk assessment and early warning services for safe drone flight operations.

[0034] This embodiment constructs a dynamic evolution model of the fire scene to dynamically quantify the evolution process of risk indicators in the fire scene, enabling the determination of the risk status of specific grids in the fire scene at different times, thereby achieving an overall risk assessment of the UAV in the fire scene and providing dynamic risk assessment for the safe flight of the UAV.

[0035] Further details will follow. Figure 3 In the second embodiment of the fireground drone risk assessment method of the present invention based on the first embodiment, step S20 includes the following steps: Step S21: Obtain wind field data from the measured fire field data, and construct a wind field projection model based on the wind field data; The wind field simulation model is used to construct the spatiotemporal evolution relationship of wind speed and direction driven by the environment-fire coupling.

[0036] In this embodiment, the logic of initial wind field + multi-factor correction + spatiotemporal diffusion is used to construct the wind field extrapolation model.

[0037] Wind field data indicates wind-related data in the fire area; wind field data specifically includes customs, wind direction, topography, etc.

[0038] First, the initial wind field of the grid is constructed based on the wind field data:

[0039] Where (i, j, k) indicates the coordinates of the corresponding grid; Windinit The initial wind field in the grid; subsequent values ​​(i, j, k) are all indicator grids and will not be described further; Ws init Initial wind speed; Wd init This is the initial wind direction.

[0040] A multi-factor coupled correction model for the wind field is constructed by introducing correction coefficients for topography and forest fire thermal buoyancy:

[0041]

[0042] Among them, Ws corr The corrected initial wind speed; Wd corr The initial wind direction is corrected; (i, j, k, t) indicates the relevant parameters of the grid at the target time t under this coordinate system, and will not be repeated hereafter; K ter The terrain correction parameters are determined based on the collected terrain data and are positively correlated with wind speed enhancement; K heat This is the thermal buoyancy correction factor, which is positively correlated with the wind speed enhancement caused by thermal convection; ΔWd ter Terrain-guided correction angle; ΔWd heat This refers to the thermal buoyancy deflection angle; the above three parameters can be calibrated using test flight data of fire-fighting drones or historical data.

[0043] A wind field projection model is constructed based on the corrected wind field:

[0044] Among them, f σ ( ) represents the spatial neighborhood wind vector action function of the grid; Δt is the time step of the derivation.

[0045] Step S22: Obtain fire data from the measured fire data, and construct a fire simulation model based on the fire data; Fire simulation models are used to construct the evolutionary relationship between fire spread rate and the transition of combustion state in the grid.

[0046] Fire data indicates relevant information about the combustion status in a fire; specific data in fire data include fuel type, terrain, wind field, baseline spread rate, etc.

[0047] First, the forest fire spread rate RoS0 is determined by combining factors such as fuel, topography, and wind field, specifically using a multi-factor linear weighted correction model:

[0048]

[0049] Among them, Fuel grade The flammability rating of the fuel is determined based on the fuel type. w1 is the model hyperparameter; β0 is the baseline propagation rate; these two parameters can be set based on ignition experiments or historical data in different scenarios; K wind This is the wind field correction factor, determined based on the terrain at the fire site.

[0050] Combustion evolution rules are defined using state transition functions, and a grid combustion state transition model is constructed driven by both neighborhood combustion states and spread rates.

[0051]

[0052] Among them, B s B1 indicates the combustion state; B2 indicates the unignited state; B3 indicates the initial ignition state; B1, B2, and B3 can be dynamically adjusted to reflect the actual needs of the fire simulation. f h ( () is a mathematical mapping function, and its calculation methods include, but are not limited to, nonlinear deep neural networks, linear polynomial functions, etc. σ(RoS t σ(B) represents the fire spread rate from the neighboring grid to the target point; t b1 and b2 are hyperparameters that can be determined by ignition experiments or historical data.

[0053] The forest fire spread rate and the grid combustion state transition model constitute a fire field simulation model.

[0054] Step S23: Obtain smoke data from the measured fire data and construct a smoke diffusion simulation model based on the smoke data; Smoke diffusion simulation models are used for the transport of pollutants.

[0055] This embodiment combines the diffusion characteristics of particulate matter and toxic gases to achieve spatiotemporal dynamic quantification of smoke concentration and complete smoke diffusion simulation:

[0056] Where S is the smoke concentration of the grid; f Sc ( ) is a mathematical mapping function that can calculate the diffusion dynamics of gaseous particulate matter concentration. Its calculation methods include, but are not limited to, nonlinear deep neural networks and linear polynomial functions.

[0057] Step S24: Combine the wind field simulation model, the fire field simulation model, and the smoke diffusion simulation model to obtain the dynamic evolution model of the fire field.

[0058] In this embodiment, the overall situation of the fire is simulated by specifically constructing wind field simulation models, fire field simulation models and smoke diffusion simulation models, so as to accurately reflect the changes of the fire in time and space and provide a basis for risk estimation.

[0059] Furthermore, in the third embodiment of the fire scene drone risk assessment method of the present invention based on the first embodiment, the disaster-causing factors include environmental disaster-causing factors, and step S30 includes the following steps: Step S31: Determine the wind turbulence index corresponding to the grid at the target time in the fire dynamic evolution model, and determine the corresponding turbulence hazard factor based on the wind turbulence index; Environmental disaster-causing factors indicate the degree of risk inherent in the environment itself; The turbulence hazard factor specifically characterizes the threat intensity of fire wind vectors to drones.

[0060] The wind turbulence index indicates the characteristics of meteorological turbulence in a fire.

[0061] In this embodiment, the turbulence hazard factor is calculated by combining factors such as initial wind speed, terrain changes, and flame thermal buoyancy. Specifically, it can be quantified using methods such as linear polynomial functions.

[0062]

[0063] Among them, H w0 The initial wind turbulence index is determined by topography-wind dominance; W s It refers to wind speed, and Ter is the terrain slope.

[0064] w1, w2, and β0 are model hyperparameters, which can be derived from wind tunnel test calibration parameters or historical observation data.

[0065] ΔT f w3 represents the temperature change, and w3 is the thermal buoyancy correction factor.

[0066] H w The value of the turbulence hazard factor is the higher the value of the meteorological turbulence factor, the stronger the threat to the flight safety of the UAV, and vice versa.

[0067] Step S32: In the fire dynamic evolution model, determine the heat release index of the grid at the target time and the distance between the UAV and the grid, and determine the corresponding thermal radiation factor based on the heat release index and the distance; The thermal radiation factor indicates the threat level of unmanned aerial vehicle (UAV) systems to the thermal radiation released from burning forest fuels.

[0068] In this embodiment, the thermal radiation factor is determined based on the fire state and propagation distance. Specifically, an exponential decay function can be used to establish a quantification equation, such as:

[0069] Where Q is the fireline heat release rate, which is determined by fuel type, combustion load and combustion state; d is the Euclidean distance between the grid where the UAV is located and the grid currently being analyzed. k smoke w4 is the smoke attenuation coefficient; w4 is the weighting parameter. Both of these can be calculated from thermal radiation experimental calibration parameters or historical observation data. H h The thermal radiation factor is the factor that determines the thermal stress exerted on the UAV system by thermal radiation. A higher thermal radiation factor indicates a stronger thermal stress on the UAV system.

[0070] Step S33: In the fire dynamic evolution model, determine the particle concentration data of the grid at the target time and the particle resistance of the UAV, and determine the corresponding smoke factor based on the particle concentration data and the particle resistance; Smoke factor characterizes the threat intensity of particulate matter and harmful gases released from burning vegetation along the fire line to drone sensors.

[0071] In this embodiment, considering factors such as particulate shielding effect, chemical circuit contamination, and physical intrusion, the main smoke components affecting UAV system components, such as PM2.5 and CO, are selected as risk indicators, and a quantitative mapping equation is established using a linear weighted function:

[0072] Among them, C PM C co These represent PM2.5 and CO concentrations from a three-dimensional airspace grid, respectively. α vis With α tox The weighting coefficient can be calculated from sensor failure test calibration parameters or historical observation data; H s The smoke factor is the factor that determines the level of interference from smoke to the drone system. A higher smoke factor value indicates a greater level of interference from smoke.

[0073] Step S34: Combine the turbulence hazard factor, the thermal radiation factor, and the smoke factor to obtain the environmental disaster factor corresponding to the grid.

[0074] Specifically, environmental disaster-causing factors can be obtained by summing or weighting the turbulence hazard factor, thermal radiation factor, and smoke factor.

[0075] In this embodiment, by constructing turbulence hazard factors, thermal radiation factors, and smoke factors, it is possible to determine three key environmental disaster-causing factors in the fire field: meteorological turbulence, high-temperature thermal radiation, and combustion smoke. This allows for the accurate determination of the environmental risks posed by the grid and the acquisition of environmental disaster-causing factors.

[0076] Furthermore, in the fourth embodiment of the fire scene drone risk assessment method proposed based on the first embodiment of the present invention, the disaster-causing factor includes the drone-borne vulnerability factor, and step S30 includes the following steps: Step S35: Determine the dominant turbulence frequency in the grid in the fire dynamic evolution model, and obtain the flight component parameters of the UAV; Step S36: Determine the aerodynamic stability vulnerability of the UAV based on the dominant turbulence frequency and the flight component parameters; The vulnerability factor carried by a drone indicates the drone's resilience to risks.

[0077] Aerodynamic stability vulnerability describes the ability of an unmanned aerial vehicle (UAV) to maintain flight control stability under different turbulent disturbances; in this embodiment, it is quantified using a frequency response model.

[0078] Among them, f t The dominant turbulence frequency is determined by real-time wind speeds obtained from fire turbulence monitoring data or calculated using wind field modeling. This can be combined with inherent parameters of the UAV and empirical coefficients to quickly estimate the dominant turbulence frequency of each three-dimensional grid. r Where f is the rotor diameter of the UAV, and f is an inherent parameter of the device; c The inherent frequency of the flight control system is determined by the equipment hardware design. The hyperparameter k is the optimal empirical value obtained by fitting wind tunnel experiments and fire field measured data. Different UAVs can be adjusted according to their differences in anti-interference performance.

[0079] V a The value of aerodynamic stability vulnerability indicates that the larger the value, the weaker the ability of the UAV system to maintain flight control stability under turbulent disturbances.

[0080] Step S37: Determine the simulated motor temperature of the UAV in the dynamic evolution model of the fire scene, and obtain the motor operating temperature of the UAV. Step S38: Determine the thermal vulnerability of the UAV based on the predicted motor temperature and the motor operating temperature; Thermal vulnerability characterizes the ability of a drone's power system to resist the heat of a fire environment, and at its core reflects the risk of power performance degradation or failure caused by high temperatures. In this embodiment, a quantitative model is constructed by combining the high-temperature failure threshold with the real-time temperatures of the drone motor and battery.

[0081] Among them, T real T is the real-time temperature of the drone's power motor. c This is the critical temperature for the drone's motor; when this temperature is reached, the motor's performance begins to decline. This temperature is set based on the motor type and historical experimental data. (T) f This refers to the failure temperature of the drone's motor, the extreme temperature at which the power system cannot function properly. It can be determined by referring to the drone manufacturer's technical parameters and high-temperature failure test data.

[0082] V t Thermal vulnerability is a metric, and a higher thermal vulnerability value indicates a weaker ability of the UAV's power system to withstand the heat of the fire environment.

[0083] Step S39: Determine the smoke concentration of the grid in the fire dynamic evolution model and obtain the optical detection parameters of the UAV; Step S310: Determine the sensor vulnerability of the UAV based on the smoke concentration and the optical detection parameters; Sensor vulnerability refers to the ability of optical sensors for drones to maintain perception accuracy in smoke and particulate environments at fire scenes.

[0084] This embodiment constructs an exponential decay model based on particulate matter concentration and optical path length:

[0085] Where β1 is the concentration attenuation coefficient, which can be estimated from optical shielding experimental calibration parameters or historical observation data; ρ is the concentration of particulate matter in the fire smoke, which is obtained in real time by the smoke monitoring equipment; L p is the optical path length of the optical sensor, and is the inherent structural parameter of the device.

[0086] V s This refers to sensor vulnerability. A higher sensor vulnerability value indicates a more severe degradation in sensing accuracy and a poorer ability to maintain sensing accuracy.

[0087] Step S311: Combine the aerodynamic stability vulnerability, thermal vulnerability and sensor vulnerability to obtain the UAV bearing vulnerability factor corresponding to the grid.

[0088] Specifically, environmental disaster factors can be obtained by summing or weighting the aerodynamic stability vulnerability, thermal vulnerability, and the sensor vulnerability.

[0089] In this embodiment, by combining aerodynamic stability vulnerability, thermal vulnerability, and sensor vulnerability, the ability of the UAV system to resist the destructive effects of environmental disaster factors is analyzed from the perspectives of aerodynamic stability, dynamic thermal resistance, and sensor performance, thereby obtaining accurate UAV bearing vulnerability factors.

[0090] Furthermore, in the fifth embodiment of the fire scene drone risk assessment method proposed in the first embodiment of the present invention, the disaster-causing factor includes the drone's exposure factor to the disaster-bearing body, and step S30 includes the following steps: Step S312: Determine the expected arrival time of the UAV at the grid in the fire dynamic evolution model; Step S313: Determine the trajectory correction coefficient based on the historical flight data of the UAV; Step S314: Determine the UAV disaster exposure factor corresponding to the grid based on the expected arrival time and the trajectory correction coefficient.

[0091] The drone-borne disaster exposure factor indicates the degree of interaction between the drone and the environment; it is used for the spatiotemporal matching quantification of the drone flight position and the three-dimensional grid of the fire risk field, characterizing the probability of the drone coming into contact with the risk source within a specific spatiotemporal range, and is a key link connecting the hazard of the disaster-causing factor and the vulnerability of the disaster-borne body.

[0092] The exposure of drone-borne disaster sites can be constructed using nonlinear deep neural networks or linear polynomial functions; for example, it can be quantified using the Logistic function.

[0093] Among them, T arr This is the estimated arrival time; hyperparameter k e The trajectory matching correction coefficient, used to control the slope of the exposure curve, can be further calibrated using test flight data of the fireground drone or historical data. In the sixth embodiment of the fireground drone risk assessment method of the present invention based on the first embodiment, step S40 includes the following steps: Step S41: For each disaster-causing factor, obtain the expert weight and historical data weight corresponding to the disaster-causing factor; Step S42: Combine the expert weights and the historical data weights to obtain the target weights corresponding to the disaster-causing factors; Step S43: Based on the target weight, perform a weighted calculation on each of the disaster-causing factors to obtain the risk index corresponding to the grid. Step S44: Determine the risk level corresponding to the risk index.

[0094] Expert weights are pre-set weights for disaster-causing factors based on the experience of domain experts; for example, weights are obtained by evaluating the importance of environmental disaster-causing factors, UAV-borne vulnerability factors, and UAV-borne vulnerability factors based on expert experience; for example, expert weight W0 is:

[0095] Among them, w H0 Expert weights corresponding to environmental disaster-causing factors; w V0 Expert weights corresponding to the exposure factors of drone-borne disaster-bearing bodies; w E0 The expert weights corresponding to the vulnerability factors of drones are assigned.

[0096] Historical data weights are the weights of disaster-causing factors determined based on historical fire data. For example, based on failure cases of various drone models in different fire scenes, the normalized values ​​and whether a failure occurred in each case are recorded. A logistic regression model is used, with the failure label as the dependent variable and whether a failure occurred as the independent variable. After normalizing the regression coefficients output by the logistic regression model, the historical data weights W1 are obtained.

[0097] Among them, w H1 The historical data weights corresponding to environmental disaster-causing factors; w V1 Historical data weights corresponding to the exposure factors of drone-borne disaster-bearing bodies; w E1 The historical data weights corresponding to the vulnerability factors of drones are assigned.

[0098] After obtaining the expert weights and historical data weights, the two are combined to obtain the target weight:

[0099]

[0100]

[0101] Among them, w H The target weights corresponding to environmental disaster-causing factors; w V The target weights corresponding to the exposure factors of drone-borne disaster-bearing bodies; w E The target weights corresponding to the vulnerability factors of drones are defined; α is the confidence level parameter, which can be set according to the actual need for correction of the expert weights. For example, it can be set to 0.5 to balance expert experience and data patterns.

[0102] After obtaining the target weights of each disaster-causing factor, the final risk index of the grid can be calculated based on a weighted method:

[0103] Wherein, CRI is the risk index; t n Indicates the target time; H norm Environmental disaster-causing factors; V norm To provide vulnerability factors for drones; E norm Exposure factors for drone-borne disaster bodies.

[0104] It is understandable that this application is applied to spatiotemporal simulation; therefore, the disaster-causing factors corresponding to the grid at different times are different, and the disaster-causing factors of grids at different locations are different. In order to eliminate the dimensional differences of different types of disaster-causing factors, the disaster-causing factors can be normalized to unify the value range of [0, 1]. For example, the minimum-maximum normalization method can be used to normalize the hazard index, vulnerability index, and exposure index of disaster-causing factors with different dimensions. Taking the vulnerability factor carried by the UAV as an example:

[0105] Among them, V norm The normalized vulnerability factor for drones; V min V represents the minimum vulnerability factor of the drone across the entire time and space range; max V represents the maximum value of the vulnerability factor of the UAV across the entire time and space range; V is the vulnerability factor of the UAV before normalization.

[0106] In this embodiment, the final target weight is obtained by combining expert weights and historical data weights. This allows the target weight to not only conform to the principle setting of disaster-causing factors, but also to set the importance of disaster-causing factors based on the experience of actual fires.

[0107] Further, step S44 includes the following steps: Step S441: Determine the risk range in which the risk index is located; Step S442: The risk level corresponding to the risk interval is taken as the risk level corresponding to the risk index.

[0108] The risk range is an index range that indicates a specific risk level; the risk range can be set in terms of quantity and value based on actual needs.

[0109] For example, risk levels can be divided into three levels: low risk, medium risk, and high risk. The risk range corresponding to low risk is [0, T1], the risk range corresponding to medium risk is [T1, T2], and the risk range corresponding to high risk is [T2, ∞].

[0110] The risk level of the risk index is determined by the risk range in which it falls, and the risk level of that range is defined as the risk level of the risk index. For example:

[0111] T1 and T2 are risk level classification thresholds, and the number and value of these thresholds can be set based on the characteristics of drone operations and the forest fire environment.

[0112] After determining the risk level of the grid, a dynamic risk map is constructed based on the fire scene model using different rendering techniques. The risk level is identified by different colors, such as L1→green, L2→yellow, and L3→red, forming a spatiotemporal risk map for airspace flight safety in forest fire areas, providing dynamic risk assessment and early warning services for safe drone flight operations.

[0113] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0114] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0115] This application also provides a fireground drone risk assessment device for implementing the above-mentioned fireground drone risk assessment method, the fireground drone risk assessment device comprising: The first acquisition module is used to acquire measured data of the fire scene; The first construction module is used to construct a dynamic evolution model of the fire scene based on the measured fire scene data; The second acquisition module is used to acquire UAV parameters and determine the disaster-causing factors corresponding to the grid in the fire field model based on the UAV parameters and the fire field dynamic evolution model. The fire field model is a model constructed based on the space where the fire field is located, and the grid is a discrete grid that constitutes the fire field model. The first determining module is used to determine the risk level corresponding to each grid based on the disaster-causing factor. The first comprehensive module is used to integrate the risk levels corresponding to each grid to obtain the fire risk assessment result of the UAV.

[0116] This fireground drone risk assessment device constructs a dynamic evolution model of the fireground to dynamically quantify the evolution process of risk indicators in the fireground. This enables the determination of the risk status of specific grids in the fireground at different times, thereby achieving an overall risk assessment of drones in the fireground and providing dynamic risk assessment for the safe flight of drones.

[0117] It should be noted that the first acquisition module in this embodiment can be used to execute step S10 in this application embodiment, the first construction module in this embodiment can be used to execute step S20 in this application embodiment, the second acquisition module in this embodiment can be used to execute step S30 in this application embodiment, the first determination module in this embodiment can be used to execute step S40 in this application embodiment, and the first synthesis module in this embodiment can be used to execute step S50 in this application embodiment.

[0118] Furthermore, the first building module includes: The first acquisition unit is used to acquire wind field data from the measured fire field data and construct a wind field projection model based on the wind field data. The second acquisition unit is used to acquire fire data from the measured fire data and construct a fire simulation model based on the fire data. The third acquisition unit is used to acquire smoke data from the measured fire data and to construct a smoke diffusion simulation model based on the smoke data. The first combining unit is used to combine the wind field simulation model, the fire field simulation model, and the smoke diffusion simulation model to obtain the fire field dynamic evolution model.

[0119] Furthermore, the disaster-causing factors include environmental disaster-causing factors, and the second acquisition module includes: The first determining unit is used to determine the wind turbulence index corresponding to the grid at the target time in the fire dynamic evolution model, and to determine the corresponding turbulence hazard factor based on the wind turbulence index. The second determining unit is used to determine the heat release index of the grid at the target time and the distance between the UAV and the grid in the fire dynamic evolution model, and to determine the corresponding thermal radiation factor based on the heat release index and the distance. The third determining unit is used to determine the particle concentration data of the grid at the target time and the particle resistance of the UAV in the fire dynamic evolution model, and to determine the corresponding smoke factor based on the particle concentration data and the particle resistance. The first integration unit is used to integrate the turbulence hazard factor, the thermal radiation factor, and the smoke factor to obtain the environmental disaster factor corresponding to the grid.

[0120] Furthermore, the disaster-causing factor includes the vulnerability factor of the drone, and the second acquisition module includes: The fourth determining unit is used to determine the dominant turbulence frequency in the grid in the fire dynamic evolution model and to obtain the flight component parameters of the UAV. The fifth determining unit is used to determine the aerodynamic stability vulnerability of the UAV based on the dominant turbulence frequency and the flight component parameters; The sixth determining unit is used to determine the simulated motor temperature of the UAV in the dynamic evolution model of the fire scene, and to obtain the motor operating temperature of the UAV. The seventh determining unit is used to determine the thermal vulnerability of the UAV based on the inferred motor temperature and the motor operating temperature; The eighth determining unit is used to determine the smoke concentration of the grid in the fire dynamic evolution model and to obtain the optical detection parameters of the UAV; The ninth determining unit is used to determine the sensor vulnerability of the UAV based on the smoke concentration and the optical detection parameters; The second integration unit is used to integrate the aerodynamic stability vulnerability, thermal vulnerability, and sensor vulnerability to obtain the UAV bearing vulnerability factor corresponding to the grid.

[0121] Furthermore, the disaster-causing factor includes the drone's exposure factor to the disaster-bearing body, and the second acquisition module includes: The tenth determining unit is used to determine the expected arrival time of the UAV at the grid in the fire dynamic evolution model; The eleventh determining unit is used to determine the trajectory correction coefficient based on the historical flight data of the UAV. The twelfth determining unit is used to determine the UAV disaster exposure factor corresponding to the grid based on the expected arrival time and the trajectory correction coefficient.

[0122] Furthermore, the first determining module includes: The fourth acquisition unit is used to acquire the expert weight and historical data weight corresponding to each disaster-causing factor. The third integration unit is used to integrate the expert weights and the historical data weights to obtain the target weights corresponding to the disaster-causing factors. The first calculation unit is used to calculate the risk index corresponding to the grid by weighting each of the disaster-causing factors based on the target weight; The thirteenth determining unit is used to determine the risk level corresponding to the risk index.

[0123] Furthermore, the thirteenth determining unit includes: The first determining subunit is used for the risk range in which the risk index is located; The first execution subunit is used to take the risk level corresponding to the risk interval as the risk level corresponding to the risk index.

[0124] Reference Figure 4 In terms of hardware structure, the electronic device may include components such as a communication module 10, a memory 20, and a processor 30. In the electronic device, the processor 30 is connected to both the memory 20 and the communication module 10. The memory 20 stores a computer program, which is executed by the processor 30. When the computer program is executed, it implements the steps of the above-described method embodiments.

[0125] The communication module 10 can connect to external communication devices via a network. The communication module 10 can receive requests from the external communication devices and can also send requests, instructions, and information to the external communication devices. The external communication devices can be other electronic devices, servers, or IoT devices, such as televisions, etc.

[0126] The memory 20 can be used to store software programs and various data. The memory 20 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as acquiring measured fire data), etc.; the data storage area may include a database, and may store data or information created based on system usage. Furthermore, the memory 20 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0127] The processor 30 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 20, and by calling data stored in the memory 20, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. The processor 30 may include one or more processing units; optionally, the processor 30 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 30.

[0128] although Figure 4Not shown, but the above-described electronic device may further include a circuit control module for connecting to a power supply to ensure the normal operation of other components. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0129] The present invention also proposes a computer-readable storage medium having a computer program stored thereon. The computer-readable storage medium may be... Figure 4 The memory 20 in the electronic device may also be at least one of ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk, etc. The computer-readable storage medium includes a number of instructions to cause a terminal device with a processor (which may be a television, automobile, mobile phone, computer, server, terminal, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0130] In this invention, the terms "first," "second," "third," "fourth," and "fifth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0131] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0132] Although embodiments of the present invention have been shown and described above, the scope of protection of the present invention is not limited thereto. It is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, and substitutions to the above embodiments within the scope of the present invention, and such changes, modifications, and substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for risk assessment of fire scene using unmanned aerial vehicles (UAVs), characterized in that, The method for assessing the risk of drones operating at fire sites includes: Obtain actual fire scene measurement data; A dynamic evolution model of the fire scene is constructed based on the measured fire scene data. The parameters of the UAV are obtained, and the disaster-causing factors corresponding to the grids in the fire field model are determined based on the UAV parameters and the fire field dynamic evolution model. The fire field model is a model constructed based on the space where the fire field airspace is located. The fire field model is divided into a set of regular and closely adjacent three-dimensional grids. The grids are discrete grids that constitute the fire field model. The disaster-causing factors reflect the factors that pose a risk to the UAV in the fire field. For each grid, the risk level corresponding to the grid is determined based on the disaster-causing factor; The fire risk assessment result of the UAV is obtained by combining the risk levels corresponding to each grid. The disaster-causing factors include the drone-borne disaster exposure factor, and the step of determining the disaster-causing factors corresponding to the grid in the fire field model based on the drone parameters and the fire field dynamic evolution model includes: The expected arrival time of the UAV at the grid is determined in the dynamic evolution model of the fire site; The trajectory correction coefficient is determined based on the historical flight data of the UAV; The exposure factor of the UAV disaster-bearing body corresponding to the grid is determined based on the expected arrival time and the trajectory correction coefficient. The disaster-causing factors include the vulnerability factor of the UAV, and the step of determining the disaster-causing factors corresponding to the grid in the fire field model based on the UAV parameters and the fire field dynamic evolution model includes: In the dynamic evolution model of the fire site, the dominant turbulence frequency in the grid is determined, and the flight component parameters of the UAV are obtained; The aerodynamic stability vulnerability of the UAV is determined based on the dominant turbulence frequency and the parameters of the flight components. The simulated motor temperature of the UAV is determined in the dynamic evolution model of the fire scene, and the operating temperature of the UAV motor is obtained. The thermal vulnerability of the UAV is determined based on the predicted motor temperature and the motor operating temperature. The smoke concentration of the grid is determined in the fire dynamic evolution model, and the optical detection parameters of the UAV are obtained; The sensor vulnerability of the UAV is determined based on the smoke concentration and the optical detection parameters; The vulnerability factor of the UAV bearing corresponding to the grid is obtained by combining the aerodynamic stability vulnerability, thermal vulnerability and sensor vulnerability.

2. The fire scene drone risk assessment method as described in claim 1, characterized in that, The step of constructing a dynamic evolution model of the fire scene based on the measured fire scene data includes: Obtain wind field data from the measured fire field data, and construct a wind field projection model based on the wind field data; Obtain fire data from the measured fire data, and construct a fire simulation model based on the fire data; Obtain smoke data from the measured fire scene data, and construct a smoke diffusion simulation model based on the smoke data; The dynamic evolution model of the fire field is obtained by combining the wind field simulation model, the fire field simulation model, and the smoke diffusion simulation model.

3. The fire scene drone risk assessment method as described in claim 1, characterized in that, The disaster-causing factors include environmental disaster-causing factors, and the step of determining the disaster-causing factors corresponding to the grid in the fire field model based on the UAV parameters and the fire field dynamic evolution model includes: In the fire dynamic evolution model, the wind turbulence index corresponding to the grid at the target time is determined, and the corresponding turbulence hazard factor is determined based on the wind turbulence index; In the fire dynamic evolution model, the heat release index of the grid at the target time and the distance between the UAV and the grid are determined, and the corresponding thermal radiation factor is determined based on the heat release index and the distance. In the fire dynamic evolution model, the particle concentration data of the grid at the target time and the particle resistance of the UAV are determined, and the corresponding smoke factor is determined based on the particle concentration data and the particle resistance. The environmental disaster factor corresponding to the grid is obtained by combining the turbulence hazard factor, the thermal radiation factor, and the smoke factor.

4. The fire scene drone risk assessment method as described in claim 1, characterized in that, The determination of the risk level corresponding to each grid based on the hazard factor includes: For each of the disaster-causing factors, obtain the expert weight and historical data weight corresponding to the disaster-causing factor; The target weight corresponding to the disaster-causing factor is obtained by combining the expert weights and the historical data weights. The risk index corresponding to the grid is obtained by weighting each of the disaster-causing factors based on the target weights. Determine the risk level corresponding to the risk index.

5. The fire scene drone risk assessment method as described in claim 4, characterized in that, Determining the risk level corresponding to the risk index includes: Determine the risk range in which the risk index falls; The risk level corresponding to the risk interval is taken as the risk level corresponding to the risk index.

6. A fire scene drone risk assessment device, characterized in that, The fire scene drone risk assessment device includes: The first acquisition module is used to acquire measured data of the fire scene; The first construction module is used to construct a dynamic evolution model of the fire scene based on the measured fire scene data; The second acquisition module is used to acquire UAV parameters and determine the disaster-causing factors corresponding to the grid in the fire field model based on the UAV parameters and the fire field dynamic evolution model. The fire field model is a model constructed based on the space where the fire field airspace is located, the grid is a discrete grid that constitutes the fire field model, and the disaster-causing factors reflect the factors that pose a risk to the UAV in the fire field. The first determining module is used to determine the risk level corresponding to each grid based on the disaster-causing factor. The first comprehensive module is used to integrate the risk levels corresponding to each grid to obtain the fire risk assessment result of the UAV; The disaster-causing factors include the drone's exposure factor to the disaster-bearing body, and the second acquisition module includes: The tenth determining unit is used to determine the expected arrival time of the UAV at the grid in the fire dynamic evolution model; The eleventh determining unit is used to determine the trajectory correction coefficient based on the historical flight data of the UAV. The twelfth determining unit is used to determine the UAV disaster-bearing body exposure factor corresponding to the grid based on the expected arrival time and the trajectory correction coefficient; The disaster-causing factors include the vulnerability factors of the UAV, and the second acquisition module includes: The fourth determining unit is used to determine the dominant turbulence frequency in the grid in the fire dynamic evolution model and to obtain the flight component parameters of the UAV. The fifth determining unit is used to determine the aerodynamic stability vulnerability of the UAV based on the dominant turbulence frequency and the flight component parameters; The sixth determining unit is used to determine the simulated motor temperature of the UAV in the dynamic evolution model of the fire scene, and to obtain the motor operating temperature of the UAV. The seventh determining unit is used to determine the thermal vulnerability of the UAV based on the inferred motor temperature and the motor operating temperature; The eighth determining unit is used to determine the smoke concentration of the grid in the fire dynamic evolution model and to obtain the optical detection parameters of the UAV; The ninth determining unit is used to determine the sensor vulnerability of the UAV based on the smoke concentration and the optical detection parameters; The second integration unit is used to integrate the aerodynamic stability vulnerability, thermal vulnerability, and sensor vulnerability to obtain the UAV bearing vulnerability factor corresponding to the grid.

7. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the fireground drone risk assessment method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the fireground drone risk assessment method as described in any one of claims 1 to 5.

Images