Artificial Intelligence-Based Forest Fire Risk Status Monitoring and Analysis System and Method

Abstract

The present invention discloses a forest fire risk status monitoring and analysis system and method based on artificial intelligence, belonging to the technical field of artificial intelligence. The present invention constructs a forest wind prediction model to predict the wind speed and wind direction of the forest wind through the forest wind prediction model. The forest is divided into checkerboards according to the detection range of the forest fire monitoring equipment, and the content of forest combustion-supporting substances in a single checkerboard is calculated according to the vegetation distribution of the forest. A forest fire spread speed calculation model is established to simulate different checkerboards as the ignition points and the process of fire spread, so as to find the key monitoring areas of forest fires; the wind force data of the forest are predicted through a time series prediction algorithm to ensure the specific reference value of the fire spread simulation data; a forest fire spread speed calculation model is established to further realize the simulation of forest fires, find the key points of forest fire detection through forest fire simulation, save manpower and material resources, and effectively investigate the key areas of fire danger.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and specifically to a forest fire risk status monitoring and analysis system and method based on artificial intelligence. Background Art

[0002] Forests are the lungs of the earth. A forest is an ecosystem composed of plant communities, animals living in the forest, and the abiotic environment in the space it occupies. The life cycle and succession series of forests are long, and forests have high productivity, multiple uses, and great benefits. However, forest fires without human control can severely damage the trees accumulated in the forest over the years and the forest environment, leading to the imbalance of the forest ecosystem, damaging the performance of the forest soil, and changing the landform. The damage of forest fires to the ecosystem is the most direct and the most disastrous; a large number of plants are burned, which will cause adverse effects such as soil erosion and aggravated land desertification, and ultimately may even be unable to maintain the ecological balance.

[0003] There are many factors in the spread of forest fires, which are closely related to weather, terrain, and forest fuel types. When forest fires occur in different regions of the forest, the time it takes for the rescue team to arrive after discovery and carry out the rescue is also different. That is to say, once a forest fire occurs, the difficulty of the rescue team to carry out the rescue is also different. However, the existing fire monitoring is all-round coverage. Due to the large forest area and the shelter of trees, it is very difficult to achieve full-area coverage, which requires a large amount of manpower and material resources. How to find the key points of forest fire monitoring in different environments is an urgent problem for us to solve. Summary of the Invention

[0004] The purpose of the present invention is to provide a forest fire risk status monitoring and analysis system and method based on artificial intelligence to solve the problems raised in the above background art.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: A forest fire risk status monitoring and analysis method based on artificial intelligence. The specific steps of the fire risk status monitoring and analysis method include:

[0006] Step 1: Collect forest wind force data, where the forest wind force data includes the wind direction of the forest wind, the wind speed of the forest wind, and the atmospheric pressure of the forest.

[0007] Step 2: Normalize the atmospheric pressure of the forest and output the normalized value of the atmospheric pressure of the forest.

[0008] Step 3: Build a forest wind prediction model based on LSTM, input the wind direction of the forest wind, the wind speed of the forest wind, and the normalized value of the atmospheric pressure of the forest, train the model, and output the forest wind prediction model with the highest mAP value. Predict the wind speed and wind direction of the forest wind through the forest wind prediction model.

[0009] When a forest fire occurs, it is greatly affected by the current wind speed and direction. Especially in the initial stage of the fire, the influence of wind speed and direction is the greatest. The fire spreads with the help of the wind, and the wind dominates the spreading direction of the fire.

[0010] China is located in the southeast of the Eurasian continent. Due to the different physical properties of the sea and land, obvious seasonal climate changes are formed. Throughout the year, the overall monsoon is regular, the wind direction is fixed, and in the case of strong wind directions, there will be specific changes in the local wind situation.

[0011] When studying forest fires, we need more specific wind directions. When the forest fire is not completely out of control, the wind direction dominates the fire. By using the time prediction algorithm to predict the possible wind direction, it can ensure that the further simulated data of fire spread has specific reference value.

[0012] Step 4: Collect forest data, where the forest data includes forest maps, forest slopes, forest vegetation distributions, forest rainfall data, and forest evaporation amounts.

[0013] Step 5: Divide the forest into checkerboards according to the detection range of forest fire monitoring equipment.

[0014] Step 6: Calculate the content of combustion-supporting substances in a single checkerboard forest according to the vegetation distribution of the forest.

[0015] Step 7: Establish a calculation model for the forest fire spread speed.

[0016] Step 8: Simulate different checkerboards as the ignition points, simulate the process of fire spread, and count the number of checkerboards burned before the rescue team arrives.

[0017] Step 9: Set a threshold for the number of burned checkerboards. When the number of burned checkerboards exceeds the threshold, it is determined that the current ignition point is a key monitoring area.

[0018] The wind direction of the forest will change continuously with time. Through the forest wind prediction algorithm, key monitoring areas can be continuously output. Once a fire occurs in a key monitoring area, it is an area where it is difficult for the rescue team to carry out rescues, and it is very likely to further develop into a major fire, with serious impacts.

[0019] In Step 2, the forest atmospheric pressure is normalized, further strengthening the characteristics of the forest atmospheric pressure, and effectively improving the accuracy of the forest wind prediction model. The specific calculation formula for outputting the normalized value of the forest atmospheric pressure is: Normalize the forest atmospheric pressure

[0020]

[0021] Among them, represents the value of the forest atmospheric pressure, represents the mean value of the forest atmospheric pressure, Represents the normalized atmospheric pressure value of the forest.

[0022] The specific content of constructing the forest wind prediction model based on LSTM in Step 3 includes: an attention mechanism is set on the LSTM network layer of the forest wind prediction model, and the output layer is connected to the fully connected layer.

[0023] The specific steps of calculating the content of the forest combustion-supporting substances in a single checkerboard grid according to the vegetation distribution of the forest in Step 6 include:

[0024] Based on the water content of the plants themselves, calculate the precipitation and evaporation of the forest, effectively predicting the water content of the vegetation in a short time; in case of a sudden fire, through the effective calculation of the predicted water content of the vegetation in a short time, accurately judge the content of the vegetation combustion-supporting substances;

[0025] Step 1.1: Collect forest rainfall data and forest evaporation, where the forest rainfall data includes rainfall amount and rainfall interval time;

[0026] Step 1.2: Calculate the oil content of the trees within the checkerboard grid ;

[0027] Step 1.3: Calculate the water content of the plants within the checkerboard grid, and the specific calculation formula is:

[0028]

[0029] Among them, represents the water content of the plants within the checkerboard grid, represents the forest precipitation, represents the forest evaporation, represents the water content of the plants within the checkerboard grid;

[0030] Step 1.4: Calculate the content of the combustion-supporting substances within the checkerboard grid, and the specific calculation formula is:

[0031]

[0032] Among them, represents the content of the combustion-supporting substances within the checkerboard grid, represents the oil content of the trees within the checkerboard grid, represents the hardness of the plants within the checkerboard grid.

[0033] Forest fires do not destroy everything. Some plants have chosen to adapt to forest fires. Plants reduce their oil content by increasing their water content. When a forest fire occurs, they use the evaporation of the water in their bodies to carry away a large amount of heat, reduce the temperature, and preserve their vitality. By calculating the water content and oil content of the plant body, and the characteristics of the plant itself regarding combustion, it is possible to effectively reflect the effects of the objects within the checkerboard on a fire when a fire occurs within the checkerboard, and visually display it through the quantification of the content of the combustion-supporting substances within the checkerboard.

[0034] Plants in the forest rely on sunlight, soil, altitude, monsoon climate, etc. Therefore, the distribution of plants in the forest is fixed, and the life forms dependent on the forest growth are relatively fixed as well. At the same time, the life forms dependent on plant growth show consistency with the characteristics of the plants through the enrichment phenomenon. Therefore, the plants within the checkerboard can represent the distribution of the combustion-supporting substances within the entire checkerboard. By calculating the oil content and water content of the plants in the forest, the content of the combustion-supporting substances within the checkerboard is output.

[0035] The content of the combustion-supporting substances within the checkerboard is directly proportional to the likelihood of a forest fire occurring and the spread speed of the forest fire. The content of the combustion-supporting substances within the checkerboard.

[0036] The specific content of Step 7 for establishing the forest fire spread speed calculation model includes:

[0037]

[0038] Among them, represents the fire spread rate, represents the initial fire source spread speed, represents the slope, represents the slope angle, represents the wind direction, represents the wind speed, represents the slope correction factor, represents the forest fire spread coefficient, represents the granular density of the combustible distribution, represents the effective heat number, represents the ignition heat.

[0039] Both the wind and the slope can effectively fuel the fire. When the wind direction and the slope angle are opposite, the slope will not fuel the fire.

[0040] The specific steps of Step 8 for simulating different checkerboards as the ignition points, simulating the process of the fire spreading, and counting the number of checkerboards burned and the number of checkerboards burned before the rescue team arrives include:

[0041] Step 2.1: Set the initial ignition point, and the forest fire spread speed calculation model calculates the time for the checkerboard to be completely burned;

[0042] Step 2.2: Collect the current address and traveling speed of the rescue team;

[0043] Step 2.3: Calculate the time and location when the rescue team meets the forest fire;

[0044] Step 2.4: Count the number of burned checkerboards and the number of burned checkerboards before the rescue team arrives.

[0045] The determination method for complete combustion of the checkerboard is: set an area threshold and an edge threshold, where the area threshold and the edge threshold are greater than or equal to 50%, the burned area of the checkerboard is greater than the area threshold, and the burned edge of the checkerboard is greater than the edge threshold.

[0046] Forest fire risk status monitoring and analysis system based on artificial intelligence. The fire risk monitoring and analysis system includes: a forest data collection module, a forest wind prediction module, a forest fire simulation model, and a data display module;

[0047] After the forest data collection module collects and structurally stores the forest data, it transmits it to the forest wind prediction module and the forest fire simulation model;

[0048] The forest data collection module includes a forest wind data collection module and a forest information collection module. The forest wind data collection module collects the wind direction of the forest wind, the wind speed of the forest wind, and the atmospheric pressure of the forest, normalizes the atmospheric pressure of the forest, and transmits it to the forest wind prediction module;

[0049] The forest information collection module collects the forest map, forest slope, forest vegetation distribution, forest rainfall data, and forest evaporation amount, and after structurally storing them, transmits them to the forest fire simulation module;

[0050] The forest wind prediction module constructs a forest wind prediction model based on the time series prediction algorithm, sets a training set by collecting forest wind force data through the forest data collection module, and outputs the model with the highest mAP value as the forest wind prediction model. The predicted wind direction value and the predicted wind direction value of the forest wind are output through the forest wind prediction model;

[0051] The forest fire simulation module simulates forest fires occurring at different locations in the forest as the ignition points, counts the burned areas of forest fires with different locations as the ignition points before the rescue team arrives, and outputs them to the data display module.

[0052] The data display module sets an area combustion threshold, and when the burned area exceeds the area combustion threshold, it is displayed as a key monitoring area on the user terminal.

[0053] The key monitoring area is an area where, once a fire breaks out, there is sufficient time for it to develop fully before the rescue team arrives, and it is very easy to further develop into a major fire. Through the display of the key monitoring area, users can easily find the key points of monitoring and further effectively investigate the key areas of fire hazards.

[0054] The forest fire simulation module includes a forest combustion-supporting substance content model and a forest fire spread speed calculation model;

[0055] The forest combustion-supporting substance content model calculates the distribution of forest nutrients through plant distribution and outputs the content of forest combustion-supporting substances in the checkerboard grid to the forest fire spread speed calculation model;

[0056] The forest fire spread speed calculation model calculates the forest fire spread speed based on the predicted value of the wind direction of the forest wind, the predicted value of the wind direction of the forest wind, and the forest terrain.

[0057] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: The present invention predicts the wind force data of the forest through the time series prediction algorithm to ensure that the fire spread simulation data has specific reference value; the present invention calculates the distribution of forest combustion-supporting substances according to the vegetation distribution of the forest, further establishes a forest fire spread speed calculation model, further realizes the simulation of forest fires, finds the key points of forest fire detection through forest fire simulation, saves manpower and material resources, and effectively investigates the key areas of fire hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0059] Figure 1 is a schematic structural diagram of the forest fire risk status monitoring and analysis system based on artificial intelligence of the present invention; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0061] Please refer to Figure 1 , the present invention provides the following technical solutions:

[0062] Embodiment 1: A forest fire risk status monitoring and analysis method based on artificial intelligence. The specific steps of the fire risk status monitoring and analysis method include:

[0063] Step 1: Collect forest wind data, which includes the wind direction of the forest wind, the wind speed of the forest wind, and the atmospheric pressure of the forest;

[0064] Step 2: Normalize the atmospheric pressure of the forest and output the normalized value of the atmospheric pressure of the forest;

[0065] Step 3: Build a forest wind prediction model based on LSTM. Input the wind direction of the forest wind, the wind speed of the forest wind, and the normalized value of the atmospheric pressure of the forest, train the model, and output the forest wind prediction model with the highest mAP value. Predict the wind speed and wind direction of the forest wind through the forest wind prediction model;

[0066] When a forest fire occurs, it is greatly affected by the current wind speed and wind direction. Especially in the initial stage of the fire, the wind speed and wind direction have the greatest impact. The fire spreads with the help of the wind, and the wind dominates the spreading direction of the fire;

[0067] China is located in the southeast of the Eurasian continent. Due to the different physical properties of the sea and land, obvious seasonal climate changes are formed. Throughout the year, the overall monsoon is regular and the wind direction is fixed. In the case of strong wind directions, there will be specific changes in the local wind force;

[0068] When studying forest fires, we need more specific wind directions. When the forest fire is not completely out of control, the wind direction dominates the fire. Predict the possible wind direction through the time prediction algorithm to ensure the specific reference value of the further forest fire spreading simulation data.

[0069] Step 4: Collect forest data, which includes forest maps, forest slopes, forest vegetation distributions, forest rainfall data, and forest evaporation amounts;

[0070] Step 5: Divide the forest into checkerboards according to the detection range of the forest fire monitoring equipment;

[0071] Step 6: Calculate the content of forest combustion-supporting substances in a single checkerboard according to the vegetation distribution of the forest;

[0072] Step 7: Establish a calculation model for the forest fire spreading speed;

[0073] Step 8: Simulate different checkerboards as the ignition points, simulate the process of the fire spreading, and count the number of checkerboards burned before the rescue team arrives;

[0074] Step 9: Set a threshold for the number of burned checkerboards. When the number of burned checkerboards exceeds the threshold for the number of burned checkerboards, determine that the current ignition point is a key monitoring area.

[0075] The wind direction in the forest changes continuously over time. Through the forest wind prediction algorithm, the key monitoring areas can be continuously output. Once a fire breaks out in the key monitoring area, it is difficult for the rescue team to reach the area, and it is very likely to further develop into a major fire, resulting in serious impacts.

[0076] Step 2: Normalize the forest atmospheric pressure to further enhance the characteristics of the forest atmospheric pressure and effectively improve the accuracy of the forest wind prediction model. The specific calculation formula for outputting the normalized value of the forest atmospheric pressure is: Normalize the forest atmospheric pressure

[0077]

[0078] where, represents the value of the forest atmospheric pressure, represents the mean value of the forest atmospheric pressure, represents the normalized value of the forest atmospheric pressure.

[0079] Step 3: Build a forest wind prediction model based on LSTM. The specific content includes: setting an attention mechanism on the LSTM network layer of the forest wind prediction model, and connecting the output layer to the fully connected layer.

[0080] Step 6: Calculate the content of the forest combustion-supporting substances in a single checkerboard grid according to the vegetation distribution in the forest. The specific steps include: based on the water content of the plants themselves, calculate the precipitation and evaporation in the forest to effectively predict the water content of the vegetation in a short period of time; in case of a sudden fire, accurately calculate through the prediction of the water content of the vegetation in a short period of time to accurately judge the content of the vegetation combustion-supporting substances;

[0081] Step 1.1: Collect forest rainfall data and forest evaporation. The forest rainfall data includes rainfall and rainfall interval time;

[0082] Step 1.2: Calculate the oil content of the trees in the checkerboard grid ;

[0083] Step 1.3: Calculate the water content of the plants in the checkerboard grid. The specific calculation formula is:

[0084]

[0085] where, represents the water content of the plants in the checkerboard grid, represents the forest precipitation, represents the forest evaporation, represents the water content of the plants in the checkerboard grid;

[0086] Step 1.4: Calculate the content of the combustion-supporting substances in the checkerboard grid. The specific calculation formula is:

[0087]

[0088] Among them, represents the content of the combustion-supporting substance in the checkerboard grid, represents the oil content of the trees in the checkerboard grid, represents the hardness of the plants in the checkerboard grid.

[0089] Forest fires do not destroy everything. Some plants choose to adapt to forest fires. Plants reduce the oil rate by increasing the water content. When encountering a forest fire, they use the evaporation of the water in their bodies to take away a large amount of heat, reduce the temperature, and preserve their own vitality. By calculating the water content and oil content of the plant body, and the characteristics of the plant itself for combustion, it can effectively reflect the effect of the objects in the checkerboard grid on the fire when a fire occurs in the checkerboard grid, and is visually displayed through the quantification of the content data of the combustion-supporting substance in the checkerboard grid.

[0090] The plants in the forest rely on sunlight, soil, altitude, monsoon climate, etc. Therefore, the distribution of plants in the forest is fixed, and the life depending on the forest growth is relatively fixed. At the same time, the life depending on the plant growth shows consistency with the characteristics of the plant through the enrichment phenomenon. Therefore, the plants in the checkerboard grid can represent the distribution of the combustion-supporting substance in the entire checkerboard grid. By calculating the oil content and water content of the plants in the forest, the content of the combustion-supporting substance in the checkerboard grid is output.

[0091] The content of the combustion-supporting substance in the checkerboard grid is directly proportional to the possibility of a forest fire occurring and the spreading speed of the forest fire. The content of the combustion-supporting substance in the checkerboard grid.

[0092] Step 7: Establish a calculation model for the spreading speed of forest fires. The specific content includes:

[0093]

[0094] Among them, represents the fire spreading rate, represents the spreading speed of the initial fire source, represents the slope, represents the slope angle, represents the wind direction, represents the wind speed, represents the slope correction factor, represents the forest fire spreading coefficient, represents the granular density of the combustible distribution, represents the effective heat number, represents the ignition heat.

[0095] Both the wind and the slope can effectively fuel the fire. When the wind direction and the slope angle are opposite, the slope will not fuel the fire.

[0096] Step 8: Simulate different checkerboards as the ignition points, simulate the process of fire spread, and count the number of burned checkerboards and the quantity of burned checkerboards before the rescue team arrives. The specific steps are as follows:

[0097] Step 2.1: Set the initial ignition point, and the forest fire spread speed calculation model calculates the time for the checkerboard to burn completely.

[0098] Step 2.2: Collect the current location and traveling speed of the rescue team.

[0099] Step 2.3: Calculate the time and location where the rescue team meets the forest fire.

[0100] Step 2.4: Count the number of burned checkerboards and the quantity of burned checkerboards before the rescue team arrives.

[0101] The determination method for the checkerboard to burn completely is as follows: Set the area threshold and the edge threshold. The area threshold and the edge threshold are greater than or equal to 50%. The burned area of the checkerboard is greater than the area threshold, and the burned edge of the checkerboard is greater than the edge threshold.

[0102] The forest fire risk status monitoring and analysis system based on artificial intelligence. The fire risk monitoring and analysis system includes: a forest data collection module, a forest wind prediction module, a forest fire simulation model, and a data display module.

[0103] After the forest data collection module collects and structurally stores the forest data, it transmits it to the forest wind prediction module and the forest fire simulation model.

[0104] The forest data collection module includes a forest wind data collection module and a forest information collection module. The forest wind data collection module collects the wind direction of the forest wind, the wind speed of the forest wind, and the atmospheric pressure of the forest, normalizes the atmospheric pressure of the forest, and transmits it to the forest wind prediction module.

[0105] The forest information collection module collects the forest map, forest slope, forest vegetation distribution, forest rainfall data, and forest evaporation amount, and after structurally storing them, transmits them to the forest fire simulation module.

[0106] The forest wind prediction module constructs a forest wind prediction model based on the time series prediction algorithm, sets the training set by collecting the forest wind force data through the forest data collection module, and outputs the model with the highest mAP value as the forest wind prediction model. Through the forest wind prediction model, the predicted value of the wind direction of the forest wind and the predicted value of the wind direction of the forest wind are output.

[0107] The forest fire simulation module simulates forest fires occurring at different locations in the forest as ignition points, counts the burned area of the forest fires at different locations as ignition points before the rescue team arrives, and outputs it to the data display module.

[0108] The data display module sets an area combustion threshold. When the burned area exceeds the area combustion threshold, the key monitoring area is displayed on the user terminal.

[0109] Once a fire breaks out in the key monitoring area, there is sufficient development time before the rescue team arrives, and it is very easy to further develop into a major fire. By displaying the key monitoring area, users can easily find the key points of monitoring and further effectively investigate the key areas of fire danger.

[0110] The forest fire simulation module includes a forest combustion-supporting substance content model and a forest fire spread speed calculation model;

[0111] The forest combustion-supporting substance content model calculates the distribution of forest nutrients through plant distribution and outputs the content of forest combustion-supporting substances in the checkerboard grid to the forest fire spread speed calculation model;

[0112] The forest fire spread speed calculation model calculates the forest fire spread speed based on the predicted wind direction value of the forest wind, the predicted wind direction value of the forest wind, and the forest terrain.

[0113] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0114] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An artificial intelligence-based method for monitoring and analyzing forest fire risk status, Characterized in that: The specific steps of the fire risk status monitoring and analysis method include: Step 1: Collect forest wind data, where the forest wind data includes the wind direction of the forest wind, the wind speed of the forest wind, and the atmospheric pressure of the forest; Step 2: Normalize the forest atmospheric pressure and output the normalized value of the forest atmospheric pressure; Step 3: Build a forest wind prediction model based on LSTM, input the wind direction of the forest wind, the wind speed of the forest wind, and the normalized value of the forest atmospheric pressure, train the model, and output the forest wind prediction model with the highest mAP value. Predict the wind speed and wind direction of the forest wind through the forest wind prediction model; Step 4: Collect forest data, where the forest data includes forest maps, forest slopes, forest vegetation distributions, forest rainfall data, and forest evaporation amounts; Step 5: Divide the forest into checkerboards according to the detection range of forest fire monitoring equipment; Step 6: Calculate the content of forest combustion-supporting substances in a single checkerboard according to the vegetation distribution of the forest. The specific steps are as follows: Step 1.1: Collect forest rainfall data and forest evaporation amounts. The forest rainfall data includes rainfall and rainfall interval time, Step 1.2: Calculate the grease content of the trees within the checkerboard , Step 1.3: Calculate the water content of plants in the checkerboard, Step 1.4: Calculate the content of combustion-supporting substances in the checkerboard; Step 7: Establish a calculation model for the forest fire spread speed; Step 8: Simulate different checkerboards as the ignition points, simulate the process of fire spread, and count the number of checkerboards burned before the rescue team arrives. The specific steps are as follows: Step 2.1: Set the initial ignition point, and the forest fire spread speed calculation model calculates the time for the checkerboard to burn completely, Step 2.2: Collect the current address and traveling speed of the rescue team, Step 2.3: Calculate the time and location where the rescue team meets the forest fire, Step 2.4: Count the number of checkerboards burned and the number of burned checkerboards before the rescue team arrives; Step 9: Set the threshold for the number of burned checkerboards. When the number of burned checkerboards exceeds the threshold for the number of burned checkerboards, it is determined that the current ignition point is a key monitoring area.

2. The artificial intelligence-based method for monitoring and analyzing forest fire risk status according to claim 1, Characterized in that: The specific calculation formula for normalizing the forest atmospheric pressure in Step 2 and outputting the normalized value of the forest atmospheric pressure is: Normalize the atmospheric pressure of the forest Among them, represents the value of the atmospheric pressure in the forest, represents the mean value of the atmospheric pressure in the forest, represents the normalized value of the atmospheric pressure in the forest.

3. The artificial intelligence-based method for monitoring and analyzing forest fire risk status according to claim 1, Characterized in that: The specific content of building a forest wind prediction model based on LSTM in Step 3 includes: The forest wind prediction model sets an attention mechanism on the LSTM network layer, and the output layer is connected to the fully connected layer.

4. The artificial intelligence-based method for monitoring and analyzing forest fire risk status according to claim 1, Characterized in that: In Step 6: Step 1.3: The specific calculation formula for calculating the water content of plants in the checkerboard is: Among them, represents the water content of plants within the checkerboard grid, represents the forest precipitation, represents the forest evaporation, represents the water content of plants within the checkerboard grid; Step 1.4: The specific calculation formula for calculating the content of combustion-supporting substances in the checkerboard is: Among them, represents the content of the combustion-supporting substance in the checkerboard grid, represents the grease content of the trees in the checkerboard grid, represents the hardness of the plants in the checkerboard grid.

5. The artificial intelligence-based method for monitoring and analyzing forest fire risk status according to claim 1, Characterized in that: The determination method for the complete combustion of the checkerboard is as follows: Set an area threshold and an edge threshold, where the area threshold and the edge threshold are greater than or equal to 50%. The burned area of the checkerboard is greater than the area threshold, and the burned edge of the checkerboard is greater than the edge threshold.

6. An artificial intelligence-based forest fire risk status monitoring and analysis system, characterized in that: The fire risk monitoring and analysis system includes: a forest data collection module, a forest wind prediction module, a forest fire simulation model, and a data display module; After the forest data collection module collects and structurally stores forest data, it transmits the data to the forest wind prediction module and the forest fire simulation model; The forest data collection module includes a forest wind data collection module and a forest information collection module. The forest wind data collection module collects the wind direction of the forest wind, the wind speed of the forest wind, and the atmospheric pressure of the forest, normalizes the atmospheric pressure of the forest, and transmits it to the forest wind prediction module; The forest information collection module collects forest maps, forest slopes, forest vegetation distributions, forest rainfall data, and forest evaporation amounts, and after structural storage, transmits them to the forest fire simulation module; The forest wind prediction module constructs a forest wind prediction model based on a time series prediction algorithm, sets a training set through the forest data collection module to collect forest wind force data, and outputs the model with the highest mAP value as the forest wind prediction model. The wind direction prediction value of the forest wind and the wind direction prediction value of the forest wind are output through the forest wind prediction model; The forest fire simulation module simulates forest fires occurring at different locations in the forest as the ignition points, and counts the burned areas of forest fires at different locations as the ignition points before the rescue team arrives and outputs them to the data display module. The data display module sets an area burning threshold, and when the burned area exceeds the area burning threshold, it is displayed as a key monitoring area on the user terminal.

7. The artificial intelligence-based forest fire risk status monitoring and analysis system according to claim 6, characterized in that: The forest fire simulation module includes a forest combustion aid substance content model and a forest fire spread speed calculation model; The forest combustion aid substance content model calculates the distribution of forest nutrient substances through plant distribution, and outputs the content of forest combustion aid substances in the checkerboard to the forest fire spread speed calculation model; The forest fire spread speed calculation model calculates the forest fire spread speed based on the wind direction prediction value of the forest wind, the wind direction prediction value of the forest wind, and the forest terrain.

Images