A forest fire prevention multi-layer early warning monitoring system and method using big data technology
Through the multi-layer early warning and monitoring system of big data technology, combined with geographical areas, weather conditions and aerial photography monitoring, the problems of equipment prone to failure and improper management in the existing technology have been solved, accurate monitoring and timely warning of forest fires have been achieved, and rescue efforts have been improved. Efficiency and equipment management efficiency.
Patent Information
- Application Number
- CN202310192247.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-03-02
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2043-03-02
AI Technical Summary
Existing forest fire monitoring equipment is prone to failure and improper management, resulting in missing the best window for fire extinguishing, and the equipment is arranged casually, so that each equipment cannot be accurately monitored, resulting in insufficient fire monitoring range.
Using big data technology to develop multi-layer early warning and monitoring systems, including geographic area monitoring modules, weather conditions warning modules and aerial photography monitoring modules. Through positioning systems, infrared lenses, temperature and humidity detectors and aircraft equipment, the accuracy of flammable areas and fires is achieved. Monitor and early warning, optimize equipment layout and management.
It improves the accuracy and timeliness of fire monitoring, ensures that fire extinguishers can intervene at the best time, reduces fire spread and casualties, optimizes equipment management, and extends the service life of the equipment.
Smart Images

Figure CN116189372B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of forest fire prevention, and specifically to a forest fire prevention multi-layer early warning monitoring system and method using big data technology. Background Art
[0002] The Internet of Things technology is based on the rapid development of communication technology and hardware updates. It interconnects hardware through the network to achieve network access to devices, so that hardware devices can be controlled and monitored through the network and receive instructions from the remote control terminal. As terminals, they form an interconnected Internet of Things network through different topological structures, and unified management of hardware devices is performed based on the Internet.
[0003] my country's forest resources are very rich, divided into protective forests, timber forests, firewood forests and special-purpose forests; due to differences in latitude and geographical environment, forest fires occur frequently in some forest areas. Once forest fires form a scale, they are usually difficult to contain, and the scale of burning is large and the duration is long, causing a large amount of natural resources and economic property losses; fighting forest fires requires a lot of human resources, and fighting is difficult, which can easily cause casualties to rescuers. In order to prevent the occurrence and spread of forest fires, they should be discovered and warned in time when forest fires begin to burn. Existing alarm facilities monitor whether there is burning in the forest by detecting smoke and temperature in the forest, and quickly issue an alarm; existing facilities can detect the occurrence of forest fires, but the equipment is prone to failure due to long-term outdoor placement, and the management of the equipment is not fine enough, and each device cannot be accurately maintained and monitored, resulting in equipment failure that cannot be detected. There are also problems such as random arrangement, and the alarm monitoring equipment is far from the fire point and exceeds the monitoring range, resulting in the best window period for fire extinguishing.
[0004] A forest fire prevention multi-layer early warning monitoring system and method using big data technology is proposed to solve the above problems. Summary of the invention
[0005] The purpose of the present invention is to provide a forest fire prevention multi-layer early warning monitoring system and method using big data technology to solve the problems raised in the above background issues.
[0006] To solve the above problems, the present invention provides the following technical solutions: a multi-layer early warning monitoring system and method for forest fire prevention using big data technology, characterized in that: the early warning monitoring system includes a three-layer early warning monitoring mechanism, a forest area monitoring module based on geographical areas, an environmental early warning module based on weather conditions, and an aerial photography monitoring module based on aviation equipment;
[0007] The forest monitoring module places alarm devices based on the current geographical location and forest characteristics to monitor and check for flammable objects in areas with frequent human activities.
[0008] The equipment focuses on monitoring flammable areas in the forest, focusing on monitoring flammable trees, capturing fire-prone points in the forest, and promptly discovering abnormally high temperatures and fire points;
[0009] The environmental early warning module issues early warnings for high temperatures and persistent droughts based on weather forecasts;
[0010] It enables forest rangers to arrange patrol plans in advance based on weather forecast information and check for potential fire spots;
[0011] The aerial monitoring module performs fire monitoring and fire spread analysis based on images taken by aviation equipment;
[0012] It enables firefighters to safely and effectively build isolated firebreaks to prevent the fire from spreading further;
[0013] According to the above technical solution, the forest monitoring module includes a regional monitoring unit, an image monitoring unit and a device management unit;
[0014] The regional monitoring unit uses a positioning system to collect the geographical location of the current forest area, reads the number, type and quantity of forests in the corresponding forest area from the database, and arranges monitoring and alarm devices accordingly to improve the pertinence of monitoring alarms; the image monitoring unit uses an infrared lens to identify flammable points and high-temperature locations to detect combustion hazards; the device management unit uniformly manages the alarm devices in the field and optimizes the layout of the devices.
[0015] According to the above technical solution, the environmental warning module includes a temperature warning unit, a drought warning unit and a wind warning unit;
[0016] The temperature warning unit monitors the temperature conditions in the forest through the temperature detection instruments in the forest farm, issues alarms for continuous high temperature conditions and extreme high temperature conditions, collects weather information through an interface, and adjusts the risk level of fire warning by predicting future temperature conditions; the drought warning unit detects the humidity conditions in the forest farm through the humidity detector in the forest farm, collects precipitation information, and predicts the duration of drought; the wind warning unit is turned on in drought and high temperature conditions, monitors the wind speed and wind direction of the day, and determines the impact of wind on the area.
[0017] According to the above technical solution, the aerial photography monitoring module includes an information acquisition unit, an image analysis unit and an intelligent pre-judgment unit;
[0018] The information collection unit is connected to the entry points of different satellites and aerospace equipment to obtain open remote sensing image resources in real time and download them to the local system; the image analysis unit retrieves images of the fire scene, analyzes the images, and determines the burning range and spreading direction of the fire; the intelligent prediction unit determines the burning time of the fire and the impact range of the fire smoke based on the image analysis results, and provides a plan direction for the fire fighting plan.
[0019] A forest fire prevention multi-layer early warning monitoring method using big data technology includes the following steps:
[0020] S1. Collect regional geographic and forest information, arrange monitoring and alarm devices and manage them in a unified manner;
[0021] S2. Patrol and identify flammable spots and high-temperature areas in the forest farm to check for burning hazards;
[0022] S3. Capture temperature and humidity data in the forest farm, and combine it with meteorological conditions to determine the fire warning level;
[0023] S4. Monitor the fire scene, analyze fire information through thermal imaging, determine the direction of fire spread, and predict the duration of continuous burning and the final scope of fire impact.
[0024] In step S1, the current forest farm is located, and according to the type and size of the forest farm, the size of the loss caused by the forest farm fire is calculated, the priority of early warning monitoring is arranged, and the detection device and the alarm device are arranged according to the characteristics and distribution positions of the trees in the forest farm; according to the classification of forest types, they are generally divided into protective forests, timber forests, firewood forests and special-purpose forests, among which timber forests and special-purpose forests have high economic and scientific research value, and firewood forests are flammable. High-value timber forests and special-purpose forests and flammable firewood forests are monitored in key areas; the device management unit monitors and manages all alarm monitoring devices in the forest farm and dispatches them in a unified manner; according to the frequency of human activities in the forest farm, the devices are arranged, the target points of the activity area in the forest farm are calibrated, the biological activity path is predicted and planned, the monitoring alarm devices are arranged along the path, and the artificial potential field algorithm is used to establish the path. The specific steps are as follows:
[0025] S101: Constructing the gravitational field in the potential field, using the parabolic gravitational field model ; is the gravitational gain, that is, the gain is the attraction of the target to the organism, expressed as a constant, q is the current point position, is the target point location, and the gravitational field expresses the attraction of the target to the organism;
[0026] S102: Construct a repulsive field in the potential field to express the repulsive effect of obstacles on organisms; when approaching the obstacle boundary, the potential field value approaches infinity, and when the distance from the obstacle boundary exceeds a certain specified distance, the potential field value decreases to 0. Set p(q) as the distance from the organism to the obstacle boundary. , is the obstacle position, is located within the boundary of the obstacle region, which gives the function of the repulsive potential field, is the distance affected by the obstacle, is the repulsion gain;
[0027] S103: Constructing potential field, potential field ; The potential field is the sum of the gravitational field and the repulsive field;
[0028] S104: Solve the trajectory using the gradient descent method: Starting from the initial structure, move forward along the negative gradient direction of the potential field for a length of m, repeat at the new structure, move forward along the negative gradient direction of the potential field for a length of m, until reaching the final structure point. The iterative algorithm of the gradient descent method is described in detail below:
[0029] S104_1: Starting from the starting point, move forward in the direction of the negative gradient of the potential field by a length of m, i; from the initial point to the first construction point, i is assigned a value of 0;
[0030] S104_2: If , iterate so that ; The value of i increases by 1; if , when the target end point is reached, the output sequence is ; is the coefficient that determines the advancement distance at the i-th iteration, represents the current structure after the i-th iteration, and the structure sequence containing the result of each iteration is the final path obtained by planning;
[0031] S104_3: Return to step S104_2;
[0032] After planning the path, the monitoring and alarm devices are arranged. On the active path, the monitoring and alarm devices are arranged according to the construction points. At the same time, according to the distance between the construction points, the monitoring and alarm devices are arranged at the middle point between two construction points that are far apart. The monitoring and alarm devices include humidity sensors, high temperature monitors, movable cameras and infrared monitors. Each alarm device includes a signal module to communicate with the system.
[0033] In step S2, the monitoring and alarm device is used to check the area in the forest farm, check for high-temperature flammable points, and identify reflective objects and flammable objects by image. The image detection unit is based on an intelligent camera, and the temperature-sensitive lens initially monitors the abnormal temperature of the object surface, and performs alarm processing for temperatures exceeding the threshold. In the smoldering stage, the smoke emitted by the object is identified based on the smoke detection instrument and an alarm is issued. In the burning stage, the fire is monitored based on the visible light camera. At this time, the fire is initially formed, and the equipment alarms and uploads fire information, including location information and fire pictures. Some light-gathering objects gather light and project it onto dead branches and fallen leaves, which also poses a risk of causing fire. These light-gathering objects are now found through image recognition, and high-definition cameras are set up at high points and run 360° around to collect images in the forest farm, and the images are processed as follows:
[0034] S201, converting the image into a grayscale image and performing parallel filtering to reduce high-frequency noise;
[0035] S202, thresholding the grayscale image, setting the pixel value P exceeding x to 255, and setting the pixel value P less than x to 0, so as to highlight the bright area in the blurred image;
[0036] S203, corrosion and expansion operations to remove noise in the image;
[0037] S204, drawing spots in the image, where the spots are highlight areas;
[0038] S205, mark the spot as Connect the spots in a unified direction, remove the spots near the straight line, and select the scattered spots. According to the discrete law, random light-gathering objects will not form a straight line. The spots that form a straight line are judged to be image highlights caused by direct sunlight.
[0039] The image processing process can be implemented in multiple image processing software. The system can embed these software to achieve fast processing, obtain the highlighted parts, and upload the images containing the highlights to the system, and arrange personnel to process these light-gathering objects.
[0040] In step S3, the environmental warning module accesses the interface of the meteorological website, and captures the meteorological information of the forest farm in the next n days according to the forest farm area information located by the system, and judges the possibility of inducing forest fires through precipitation, temperature, wind speed and relative humidity; among them, when the precipitation in 30 days is less than the threshold, the forest farm is relatively dry, and the probability of fire increases. Combined with the humidity monitoring data in the forest farm, when the humidity is less than the critical point, the warning level is increased; when high temperature climate occurs, the system monitors the maximum temperature during the day, and combines the influence of wind force and direction on forest fires. When the daily temperature difference is greater than m°C, the greater the wind force, the more conducive the fire is to burn; combined with the forest fire spread speed calculation formula, , is the correction coefficient of the fuel configuration pattern in the forest farm, is the wind speed adjustment factor, is the terrain slope adjustment factor, where Initial spread rate, , where T is the highest daily temperature, W is the average wind level at noon, h is the lowest daily humidity, and a, b, c, and D are constant coefficients. When the forest fire spread rate R under the meteorological conditions exceeds the threshold, the fire warning level is increased.
[0041] In step S4, for the forest where a fire has occurred, after receiving the alarm information through the device, the system immediately captures the data, finds the nearest aerospace equipment, and requests the forest image information; based on the forest fire spread speed in step S3, the burning range can be obtained. Based on the burning range, a fire smoke diffusion model based on Gaussian distribution is introduced to calculate the smoke concentration at a certain point within the fire range. ,in is the smoke concentration value produced under current natural conditions, is the wind speed influence coefficient, is the wind direction influence coefficient, is the humidity influence coefficient, S is the smoke intensity at the fire point source, b is the Gaussian distribution parameter, and r is the distance from the fire point source; the smoke concentration at various locations in the fire scene and the smoke concentration at residential areas outside the fire scene are calculated, and the residents of the affected residential areas are arranged to evacuate, and rescue routes that are less affected by smoke are designed for rescue personnel.
[0042] Compared with the prior art, the beneficial effects achieved by the present invention are:
[0043] The present invention introduces a fire smoke diffusion model based on Gaussian distribution to calculate the smoke concentration in different areas within the fire combustion range and the smoke diffusion range, which helps firefighters avoid high-concentration smoke areas and organize the evacuation of people in affected residential areas near the fire in advance. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The accompanying 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 of the present invention. In the accompanying drawings:
[0045] Figure 1 This is a module composition diagram of a forest fire prevention multi-layer early warning monitoring system using big data technology in the present invention;
[0046] Figure 2 This is a schematic diagram of the steps of a multi-layer early warning monitoring method for forest fire prevention using big data technology. DETAILED DESCRIPTION
[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0048] like Figure 1-2 As shown, the present invention provides the following technical solutions: a multi-layer early warning monitoring system and method for forest fire prevention using big data technology, the early warning monitoring system includes a three-layer early warning monitoring mechanism, a forest area monitoring module based on geographical areas, an environmental early warning module based on weather conditions, and an aerial photography monitoring module based on aviation equipment;
[0049] The forest monitoring module places alarm devices based on the current geographical location and forest characteristics to monitor and check for flammable objects in areas with frequent human activities.
[0050] The equipment focuses on monitoring flammable areas in the forest, focusing on monitoring flammable trees, capturing fire-prone points in the forest, and promptly discovering abnormally high temperatures and fire points;
[0051] The environmental early warning module issues early warnings for high temperatures and persistent droughts based on weather forecasts;
[0052] It enables forest rangers to arrange patrol plans in advance based on weather forecast information and check for potential fire spots;
[0053] The aerial monitoring module performs fire monitoring and fire spread analysis based on images taken by aviation equipment;
[0054] It enables firefighters to safely and effectively build isolated firebreaks to prevent the fire from spreading further;
[0055] According to the above technical solution, the forest monitoring module includes a regional monitoring unit, an image monitoring unit and a device management unit;
[0056] The regional monitoring unit uses a positioning system to collect the geographical location of the current forest area, reads the number, type and quantity of forests in the corresponding forest area from the database, and arranges monitoring and alarm devices accordingly; the image monitoring unit uses an infrared lens to identify flammable points and high-temperature locations to detect combustion hazards; the device management unit uniformly manages the alarm devices in the field and optimizes the layout of the devices.
[0057] According to the above technical solution, the environmental warning module includes a temperature warning unit, a drought warning unit and a wind warning unit;
[0058] The temperature warning unit monitors the temperature conditions in the forest through the temperature detection instruments in the forest farm, issues alarms for continuous high temperature conditions and extreme high temperature conditions, collects weather information through an interface, and adjusts the risk level of fire warning by predicting future temperature conditions; the drought warning unit detects the humidity conditions in the forest farm through the humidity detector in the forest farm, collects precipitation information, and predicts the duration of drought; the wind warning unit is turned on in drought and high temperature conditions, monitors the wind speed and wind direction of the day, and determines the impact of wind on the area.
[0059] According to the above technical solution, the aerial photography monitoring module includes an information acquisition unit, an image analysis unit and an intelligent pre-judgment unit;
[0060] The information collection unit is connected to the entry points of different satellites and aerospace equipment to obtain open remote sensing image resources in real time and download them to the local system; the image analysis unit retrieves images of the fire scene, analyzes the images, and determines the burning range and spreading direction of the fire; the intelligent prediction unit determines the burning time of the fire and the impact range of the fire smoke based on the image analysis results, and provides a plan direction for the fire fighting plan.
[0061] A forest fire prevention multi-layer early warning monitoring method using big data technology includes the following steps:
[0062] S1. Collect regional geographic and forest information, arrange monitoring and alarm devices and manage them in a unified manner;
[0063] S2. Patrol and identify flammable spots and high-temperature areas in the forest farm to check for burning hazards;
[0064] S3. Capture temperature and humidity data in the forest farm, and combine it with meteorological conditions to determine the fire warning level;
[0065] S4. Monitor the fire scene, analyze fire information through thermal imaging, determine the direction of fire spread, and predict the duration of continuous burning and the final scope of fire impact.
[0066] In step S1, the current forest farm is located, and according to the type and size of the forest farm, the size of the loss caused by the forest farm fire is calculated, the priority of early warning monitoring is arranged, and the detection device and the alarm device are arranged according to the characteristics and distribution positions of the trees in the forest farm; according to the classification of forest types, they are generally divided into protective forests, timber forests, firewood forests and special-purpose forests, among which timber forests and special-purpose forests have high economic and scientific research value, and firewood forests are flammable. High-value timber forests and special-purpose forests and flammable firewood forests are monitored in key areas; the device management unit monitors and manages all alarm monitoring devices in the forest farm and dispatches them in a unified manner; according to the frequency of human activities in the forest farm, the devices are arranged, the target points of the activity area in the forest farm are calibrated, the biological activity path is predicted and planned, the monitoring alarm devices are arranged along the path, and the artificial potential field algorithm is used to establish the path. The specific steps are as follows:
[0067] S101: Constructing the gravitational field in the potential field, using the parabolic gravitational field model ; is the gravitational gain, that is, the gain is the attraction of the target to the organism, expressed as a constant. The data is simulated by different models, and the values obtained are different. In this method, MATLAB software is used to build the model and simulate the data; q is the current point position, is the target point location, and the gravitational field expresses the attraction of the target to the organism;
[0068] S102: Construct a repulsive field in the potential field to express the repulsive effect of obstacles on organisms; when approaching the obstacle boundary, the potential field value approaches infinity, and when the distance from the obstacle boundary exceeds a certain specified distance, the potential field value decreases to 0. Set p(q) as the distance from the organism to the obstacle boundary. , is the obstacle position, is located within the boundary of the obstacle region, which gives the function of the repulsive potential field, is the distance affected by the obstacle, is the repulsive force gain, which is simulated as a constant value by software;
[0069] S103: Constructing potential field, potential field ; The potential field is the sum of the gravitational field and the repulsive field;
[0070] S104: Solve the trajectory using the gradient descent method: Starting from the initial structure, move forward along the negative gradient direction of the potential field for a length of m, repeat at the new structure, move forward along the negative gradient direction of the potential field for a length of m, until reaching the final structure point. The iterative algorithm of the gradient descent method is described in detail below:
[0071] S104_1: Starting from the starting point, move forward in the direction of the negative gradient of the potential field by a length of m, i; from the initial point to the first construction point, i is assigned a value of 0;
[0072] S104_2: If , iterate so that ; The value of i increases by 1; if , when the target end point is reached, the output sequence is ; is the coefficient that determines the advancement distance at the i-th iteration, It represents the current structure after the i-th iteration, and the structure sequence containing the result of each iteration is the final activity path obtained by planning;
[0073] S104_3: Return to step S104_2;
[0074] After planning the path, the monitoring and alarm devices are arranged. On the active path, the monitoring and alarm devices are arranged according to the construction points. At the same time, according to the distance between the construction points, the monitoring and alarm devices are arranged at the middle point between two construction points that are far apart. The construction points are where human activities are frequent, and voice broadcast devices are arranged to prompt forest fire ban related information; the monitoring and alarm devices include humidity sensors, high temperature monitors, movable cameras and infrared monitors, and each alarm device includes a signal module to communicate with the system.
[0075] In step S2, the monitoring and alarm device is used to check the area in the forest farm, check for high-temperature flammable points, and identify reflective objects and flammable objects by image. The image detection unit is based on an intelligent camera, and the temperature-sensitive lens initially monitors the abnormal temperature of the object surface, and performs alarm processing for temperatures exceeding the threshold. In the smoldering stage, the smoke emitted by the object is identified based on the smoke detection instrument and an alarm is issued. In the burning stage, the fire is monitored based on the visible light camera. At this time, the fire is initially formed, and the equipment alarms and uploads fire information, including location information and fire pictures. Some light-gathering objects gather light and project it onto dead branches and fallen leaves, which also poses a risk of causing fire. These light-gathering objects are now found through image recognition, and high-definition cameras are set up at high points and run 360° around to collect images in the forest farm, and the images are processed as follows:
[0076] S201, converting the image into a grayscale image and performing parallel filtering to reduce high-frequency noise;
[0077] S202, thresholding the grayscale image, setting the pixel value P exceeding x to 255, and setting the pixel value P less than x to 0, so as to highlight the bright area in the blurred image;
[0078] S203, corrosion and expansion operations to remove noise in the image;
[0079] S204, drawing spots in the image, where the spots are highlight areas;
[0080] S205, mark the spot as Connect the spots in a unified direction, remove the spots near the straight line, and select the scattered spots. According to the discrete law, random light-gathering objects will not form a straight line. The spots that form a straight line are judged to be image highlights caused by direct sunlight.
[0081] The image processing process can be implemented in multiple image processing software. The system can embed these software to achieve fast processing, obtain the highlighted parts, and upload the images containing the highlights to the system, and arrange personnel to process these light-gathering objects.
[0082] In step S3, the environmental warning module accesses the interface of the meteorological website, and captures the meteorological information of the forest farm in the next n days according to the forest farm area information located by the system, and judges the possibility of inducing forest fires through precipitation, temperature, wind speed and relative humidity; among them, when the precipitation in 30 days is less than 100 mm, the forest farm is relatively dry and the probability of fire increases. Combined with the humidity monitoring data in the forest farm, when the humidity is less than 55%, the warning level is increased; when high temperature climate occurs, the system monitors the maximum temperature during the day, and combines the influence of wind force and direction on forest fires. When the daily temperature difference is greater than 7°C, the greater the wind force, the more conducive the fire is to burn; combined with the forest fire spread speed calculation formula, , is the correction coefficient of the fuel configuration pattern in the forest farm, is the wind speed adjustment factor, is the terrain slope adjustment factor, where Initial spread rate, , where T is the highest daily temperature, W is the average wind level at noon, h is the lowest daily humidity, and a, b, c, and D are constant coefficients. When the forest fire spread rate R under the meteorological conditions exceeds the threshold, the fire warning level is increased.
[0083] In step S4, for the forest where a fire has occurred, after receiving the alarm information through the device, the system immediately captures the data, finds the nearest aerospace equipment, and requests the forest image information; based on the forest fire spread speed in step S3, the burning range can be obtained. Based on the burning range, a fire smoke diffusion model based on Gaussian distribution is introduced to calculate the smoke concentration at a certain point within the fire range. ,in is the smoke concentration value produced under current natural conditions, is the wind speed influence coefficient, is the wind direction influence coefficient, is the humidity influence coefficient, S is the smoke intensity at the fire point source, b is the Gaussian distribution parameter, and r is the distance from the fire point source; this step is based on the computer to calculate each point, which can efficiently and quickly traverse, perform repeated calculations, and obtain the smoke concentration at each point, calculate the smoke concentration at each location in the fire scene and the smoke concentration at the residential area outside the fire scene, arrange the evacuation of residents in the affected residential areas, and design a disaster relief route that is less affected by smoke for the firefighters.
[0084] Example 1
[0085] In step S4, the smoke density is calculated for a point E in the forest and a point F outside the burning range of the forest. The smoke density value of the forest under natural conditions is 0; at position E in the fire field, the smoke intensity at the nearest fire point source is 30, and the smoke density calculated based on position E is =312pmm;
[0086] At position F outside the fire scene, the smoke intensity at the nearest fire source is 10. =0.25pmm; the smoke density at point E in the fire scene is high and not suitable for passage; the smoke density at point F outside the fire scene is low and not affected.
[0087] Example 2
[0088] In step S1, the gravitational field in the force field is first constructed. ;in is the gravitational gain, which is realistically 0.04; q is the current point position (2,2), is the target point position (25,27). The gravitational field expresses the attraction of the target to the organism. At (2,2), the gravitational field is expressed as = (625+259)=23.08;
[0089] Construct a repulsive field in the potential field to express the repulsive effect of obstacles on organisms; set p(q) to be 0.4, the distance from the organism to the obstacle boundary. is the obstacle position, is located within the boundary of the obstacle region, which gives the function of the repulsive potential field, The distance affected by the obstacle is 0.8. is the repulsive force gain, which is realistically 0.7; the repulsive force field at (2,2) is: = =0.546;
[0090] S103: Construct potential field, potential field at point (2,2) ; The potential field is the sum of the gravitational field and the repulsive field;
[0091] S104: Solve the trajectory using the gradient descent method: Starting from the initial structure, move forward 0.7 length along the negative gradient direction of the potential field, repeat at the new structure, move forward 0.7 length along the negative gradient direction of the potential field, until reaching the final structure point. The iterative algorithm of the gradient descent method is described in detail below:
[0092] S104_1: Starting from the starting point, move forward in the direction of the negative gradient of the potential field by a length of 1. i; from the initial point to the first construction point, i is assigned a value of 0;
[0093] S104_2: After multiple software iterations, the output sequence ; The sequence of construction points containing the results of each iteration is the final activity path obtained by planning;
[0094] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A multi-layer early warning monitoring method for forest fire prevention using big data technology. The following steps are involved: S1. Collect regional geographic and forest information, arrange monitoring and alarm devices and manage them in a unified manner; S2. Patrol and identify flammable spots and high-temperature areas in the forest farm to check for burning hazards; S3. Capture temperature and humidity data in the forest farm, combine it with meteorological conditions, and determine the fire warning level; S4. Monitor the fire scene, analyze the fire information through thermal imaging, determine the direction of fire spread, predict the duration of continuous burning and the final scope of fire impact; In step S1, the current forest farm is located, and according to the type and size of the forest farm, the loss caused by the forest farm fire is calculated, the priority of early warning monitoring is arranged, and the monitoring device and the alarm device are arranged according to the characteristics and distribution of trees in the forest farm; high-value timber forests, special-purpose forests and flammable firewood forests are monitored; all alarm monitoring devices in the forest farm are monitored and managed through the device management unit, and dispatched in a unified manner; according to the frequency of biological activities in the forest farm, the devices are arranged, the target points of the activity area in the forest farm are calibrated, the biological activity path is predicted and planned, the monitoring alarm devices are arranged along the path, and the artificial potential field algorithm is used to establish the path. The specific steps are as follows: S101: constructing a gravitational field in a potential field, wherein the gravitational field expresses the attraction of the target to the organism; S102: Construct a repulsive field in the potential field to express the repulsive effect of obstacles on the target; S103: Constructing potential field, potential field ; The potential field is the sum of the gravitational field and the repulsive field; S104: Use the gradient descent method to solve the trajectory: starting from the initial structure, move forward m length along the negative gradient direction of the potential field, repeat at the new structure, move forward m length along the negative gradient direction of the potential field, until reaching the final structure point: After planning the path, the monitoring and alarm devices are arranged. On the active path, the monitoring and alarm devices are arranged according to the construction points. At the same time, according to the distance between the construction points, the monitoring and alarm devices are arranged at the middle point between two construction points that are far apart. The monitoring and alarm devices include humidity sensors, high temperature monitors, movable cameras and infrared monitors. Each alarm device includes a signal module to communicate with the system.
2. A forest fire prevention multi-layer early warning monitoring method using big data technology according to claim 1, Features: In step S2, the monitoring and alarm device is used to check the area within the forest farm, check for high-temperature and flammable points, and identify reflective objects and flammable objects through images; the image detection unit is based on an intelligent camera, and the temperature-sensing lens initially monitors the abnormal surface temperature of the object, and performs alarm processing for temperatures exceeding the threshold; in the smoldering stage, the smoke emitted by the object is identified based on the smoke detection instrument and an alarm is issued; During the combustion stage, the visible light camera detects the flames, which is the initial stage of the fire. The equipment will sound an alarm and upload fire information, including location information and fire pictures. Some light-gathering objects gather light and also have the risk of causing fires. These light-gathering objects can be found through image recognition. High-definition cameras are set up at high points and run 360° around to collect images and identify light-gathering objects.
3. A multi-layer early warning monitoring method for forest fire prevention using big data technology according to claim 1, Features: In step S3, the interface of the meteorological website is accessed, and based on the forest area information located by the system, the meteorological information of the forest in the next n days is captured, and the possibility of inducing forest fires is judged through precipitation, temperature, wind speed and relative humidity; combined with the forest fire spread formula, when the forest fire spread speed R under the meteorological conditions exceeds the threshold, the fire warning level is increased.
4. A forest fire prevention multi-layer early warning monitoring method using big data technology according to claim 1, Features: In step S4, for the forest farm where a fire has occurred, after receiving the alarm information through the device, the system immediately captures the data, finds the nearest aerospace equipment, and requests the forest farm image information; based on the forest fire spread speed in step S3, the burning range can be obtained, and based on the burning range, a fire smoke diffusion model based on Gaussian distribution is introduced to calculate the smoke concentration in various places in the fire scene and the smoke concentration in the residential areas outside the fire scene, arrange the residents of the affected residential areas to evacuate, and design a disaster relief route that is less affected by the smoke for the rescue personnel.
Citation Information
Patent Citations
Forest-fire-prevention digital informatization integration command system
CN105096508A
Big data mining based integrated forest fire prevention informatization system
CN105719421A
Environment three-dimensional perception and safety early warning protection system and method of distribution network live working robot
CN110421557A
Remote forest fire prevention monitoring system and method
CN111476964A
Forest fire monitoring, early warning and rescue integrated system based on Beidou navigation
CN111899453A
