A method and system for automatic identification and precise management of forest stress
By using forest thermal field evolution models and heat accumulation identification models, we can accurately identify heat spot retention information in forest areas, determine precise spraying strategies, solve the problem of insufficient ability to distinguish spray intensity and timing, realize refined spraying management under forest stress, and reduce the risk of mis-spraying of canopy heat spots.
Patent Information
- Application Number
- CN202610784408.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-02
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2046-06-02
AI Technical Summary
Existing spray cooling technologies have insufficient spatial resolution of spray intensity and timing in forest management, leading to the risk of mis-spraying hot spots in the canopy, resulting in localized high-temperature retention on leaf surfaces, localized wilting, and leaf edge scorching. Furthermore, they cannot accurately identify the differences in spraying needs at different locations within the forest area.
By acquiring forest area environmental monitoring data and gridded canopy sensing data, and utilizing forest area thermal field evolution models and heat accumulation identification models, global thermal distribution and gridded hot spot retention information are calculated to determine precise spraying strategies, including independent high-frequency pulse spraying, global coordinated continuous spraying, and spray blocking strategies, to accurately identify hidden hot spots and false hot spots and reduce the risk of mis-spraying.
It enables precise spray management under forest stress conditions, reduces the risk of mis-spraying of canopy heat spots, prevents localized heat damage and diseases, and improves the reliability and precision of spray management.
Smart Images

Figure CN122594884B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of data processing technology, specifically relating to a method and system for automatic identification and precise management of forest stress. Background Technology
[0002] Spray cooling is a commonly used technique in the management of trees under high-temperature stress. It can reduce the surface temperature of the canopy and alleviate heat accumulation on leaves by absorbing heat through water mist evaporation, thereby reducing the risk of heat damage in high-temperature and low-humidity environments. However, the current control of spray cooling methods is still limited to macro-control at the level of the entire forest area or large sub-area. The observation method is based on air temperature, relative humidity, or temperature thresholds at a few fixed points for triggering. Even when multi-sensor joint control is used, the control logic still uses a few discrete observation points to represent the state of a large forest stand. This is often accompanied by the risk of mis-spraying hot spots in the canopy. That is, due to insufficient spatial resolution of spray intensity and timing, management anomalies occur in local canopy areas. These include areas where persistent high-temperature hot spots have formed in the canopy but have not been cooled in time, resulting in high temperature retention on the leaves, local wilting, and aggravated leaf edge scorching. Meanwhile, areas where there is no heat accumulation in the canopy are repeatedly sprayed, causing prolonged moisture on the canopy surface and increasing the conditions for the occurrence of local diseases.
[0003] The risk of mis-spraying canopy hot spots arises from the varying spraying needs at different locations within the forest. This typically involves inconsistencies in supplemental heating conditions, heat dissipation conditions, and transpiration baseline conditions. Therefore, even adjacent grid points operating at the same time can exhibit completely different canopy warming rates and heat retention levels. Traditional uniform spraying management often assumes that the stand's heating state is approximately synchronous. However, in reality, grid points subjected to strong direct sunlight and weak ventilation are more likely to develop high-temperature hot spots first, while grid points with higher canopy closure, better transpiration conditions, or stronger air exchange warm up relatively slowly. This exacerbates the aforementioned abnormal leaf temperature retention, hot spot residue, and repeated wetting of non-hotspot areas, leading to the risk of mis-spraying canopy hot spots. Especially in forest stress management with prominent exposed patches of low humidity and sustained high afternoon temperatures, the strong solar radiation input and the limitation of local wind fields by topography and canopy structure make the differences in canopy heat accumulation and hot spot migration more pronounced between different grid points. Even after the same round of spraying, some grid points may experience rapid reheating while others remain continuously moist, exhibiting spatial differentiation. Therefore, there is an urgent need for a method and system for automatic identification and precise management of forest stress. Summary of the Invention
[0004] The purpose of this invention is to propose an automatic identification and precise management method and system for forest stress, in order to solve one or more technical problems existing in the prior art, and at least provide a beneficial option or create conditions to effectively improve the reliability and precision of forest stress spray management, and reduce the risk of mis-spraying of canopy hot spots due to insufficient spatial resolution of spray intensity and spray timing.
[0005] To achieve the above objectives, according to a first aspect of the present invention, a method for automatic identification and precise management of forest stress is provided, the method comprising the following steps: Acquire forest environment monitoring data of the target forest area, as well as grid canopy sensing data of any target grid point in the target forest area; Forest area environmental monitoring data is input into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area; Based on the thermal accumulation identification model corresponding to the target grid point, the corresponding grid point thermal spot retention information is predicted according to the grid point canopy sensing data. Based on the spatial difference matching calculation between global thermal distribution information and grid hot spot retention information, the precise spraying strategy for the target grid point is determined. As an optional implementation, in the first aspect of the present invention, the acquisition of forest environmental monitoring data of the target forest area and grid canopy sensing data of any target grid point in the target forest area includes: acquiring forest environmental monitoring data through forest meteorological stations set up in the target forest area, wherein the forest environmental monitoring data is used to characterize the overall macro-meteorological characteristics of the target forest area; The grid canopy sensing data is acquired by miniature sensors placed within the target grid points. The grid canopy sensing data is used to characterize the local microclimate features within the target grid points. The target grid point is the smallest spatial unit that contains the tree canopy, exposed patches of forest windows, or closed areas of the understory, obtained by dividing the target forest area into two-dimensional planar grids or three-dimensional spatial grids.
[0006] As an optional implementation, in the first aspect of the present invention, the step of inputting forest area environmental monitoring data into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area includes: All forest area environmental monitoring data, corresponding monitoring locations, and the three-dimensional canopy topography model of the target forest area are input into the trained forest area thermal field evolution model to obtain the global thermal distribution information of the target forest area. Global thermal distribution information includes the macroscopic thermal stress probability and heat source driving factors corresponding to each grid point in the three-dimensional canopy topography structure model; The forest thermal field evolution model was trained using a training dataset that included multiple training environment monitoring data and corresponding monitoring location labels, three-dimensional structural model location labels, and thermal distribution labels.
[0007] As an optional implementation, in the first aspect of the present invention, the step of predicting the corresponding grid point hot spot retention information based on the grid point canopy sensing data using the heat accumulation identification model corresponding to the target grid point includes: Determine the grid point parameters corresponding to the target grid point; Based on grid parameters and a similarity algorithm, the target heat accumulation identification model is determined from multiple candidate identification models. The grid canopy sensing data is input into the target heat accumulation identification model to obtain the grid hot spot retention information corresponding to the target grid. The grid hot spot retention information includes the local hot spot retention probability and local heating rate of the target grid.
[0008] As an optional implementation, in the first aspect of the invention, the step of determining the precise spraying strategy for the target grid point based on the spatial difference matching calculation between global thermal distribution information and grid point hot spot retention information includes: From all target grid points in the target forest area, grid points whose retention severity of hot spot information is greater than a preset threshold are selected to obtain multiple high-risk hot spot grid points. Calculate the spatial matching degree between grid hot spot retention information and global thermal distribution information for all high-risk hot spot grid points; The average local hotspot residual probability of all high-risk hotspot grid points is calculated to obtain the residual probability parameter; When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is lower than the preset matching threshold, the precision spraying strategy for the target grid point is determined to be an independent high-frequency pulse spraying strategy. When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is not lower than the preset matching threshold, the precise spraying strategy for the target grid point is determined to be the global collaborative continuous spraying strategy. When the residual probability parameter is not greater than the preset safe residual threshold, the precise spraying strategy for the target grid point is determined to be the spray blocking and shielding strategy.
[0009] As an optional implementation, in the first aspect of the invention, the retention severity of the grid hotspot retention information is calculated through the following steps: For each target grid point, calculate the deviation ratio of the local heating rate of that target grid point from the local heating rate of all target grid points; Among all other target grid points, grid points whose local heating rate is in the same preset range as the target grid point are identified, resulting in multiple associated grid points; The weighted summation average of the local hot spot residual probability of the target grid point and all associated grid points is calculated to obtain the probability average parameter. Among them, the calculation weight of the local hot spot residual probability of the associated grid point is less than the calculation weight of the target grid point, and the calculation weight of the local hot spot residual probability of the associated grid point is inversely proportional to the spatial wind direction distance between the associated grid point and the target grid point. The severity of hot spot retention information at the target grid point is obtained by multiplying the probability average parameter and the deviation ratio weight; the deviation ratio weight is proportional to the deviation ratio.
[0010] As an optional implementation, in the first aspect of the present invention, the calculation of the spatial matching degree between the grid hot spot retention information and the global thermal distribution information of all high-risk hot spot grid points includes: combining the local heating rates of all high-risk hot spot grid points into a first thermal set; and combining the heat source driving factors corresponding to all grid point locations in the global thermal distribution information into a second thermal set. The local hot spot residual probability of all high-risk hot spot grid points and the corresponding target grid point location are combined into a first probability set; the grid point location and the corresponding macroscopic thermal stress probability in the global thermal distribution information are combined into a second probability set. A first similarity is calculated based on a first thermodynamic set and a second thermodynamic set; a second similarity is calculated based on a first probability set and a second probability set. The product of the first and second similarities is used as the spatial matching degree between the grid hot spot retention information and the global thermal distribution information of all high-risk hot spot grid points.
[0011] As an optional implementation, in the first aspect of the invention, after determining the precise spraying strategy for the target grid points, the method further includes: When the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is greater than the safe residual threshold, the unified spraying command for the target forest area is blocked, and an independent high-frequency spraying command for the target grid point is generated. When the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is not greater than the safe residual threshold, the spray trigger signal for the target grid point is blocked.
[0012] A second aspect of this invention discloses an automatic identification and precise management system for forest stress, the system comprising: The environmental data acquisition module is used to acquire forest environmental monitoring data of the target forest area, as well as grid canopy sensing data of any target grid point in the target forest area; The global thermal monitoring module is used to input forest area environmental monitoring data into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area; The grid hot spot monitoring module is used to predict the corresponding grid hot spot retention information based on the heat accumulation identification model corresponding to the target grid point and the grid canopy sensing data. The precision spray management module is used to determine the precision spray strategy for the target grid point by matching and calculating the spatial differences between global thermal distribution information and grid point hot spot retention information.
[0013] As an optional implementation, in the second aspect of the present invention, the specific method by which the environmental data acquisition module acquires forest environmental monitoring data of the target forest area and grid canopy sensing data of any target grid point in the target forest area includes: acquiring forest environmental monitoring data through forest meteorological stations set up in the target forest area, wherein the forest environmental monitoring data is used to characterize the overall macro-meteorological characteristics of the target forest area; The grid canopy sensing data is acquired by miniature sensors placed within the target grid points. The grid canopy sensing data is used to characterize the local microclimate features within the target grid points. The target grid point is the smallest spatial unit that contains the tree canopy, exposed patches of forest windows, or closed areas of the understory, obtained by dividing the target forest area into two-dimensional planar grids or three-dimensional spatial grids.
[0014] As an optional implementation, in the second aspect of the present invention, the specific method by which the global thermal monitoring module inputs forest area environmental monitoring data into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area includes: All forest area environmental monitoring data, corresponding monitoring locations, and the three-dimensional canopy topography model of the target forest area are input into the trained forest area thermal field evolution model to obtain the global thermal distribution information of the target forest area. Global thermal distribution information includes the macroscopic thermal stress probability and heat source driving factors corresponding to each grid point in the three-dimensional canopy topography structure model; The forest thermal field evolution model was trained using a training dataset that included multiple training environment monitoring data and corresponding monitoring location labels, three-dimensional structural model location labels, and thermal distribution labels.
[0015] As an optional implementation, in a second aspect of the invention, the specific method by which the grid hotspot monitoring module predicts the corresponding grid hotspot retention information based on the heat accumulation identification model corresponding to the target grid point and the grid canopy sensing data includes: Determine the grid point parameters corresponding to the target grid point; Based on grid parameters and a similarity algorithm, the target heat accumulation identification model is determined from multiple candidate identification models. The grid canopy sensing data is input into the target heat accumulation identification model to obtain the grid hot spot retention information corresponding to the target grid. The grid hot spot retention information includes the local hot spot retention probability and local heating rate of the target grid.
[0016] As an optional implementation, in a second aspect of the invention, the precision spray management module determines the specific method of the precision spray strategy for the target grid point based on spatial difference matching calculations between global thermal distribution information and grid point hot spot retention information, including: From all target grid points in the target forest area, grid points whose retention severity of hot spot information is greater than a preset threshold are selected to obtain multiple high-risk hot spot grid points. Calculate the spatial matching degree between grid hot spot retention information and global thermal distribution information for all high-risk hot spot grid points; The average local hotspot residual probability of all high-risk hotspot grid points is calculated to obtain the residual probability parameter; When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is lower than the preset matching threshold, the precision spraying strategy for the target grid point is determined to be an independent high-frequency pulse spraying strategy. When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is not lower than the preset matching threshold, the precise spraying strategy for the target grid point is determined to be the global collaborative continuous spraying strategy. When the residual probability parameter is not greater than the preset safe residual threshold, the precise spraying strategy for the target grid point is determined to be the spray blocking and shielding strategy.
[0017] As an optional implementation, in the second aspect of the present invention, the specific method for determining the severity of the retention of grid hot spot retention information in the precision spray management module further includes: For each target grid point, calculate the deviation ratio of the local heating rate of that target grid point from the local heating rate of all target grid points; Among all other target grid points, grid points whose local heating rate is in the same preset range as the target grid point are identified, resulting in multiple associated grid points; The weighted summation average of the local hot spot residual probability of the target grid point and all associated grid points is calculated to obtain the probability average parameter. Among them, the calculation weight of the local hot spot residual probability of the associated grid point is less than the calculation weight of the target grid point, and the calculation weight of the local hot spot residual probability of the associated grid point is inversely proportional to the spatial wind direction distance between the associated grid point and the target grid point. The severity of hot spot retention information at the target grid point is obtained by multiplying the probability average parameter and the deviation ratio weight; the deviation ratio weight is proportional to the deviation ratio.
[0018] As an optional implementation, in the second aspect of the invention, the specific method by which the precision spray management module calculates the spatial matching degree between the grid hot spot retention information and the global thermal distribution information of all high-risk hot spot grid points further includes: The local heating rates of all high-risk hotspot grid points are combined into a first thermodynamic set; the heat source driving factors corresponding to all grid point locations in the global thermodynamic distribution information are combined into a second thermodynamic set. The local hot spot residual probability of all high-risk hot spot grid points and the corresponding target grid point location are combined into a first probability set; the grid point location and the corresponding macroscopic thermal stress probability in the global thermal distribution information are combined into a second probability set. A first similarity is calculated based on a first thermodynamic set and a second thermodynamic set; a second similarity is calculated based on a first probability set and a second probability set. The product of the first and second similarities is used as the spatial matching degree between the grid hot spot retention information and the global thermal distribution information of all high-risk hot spot grid points.
[0019] As an optional implementation, in the second aspect of the invention, the specific method by which the precision spray management module determines the precision spray strategy for the target grid points includes: When the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is greater than the safe residual threshold, the unified spraying command for the target forest area is blocked, and an independent high-frequency spraying command for the target grid point is generated. When the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is not greater than the safe residual threshold, the spray trigger signal for the target grid point is blocked.
[0020] Preferably, all undefined variables in this invention, if not explicitly defined, can be manually set thresholds.
[0021] A third aspect of this invention discloses another method and system for automatic identification and precise management of forest stress, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the automatic identification and precise management method for forest stress disclosed in the first aspect of the present invention.
[0022] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the automatic identification and precise management method for forest stress disclosed in the first aspect of the present invention.
[0023] Compared with existing technologies, the embodiments of this invention have the following beneficial effects: This invention breaks through the limitations of traditional macro-level extensive control in forest areas. By offsetting the spatial differences between the thermal field and the actual micro-level heat retention, it accurately identifies hidden heat spots and false hot spots. Furthermore, by introducing spatial distance weighting of wind direction vectors, it endows the system with the ability to predict thermal convection drift. Based on this, a multi-level control matrix is output, enabling targeted application of high-frequency pulsed sprays to prevent water accumulation in truly high-risk areas, while forcibly implementing flow interruption and shielding in areas requiring no cooling. This proactively curbs irreversible thermal damage to local trees, breaks the vicious chain of canopy diseases induced by repeated ineffective wetting, and eliminates the risk of mis-spraying canopy heat spots, thereby achieving differentiated and refined spray cooling management under high-temperature stress scenarios for forest trees. Attached Figure Description
[0024] The above and other features of the present invention will become more apparent from the detailed description of the embodiments shown in conjunction with the accompanying drawings. In the accompanying drawings, the same reference numerals denote the same or similar elements. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort. In the drawings: Figure 1 The diagram shown is a flowchart of an automatic identification and precise management method for forest stress disclosed in an embodiment of the present invention; Figure 2 The diagram shown is a structural diagram of an automatic forest stress identification and precision management system disclosed in an embodiment of the present invention; Figure 3 The diagram shown is a structural diagram of another forest stress automatic identification and precision management system disclosed in an embodiment of the present invention. Detailed Implementation
[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0026] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.
[0027] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0028] This invention discloses an automatic identification and precise management method and system for forest stress. By offsetting the spatial difference between the thermal field and the actual microscopic heat retention, it accurately identifies hidden heat spots and false hot spots. Furthermore, by introducing spatial distance weighting based on wind direction vectors, it endows the system with the ability to predict thermal convection drift. Based on this, a multi-level control matrix is output to apply high-frequency pulsed sprays to truly high-risk areas to prevent water accumulation, while forcibly implementing flow interruption and shielding in areas requiring no cooling. This proactively curbs irreversible thermal damage to localized forest trees, breaks the vicious cycle of canopy diseases induced by repeated ineffective wetting, and eliminates the risk of mis-spraying canopy heat spots, thus achieving refined spray cooling management under high-temperature stress on forest trees. Detailed descriptions follow.
[0029] Example 1 Please see Figure 1 , Figure 1 The diagram shown is a flowchart of a method for automatic identification and precise management of forest stress according to an embodiment of the present invention; wherein, Figure 1 The described method for automatic identification and precise management of forest stress can be applied to data processing systems / data processing equipment / data processing servers (including local processing servers or cloud processing servers). For example... Figure 1 As shown, this method for automatic identification and precise management of forest stress may include the following operations: 101. Acquire forest environment monitoring data of the target forest area, as well as grid canopy sensing data of any target grid point in the target forest area; Optionally, the target forest area can be a natural protected forest area, an economic fruit forest area, an urban greening area, or an artificial windbreak and sand-fixing forest area; this invention does not limit such areas.
[0030] Optionally, the data acquisition trigger mechanism can be periodic continuous polling in the system background, or high-frequency acquisition automatically triggered after receiving a high temperature and low humidity warning signal issued by the local meteorological department. This invention does not limit the specific mechanisms.
[0031] 102. Input forest area environmental monitoring data into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area; Optionally, the operating environment of the forest thermal field evolution model can be deployed in a cloud computing cluster, a local high-performance server in the forest management center, or a distributed edge computing gateway; this invention does not limit this.
[0032] Optionally, the predicted global thermal distribution information can cover the predicted trend of thermal field evolution within the next 1 to 24 hours in the time dimension, so as to provide a time margin for advance scheduling of forest spray water resources. This invention does not limit this.
[0033] 103. Based on the thermal accumulation identification model corresponding to the target grid point, predict the corresponding grid point hot spot retention information according to the grid point canopy sensing data; Optionally, in order to save system computing power, the invocation mechanism of the thermal accumulation identification model corresponding to the target grid point can be triggered by the macroscopic temperature rise feature in the global thermal distribution information. This invention does not limit this.
[0034] Optionally, the predicted grid hot spot retention information can be visualized on the forest area digital twin management screen using a 3D rendering engine in the form of a 3D dynamic thermal layer or a red highlighted warning label. This invention does not limit this.
[0035] 104. Based on the spatial difference matching calculation between global thermal distribution information and grid hot spot retention information, the precise spraying strategy for the target grid point is determined. Optionally, the final physical execution vehicle for the precision spraying strategy can be a micro-irrigation network fixedly laid in the forest, a self-propelled spraying robot shuttling through forest passages, or an agricultural drone that automatically cruises along a set route. This invention does not limit the scope of the application.
[0036] Optionally, while determining the precision spraying strategy, the system can also simultaneously generate forest stress intervention record archives and push collaborative alarm information containing grid abnormal locations and automatic intervention status to the mobile terminals of forestry managers. This invention does not limit this.
[0037] As can be seen, the above-mentioned embodiments of the invention acquire forest environment monitoring data of the target forest area and grid canopy sensing data of any target grid point, respectively predict macroscopic global thermal distribution information and microscopic local grid hot spot retention information based on the model, and determine a precise spraying strategy by matching the spatial differences between the two. This enables precise hot spot identification and dynamic cooling control based on dual-track data fusion, improves the reliability and precision of forest stress management, and reduces the risk of mis-spraying of canopy hot spots due to insufficient spatial resolution of spray intensity and spraying timing.
[0038] As an optional embodiment, the steps above, obtaining forest environmental monitoring data of the target forest area and grid canopy sensing data of any target grid point in the target forest area, include: obtaining forest environmental monitoring data through forest meteorological stations set up in the target forest area, the forest environmental monitoring data being used to characterize the overall macro-meteorological characteristics of the target forest area; The grid canopy sensing data is acquired by miniature sensors placed within the target grid points. The grid canopy sensing data is used to characterize the local microclimate features within the target grid points. The target grid point is the smallest spatial unit that contains the tree canopy, exposed patches of forest windows, or closed areas of the understory, obtained by dividing the target forest area into two-dimensional planar grids or three-dimensional spatial grids.
[0039] Optionally, the forest meteorological station may include a large ultrasonic anemometer, a high-precision temperature and humidity Stevenson screen, a full radiation meter or a long-wave radiometer; this invention does not limit the scope of the invention.
[0040] Optionally, the miniature sensor may include an infrared leaf temperature thermocouple, a canopy miniature anemometer, a transmittance meter, or a patch-type leaf surface humidity sensor; this invention does not limit the type of sensor.
[0041] Optionally, the forest area environmental monitoring data may include the total solar radiation input above the forest area, the macroscopic air temperature above the canopy, the overall relative humidity, or the macroscopic wind field vector, which is not limited in this invention.
[0042] Optionally, the grid canopy sensing data may include the actual leaf surface temperature, local micro-wind velocity, leaf surface water retention, or photosynthetically active radiation distribution within the grid, which is not limited in this invention.
[0043] Optionally, the smallest spatial unit of the two-dimensional planar grid division can be a standard square grid with a side length of 1 to 5 meters, or an irregular polygonal planar area determined by the vertical projection of the canopy of a single target tree and the forest window boundary. This invention does not limit the scope of the grid division.
[0044] Optionally, the smallest spatial unit of the three-dimensional spatial grid division can be a standard cubic voxel with a side length of 1 to 5 meters, or an irregular spatial geometry determined by the outer envelope of the canopy and the forest window boundary of a single target tree. This invention does not limit the specific unit.
[0045] Optionally, the acquisition frequency of forest area environmental monitoring data and gridded canopy sensing data can be high-frequency continuous acquisition at the second level, periodic timed acquisition at the minute level, or adaptive frequency acquisition triggered by a specific meteorological threshold. This invention does not limit the frequency.
[0046] Optionally, the acquired forest area environmental monitoring data and gridded canopy sensing data can be in the form of raw analog signals, digital signals after analog-to-digital conversion, or structured time-series data after noise reduction and filtering preprocessing. This invention does not impose any limitations on these data formats.
[0047] Optionally, the data acquisition process can be implemented through real-time collection and uploading via forest area IoT base stations, LoRa wireless communication networks, drone patrol relays, or edge computing gateways; this invention does not limit this process.
[0048] Optionally, in order to achieve automatic identification and accurate landing of the smallest spatial unit during gridding, the canopy, exposed patches of forest windows, and closed areas of the understory can be quantitatively defined by multispectral image recognition algorithms, three-dimensional lidar point cloud classification, or canopy analyzers. This invention does not limit these methods.
[0049] Optionally, for the tree canopy, the definition criteria can be a continuous branch and leaf community area within the spatial range corresponding to the target grid point, where the echo height of the three-dimensional lidar point cloud is greater than a preset height threshold and the point cloud density is greater than a preset density threshold, or a green cover plane area in the visible light orthophoto image where the vegetation index is greater than a preset vegetation threshold. This invention does not limit the definition.
[0050] Optionally, for exposed forest gaps, the defining criteria can be bare ground, herbaceous or shrub areas within the spatial range corresponding to the target grid point that are not covered by tall tree branches and leaves and can directly receive direct solar radiation. Numerically, this is expressed as the photosynthetically active radiation intensity in this area reaching more than 80% of the photosynthetically active radiation intensity of completely unobstructed open ground outside the forest, or the echo height in the three-dimensional lidar point cloud being less than 20% of the average height of the tree canopy. This invention does not impose any limitations.
[0051] Optionally, the definition of a closed forest canopy area can be a spatial area located below the vertical projection of the tree canopy, where light and ventilation conditions are weakened due to shading by branches and leaves. Numerically, this is expressed as a forest stand canopy closure greater than 0.6, or the photosynthetically active radiation intensity in this area being less than 30% of the photosynthetically active radiation intensity of completely unshaded open space outside the forest. This invention does not impose any limitations on this.
[0052] As can be seen, the above optional embodiments define the content and differences between forest area environmental monitoring data and gridded canopy sensing data, the corresponding sensor types, and the gridding standards. This enables multi-scale monitoring data to effectively characterize the overall macroscopic heating conditions of the forest area and the actual heat accumulation and transpiration characteristics of individual micro-grid points. It also assists in achieving accurate hot spot identification and difference matching based on dual-track data, improving the reliability and precision of forest stress spraying management, and providing a data foundation for further reducing the risk of mis-spraying of canopy hot spots due to insufficient spatial resolution of spray intensity and timing.
[0053] As an optional embodiment, the step of inputting forest area environmental monitoring data into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area includes: All forest area environmental monitoring data, corresponding monitoring locations, and the three-dimensional canopy topography model of the target forest area are input into the trained forest area thermal field evolution model to obtain the global thermal distribution information of the target forest area. Global thermal distribution information includes the macroscopic thermal stress probability and heat source driving factors corresponding to each grid point in the three-dimensional canopy topography structure model; The forest thermal field evolution model was trained using a training dataset that included multiple training environment monitoring data and corresponding monitoring location labels, three-dimensional structural model location labels, and thermal distribution labels.
[0054] Optionally, the three-dimensional canopy topography model of the target forest area can be represented as a grid model obtained by vertically superimposing and fusing a digital elevation model and a digital canopy height model, a three-dimensional gridded stereo model, or a spatial voxel grid model generated by airborne lidar scanning, in order to completely reconstruct the three-dimensional spatial topographic undulations, canopy density distribution, canopy geometry, and forest window exposure boundaries of the entire target forest area. This invention does not limit the scope of the model.
[0055] Optionally, each monitoring location can be characterized by three-dimensional spatial coordinates combining latitude and longitude coordinates with absolute altitude, relative rows and columns and hierarchical geometric numbering of local physical grids, or node indexes in a spatial topology structure. This invention does not impose any limitations on these methods.
[0056] Optionally, the forest thermal field evolution model can be an algorithm architecture based on a three-dimensional spatial convolution neural network model, an STGCN spatiotemporal graph convolutional network model, a three-dimensional spatial attention mechanism model, or a spatiotemporal Transformer model specifically for fine-tuning the temporal evolution of grids. This invention does not limit the specific model.
[0057] Optionally, the macroscopic thermal stress probability can be expressed as a continuous numerical risk probability value between 0 and 1, or it can be expressed as a multi-level discrete risk level such as no risk, low risk, medium risk and high risk. This invention does not limit it.
[0058] Optionally, the heat source driving factors may include strong direct solar radiation, weak ventilation obstruction caused by local topography or forest stand dead corners, local heat accumulation caused by excessive local vegetation cover, or local low humidity coupled with high temperature caused by excessive water evaporation. This invention does not limit these factors.
[0059] Optionally, the training dataset for the forest thermal field evolution model can be derived from historical extreme high-temperature meteorological logs of the target forest area, flux observation records of long-term ecological monitoring stations, historical inversion maps of airborne thermal infrared imaging, or extreme drought and sun exposure simulation datasets generated using a high-fidelity digital twin engine. This invention does not impose any limitations on these datasets.
[0060] Optionally, the specific training and fine-tuning process of the forest area thermal field evolution model can adopt a multi-task learning architecture. The total loss function of the model can be a weighted combination of the mean squared error loss function and the multi-class cross-entropy loss function. The optimizer can be the AdamW optimizer. The learning rate can be dynamically adjusted using a cosine annealing scheduling strategy. Data augmentation can be performed during training by introducing spatial rotation, translation, and injecting Gaussian noise into meteorological parameters. This invention does not impose any limitations.
[0061] As can be seen, the above optional embodiments define the specific architecture of the model for global thermal distribution prediction, the quantitative representation of the three-dimensional canopy topographic structure model and monitoring location, the output form of macroscopic thermal stress probability and heat source driving factors, as well as the training set source and deep learning optimization mechanism of the model. Thus, based on accurate global prediction, multi-source macroscopic meteorological parameters and complex three-dimensional canopy and topographic structure features are coupled in a multidimensional way in the deep learning space, enabling the model to automatically learn and accurately simulate the three-dimensional thermal field dynamics of the macroscopic weather scene evolving on the complex forest underlying surface. This provides a highly reliable global background baseline for subsequent local micro-grid point thermal spot comparison, significantly improving the accuracy of predicting the large-scale evolution trend of forest thermal stress.
[0062] As an optional embodiment, the step above, predicting the corresponding grid point hot spot retention information based on the grid point canopy sensing data using the heat accumulation identification model corresponding to the target grid point, includes: Determine the grid point parameters corresponding to the target grid point; Based on grid parameters and a similarity algorithm, the target heat accumulation identification model is determined from multiple candidate identification models. The grid canopy sensing data is input into the target heat accumulation identification model to obtain the grid hot spot retention information corresponding to the target grid. The grid hot spot retention information includes the local hot spot retention probability and local heating rate of the target grid.
[0063] Optionally, the grid parameters may include the dominant tree species type, stand canopy closure, leaf area index, average crown height, tree age stage, slope and aspect of the terrain, basic transpiration resistance coefficient, or soil water retention performance index within the target grid point. This invention does not impose any limitations on these parameters.
[0064] Optionally, the determination of grid point parameters can be obtained by querying a pre-constructed database of basic geographic and vegetation attributes of forest areas, or by real-time inversion extraction based on high-resolution multispectral remote sensing images combined with lidar point cloud data, or by manual calibration and input on-site. This invention does not limit the scope of the determination.
[0065] Optionally, the similarity algorithm can be a cosine similarity algorithm, an Euclidean distance algorithm, a Mahalanobis distance calculation, or a feature-weighted K-nearest neighbor matching algorithm; this invention does not limit the algorithm.
[0066] Optionally, the multiple candidate identification models can be multiple differentiated dedicated time series prediction models that are pre-trained in advance using massive historical microclimate data and stored in a cloud model library based on different tree species characteristics (such as broad-leaved forests and coniferous forests), different aging stages, or different canopy closure levels. This invention does not limit these models.
[0067] Optionally, the target heat accumulation recognition model can be based on an LSTM long short-term memory network, a GRU gated recurrent unit, a 1D-CNN temporal convolutional neural network, a Transformer time series prediction model, or an ensemble learning model based on gradient boosting decision trees such as XGBoost or LightGBM. The target heat accumulation recognition model is used to specifically capture short-term time series mutations and heat accumulation features in local micro-environments, and this invention does not limit it.
[0068] Optionally, the probability of residual local heat spots can be expressed as the percentage probability that, within a set time window in the future (e.g., the next 30 minutes to 2 hours, with a default value of 30 minutes), the surface temperature of the local tree canopy will remain above the physiological safety threshold, and due to transpiration exhaustion or windless conditions, the tree will be unable to dissipate heat through its own physiological regulation, thus forming a continuous high-temperature heat injury. This invention does not limit this.
[0069] Optionally, the local heating rate can be expressed as a continuous quantitative value in degrees Celsius per minute (°C / min) or degrees Celsius per hour (°C / h), or as a normalized relative heating rate gradient, to intuitively reflect the intensity of local canopy heat accumulation. This invention does not limit the expression of the local heating rate.
[0070] As can be seen, the above optional embodiments define the specific physical composition and acquisition method of grid parameters, the classification dimension of similarity algorithm and candidate model library, the underlying temporal network architecture of target heat accumulation identification model, and the specific physical representation and numerical dimensions of various indicators in grid hot spot retention information. Through the model adaptive selection and matching mechanism based on grid parameters, the spatial heterogeneity of different grid points in the forest under the conditions of supplementary heating, heat dissipation and transpiration is fully considered, which improves the pertinence and accuracy of local hot spot feature prediction and provides highly reliable local risk feature support for subsequent spatial difference collision with macro-global thermal distribution.
[0071] As an optional embodiment, the step above, determining the precise spraying strategy for the target grid point based on the spatial difference matching calculation between global thermal distribution information and grid point hot spot retention information, includes: From all target grid points in the target forest area, grid points whose retention severity of hot spot information is greater than a preset threshold are selected to obtain multiple high-risk hot spot grid points. Calculate the spatial matching degree between grid hot spot retention information and global thermal distribution information for all high-risk hot spot grid points; The average local hotspot residual probability of all high-risk hotspot grid points is calculated to obtain the residual probability parameter; When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is lower than the preset matching threshold, the precision spraying strategy for the target grid point is determined to be an independent high-frequency pulse spraying strategy. When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is not lower than the preset matching threshold, the precise spraying strategy for the target grid point is determined to be the global collaborative continuous spraying strategy. When the residual probability parameter is not greater than the preset safe residual threshold, the precise spraying strategy for determining the target grid point is the spray blocking and shielding strategy.
[0072] Optionally, the preset severity threshold can be a fixed empirical value threshold, an adaptive dynamic threshold dynamically generated based on the current season and phenological period, or a specific severity limit specifically set for different heat-resistant tree species. This invention does not limit the threshold.
[0073] Optionally, the spatial matching degree can be represented as a comprehensive evaluation index between 0 and 1, which can be used to intuitively quantify the degree of spatial deviation between the local microscopic real heat accumulation state and the macroscopic global theoretical predicted heat field. This invention does not limit this.
[0074] Optionally, when calculating the average value of the probability of local hot spot remnants, a simple arithmetic average algorithm, a weighted average algorithm based on the topological distance between grid points, or a truncated average algorithm after removing isolated extreme values can be used. This invention does not limit the method.
[0075] Optionally, the preset safe residual threshold can be set to an occurrence probability range of 60% to 80%, and the preset matching threshold can be set to a similarity range of 0.5 to 0.7. The specific values can be calibrated on-site according to the actual evapotranspiration water consumption characteristics of the forest area, and this invention does not limit them.
[0076] Optionally, the independent high-frequency pulse spray strategy may include activating a micro-irrigation solenoid valve specific to the target grid point and driving the atomizing nozzle with a pulse width modulation signal with a control cycle of 30 seconds and a duty cycle of 33.3% (i.e., 10 seconds on and 20 seconds off in a single control cycle); wherein, the high-frequency feature is specifically manifested as the solenoid valve performing a second-level action frequency of 2 start-stop switching per minute, which is used to provide powerful and water-proof physical cooling for hidden hot spots that are not macroscopically apparent but have localized heat damage, and this invention is not limited thereto.
[0077] Optionally, the global coordinated continuous spraying strategy may include triggering the booster pump of the main pipeline in the forest area and linking the spraying units of the target grid point and its surrounding grid points to carry out continuous spraying for 3 to 5 minutes to cope with large-area high-temperature heat damage that occurs simultaneously at both the macro and micro levels. This invention does not limit this.
[0078] Optionally, the spray blocking strategy may include forcibly locking the execution node corresponding to the target grid point in the control center and cutting off any spray trigger level signal for that area to prevent canopy water accumulation in areas where transpiration conditions are still acceptable or have been repeatedly moistened, thereby cutting off the conditions for the breeding of pests and diseases. This invention is not limited to this.
[0079] As can be seen, the above optional embodiments define the physical application performance of degree thresholds and high-risk hot spot grid points, the calculation logic of spatial matching degree and residual probability parameters, the quantitative definition of various control thresholds, and the specific hardware actions and execution parameters corresponding to the three direct control spray strategies. Thus, by constructing a direct rule control matrix based on multi-dimensional threshold comparison, the real-time performance and interpretability of the system's spray decision are greatly improved, providing clear execution instructions for the hardware control gateway, promoting the solution to the problem of mismatch between macro-unified control and micro-local needs, and reducing the risk of mis-spraying of canopy hot spots.
[0080] As an optional embodiment, the severity of the retention of grid hot spot retention information in the above steps is calculated by the following steps: for each target grid point, calculate the deviation ratio of the local heating rate of the target grid point from the local heating rate of all target grid points; Among all other target grid points, grid points whose local heating rate is in the same preset range as the target grid point are identified, resulting in multiple associated grid points; The weighted summation average of the local hot spot residual probability of the target grid point and all associated grid points is calculated to obtain the probability average parameter. Among them, the calculation weight of the local hot spot residual probability of the associated grid point is less than the calculation weight of the target grid point, and the calculation weight of the local hot spot residual probability of the associated grid point is inversely proportional to the spatial wind direction distance between the associated grid point and the target grid point. The severity of hot spot retention information at the target grid point is obtained by multiplying the probability average parameter and the deviation ratio weight; the deviation ratio weight is proportional to the deviation ratio.
[0081] Optionally, the deviation ratio of the local heating rate from the local heating rate of all target grid points can be obtained by calculating the absolute difference between the local heating rate of the target grid point and the median or average local heating rate of all target grid points, and then dividing it by the standard deviation of the local heating rate of all target grid points (i.e., Z-Score normalization algorithm). This can be used to accurately quantify the degree of abnormal prominence of the target grid point in the overall forest area warming process. This invention does not limit this.
[0082] Optionally, the preset interval can be expressed as a numerical range of five to ten percent fluctuation in the local heating rate of the target grid point, or as an absolute change range of ±0.5 degrees Celsius per hour; the determined associated grid points physically represent spatial heterogeneity co-occurrence classes with highly similar thermal potential and evaporation basis conditions to the target grid point under the current macro-climate, and this invention does not limit this.
[0083] Optionally, when calculating the weighted average, the calculation weight of the target grid point can be set to a fixed dominant coefficient between 0.6 and 0.8, and the sum of the calculation weights of all associated grid points is limited to between 0.2 and 0.4, so as to ensure that the assessment of the probability of local hot spot residue is mainly determined by the state of the target grid point itself, and to avoid the distortion of the assessment results due to the disturbance of the surrounding environment. This invention does not impose any limitations.
[0084] Optionally, the spatial wind direction distance differs from the conventional Euclidean physical straight-line distance. It is represented by the corrected distance obtained by dividing the physical straight-line distance between the associated grid point and the target grid point by the wind direction influence factor. The value of the wind direction influence factor is determined based on the cosine of the angle between the current micro-wind field vector and the vector connecting the two grid points. When the associated grid point is located upwind of the target grid point, the wind direction influence factor is greater than 1, which causes the calculated spatial wind direction distance to be drastically shortened numerically. Therefore, according to the inverse proportionality principle, the upwind associated grid point is given a large calculation weight. This invention does not limit this.
[0085] Optionally, the relationship between the deviation ratio weight and the deviation ratio can be expressed as a linear proportional function mapping, or as an exponential amplification function based on the natural logarithm, to exponentially increase the final retention severity value when the temperature rise is extremely abnormal and the deviation ratio is extremely large, so as to ensure that the system will prioritize the capture of extremely high-risk situations. This invention does not limit this.
[0086] As can be seen, the above optional embodiments limit the specific statistical calculation method of the deviation ratio, the quantitative range of the preset interval, the numerical allocation boundary of the master-slave calculation weight, and the type of function mapping. In particular, it innovatively introduces the spatial wind direction distance calculation rule that integrates the vector characteristics of the micro-wind field, thereby fully considering the dynamic law of heat drift and superposition of the canopy hot spot along a specific direction under weak wind conditions. This allows the target grid point to perceive in advance the risk of increased thermal convection caused by the approach of the upwind hot spot, greatly improving the spatial coherence and physical fitting ability of the severity assessment.
[0087] As an optional embodiment, the step of calculating the spatial matching degree between the grid hotspot retention information and the global thermal distribution information of all high-risk hotspot grid points includes: The local heating rates of all high-risk hot spot grid points are combined into a first thermodynamic set; The heat source driving factors corresponding to all grid points in the global thermal distribution information are combined into a second thermal set; The local hot spot residual probability of all high-risk hot spot grid points and the corresponding target grid point location are combined into a first probability set; The locations of all grid points in the global thermal distribution information and their corresponding macroscopic thermal stress probabilities are combined into a second probability set. The first similarity is calculated based on the first and second thermodynamic sets; The second similarity is calculated based on the first probability set and the second probability set; The product of the first and second similarities is used as the spatial matching degree between the grid hot spot retention information and the global thermal distribution information of all high-risk hot spot grid points.
[0088] Optionally, the first thermal set can be represented in terms of data structure as a one-dimensional feature vector sequence composed of continuous temperature rise values of high-risk hot spot grid points; the second thermal set can be represented as a high-dimensional feature matrix formed by splicing together multiple independent heat source driving factors in the global grid, such as strong direct radiation or weak ventilation obstruction, after performing one-hot encoding or word embedding mapping respectively. This invention does not limit the scope of the feature set.
[0089] Optionally, the first probability set and the second probability set can be represented in terms of data expression form as a spatial point cloud dataset containing grid spatial geometric coordinates and corresponding probability feature dimensions, a two-dimensional probability density distribution map, or a directed graph data structure with geographic topological relationships. This invention does not limit the representation of these data.
[0090] Optionally, the first similarity can be calculated based on the first and second thermodynamic sets, using a preset physical causal relationship mapping matrix. Specifically, this can be achieved by first aligning the anomalous gradients of local heating rates with specific macroscopic heat source driving factors using vector features, and then calculating the degree of agreement between the two in terms of heating causes using a cosine similarity algorithm or a Pearson correlation coefficient algorithm. This invention does not impose any limitations on this method.
[0091] Optionally, the second similarity can be calculated based on the first and second probability sets using a distance metric algorithm specifically designed to measure the differences between two sets of spatial probability distribution fields. Specifically, this can be achieved by using the bulldozer distance algorithm to measure the deformation algebraic distance between the macroscopic thermal stress probability distribution and the local hotspot residual probability distribution in terms of spatial topology, or by using the KL divergence relative entropy algorithm to calculate the distribution divergence of the spatial probability density, and then taking the reciprocal of the distance or after normalization using a nonlinear activation function as the second similarity. This invention does not impose any limitations on this method.
[0092] Optionally, the product of the first similarity and the second similarity can be used as the spatial matching degree. Its physical control significance lies in constructing a dual verification threshold that simultaneously possesses the logic of thermal cause and the spatial physical overlap. The product value will approach the high confidence limit if and only if the macroscopic prediction of cause can reasonably explain the rapid temperature rise of micro-grid points and the macroscopic danger probability space field accurately covers the microscopic high-risk location. This invention does not impose any limitations on this.
[0093] As can be seen, the above optional embodiments define the underlying data format of each set, the specific measurement algorithms for the first and second similarities, and the cybernetics significance behind the similarity product fusion. This allows for the orthogonal fusion of semantic similarity in the causal dimension and spatial distribution similarity in the probabilistic dimension, based on accurate calculation of spatial differences. This design effectively improves the robustness and fault tolerance of the matching evaluation, ensuring that the system can accurately identify heterogeneous target points where macroscopic and microscopic realities are disconnected, even when facing extremely complex forest stand structures and microclimate disturbances. This provides irrefutable quantitative decision-making basis for ultimately blocking extensive management and triggering independent high-frequency spraying.
[0094] As an optional embodiment, after determining the precise spraying strategy for the target grid points in the above steps, the method further includes: When the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is greater than the safe residual threshold, the unified spraying command for the target forest area is blocked, and an independent high-frequency spraying command for the target grid point is generated. When the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is not greater than the safe residual threshold, the spray trigger signal for the target grid point is blocked.
[0095] Optionally, the preset matching threshold and safe residual threshold can be stored in the microcontroller memory of the on-site edge computing gateway, or they can be dynamically updated daily by the cloud server based on the current seasonal phenological characteristics, soil moisture, and the heat tolerance critical physiological indicators of the dominant tree species in the target forest area. This invention does not limit these settings.
[0096] Optionally, blocking the unified spraying command for the target forest area can be manifested at the underlying control link as the regional edge gateway intercepting and discarding the global spraying broadcast message issued by the central controller at the network communication protocol layer, or by cutting off the default start level signal of the booster pump of the main pipeline network in the forest area through the hard wire of the relay logic circuit in the power control cabinet. This invention does not limit this.
[0097] Optionally, generating independent high-frequency spray commands for target grid points can be manifested as the system sending customized messages containing specific device addresses, such as MAC addresses or node IDs, to the node controller corresponding to the target grid point through a wireless IoT communication protocol. The messages contain explicit pulse control opening and closing width parameters, driving the atomizing nozzles in that local area to operate independently in a short-time, multiple-pulse working mode, in order to strongly suppress local high temperatures without causing a large amount of water accumulation in the canopy. This invention is not limited to this.
[0098] Optionally, the spray trigger signal for the target grid point can be shielded. At the software system level, this can be manifested by setting the operation flag position of the target grid point in the control platform device mapping table to a forced sleep or locked state. At the hardware physical level, this can be manifested by pulling down the drive pin level of the corresponding micro solenoid valve, so that even if a large-area spray cooling operation is being performed in the surrounding adjacent areas of the grid point, the nozzles in the target grid point remain completely silent and physically disconnected. This invention does not limit this.
[0099] As can be seen, the above optional embodiments define the physical storage and update methods for various thresholds, the hardware and software interception methods for blocking global commands, the message generation logic for independent high-frequency commands, and the system-level locking mechanism for shielding trigger signals. This establishes a robust, secure underlying rewriting and blocking mechanism at the execution end of the precise strategy. This ensures that the system has the authority to bypass the sluggish macro-level unified control to target and extinguish hidden hot spots; simultaneously, it allows for the forced implementation of physical silencing to completely cut off the water mist supply, absolutely eliminating the problem of increased damage caused by repeated wetting of non-hotspot areas at the execution level, thus closing the loop to prevent the risk of mis-spraying of canopy hot spots.
[0100] Example 2 Please see Figure 2 , Figure 2 The diagram shown is a structural diagram of an automatic forest stress identification and precision management system disclosed in an embodiment of the present invention. Wherein, Figure 2 The described automatic identification and precision management system for forest stress can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes a local processing server or a cloud processing server). For example... Figure 2 As shown, this automatic identification and precision management system for forest stress may include: The environmental data acquisition module 201 is used to acquire forest environmental monitoring data of the target forest area, as well as grid canopy sensing data of any target grid point in the target forest area; The global thermal monitoring module 202 is used to input forest area environmental monitoring data into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area; The grid hot spot monitoring module 203 is used to predict the corresponding grid hot spot retention information based on the heat accumulation identification model corresponding to the target grid point and the grid canopy sensing data. The precision spray management module 204 is used to determine the precision spray strategy for the target grid point by matching and calculating the spatial difference between the global thermal distribution information and the grid point heat spot retention information.
[0101] As can be seen, the above-described embodiments of the invention precisely identify hidden heat spots and false hot spots by offsetting the spatial differences between the thermal field and the actual microscopic heat retention. Furthermore, by introducing spatial distance weighting based on wind direction vectors, the system gains the ability to predict thermal convection drift. Based on this, a multi-level control matrix can be output to apply high-frequency pulsed sprays to truly high-risk areas to prevent water accumulation, while forcibly implementing flow interruption and shielding in areas requiring no cooling. This can proactively curb irreversible thermal damage to local trees, break the vicious cycle of canopy diseases induced by repeated ineffective wetting, and thus eliminate the risk of mis-spraying canopy heat spots, achieving refined spray cooling management under high-temperature stress on trees.
[0102] As an optional embodiment, the environmental data acquisition module acquires forest area environmental monitoring data through forest meteorological stations set up in the target forest area. The forest area environmental monitoring data is used to characterize the overall macro-meteorological characteristics of the target forest area. It also acquires grid canopy sensing data through miniature sensors set up in the target grid points. The grid canopy sensing data is used to characterize the local microclimate characteristics within the target grid points. The target grid point is the smallest spatial unit containing tree canopy, exposed forest gaps, or closed understory areas, obtained by dividing the target forest area into two-dimensional planar grids or three-dimensional spatial grids.
[0103] As can be seen, the above-described embodiments define the content and differences between forest area environmental monitoring data and gridded canopy sensing data, the corresponding sensor types, and the gridding standards. This enables multi-scale monitoring data to effectively characterize the overall macroscopic heating conditions of the forest area and the actual heat accumulation and transpiration characteristics of individual micro-grid points. This facilitates accurate hotspot identification and difference matching based on dual-track data, improving the reliability and precision of forest stress spraying management. Furthermore, it provides a data foundation for further reducing the risk of mis-spraying canopy hotspots due to insufficient spatial resolution of spray intensity and timing. As an optional embodiment, the global thermal monitoring module inputs all forest area environmental monitoring data, corresponding monitoring locations, and the three-dimensional canopy topography structure model of the target forest area into the trained forest area thermal field evolution model to obtain the global thermal distribution information of the target forest area. The global thermal distribution information includes the macroscopic thermal stress probability and heat source driving factors corresponding to each grid point in the three-dimensional canopy topography structure model. The forest area thermal field evolution model is trained using a training dataset that includes multiple training environmental monitoring data and corresponding monitoring location labels, three-dimensional structure model location labels, and thermal distribution labels.
[0104] As can be seen, the above optional embodiments define the specific architecture of the model for global thermal distribution prediction, the quantitative representation of the three-dimensional canopy topographic structure model and monitoring location, the output form of macroscopic thermal stress probability and heat source driving factors, as well as the training set source and deep learning optimization mechanism of the model. This allows for multi-dimensional coupling of multi-source macroscopic meteorological parameters with complex three-dimensional canopy and topographic structure features in the deep learning space, enabling the model to automatically learn and accurately simulate the three-dimensional thermal field dynamics of macroscopic weather patterns evolving on complex forest underlying surfaces. This provides a highly reliable global background baseline for subsequent local micro-grid point thermal spot comparison, significantly improving the accuracy of predicting the large-scale evolution trend of forest thermal stress.
[0105] As an optional embodiment, the grid hot spot monitoring module determines the grid parameters corresponding to the target grid point; based on the grid parameters, and using a similarity algorithm, it determines the target heat accumulation identification model from multiple candidate identification models; the grid canopy sensing data is input into the target heat accumulation identification model to obtain the grid hot spot retention information corresponding to the target grid point; the grid hot spot retention information includes the local hot spot residue probability and local heating rate of the target grid point.
[0106] This study defines the specific physical composition and acquisition method of grid parameters, the classification dimension of similarity algorithms and candidate model libraries, the underlying temporal network architecture of the target heat accumulation identification model, and the specific physical representation and numerical dimensions of various indicators in grid hot spot retention information. Through a model adaptive selection and matching mechanism based on grid parameters, it fully considers the spatial heterogeneity of different grid points within the forest in terms of supplementary heating conditions, heat dissipation conditions, and transpiration basic conditions, thereby improving the pertinence and accuracy of local hot spot feature prediction. This provides highly reliable local risk feature support for subsequent spatial difference collision with macro-global thermal distribution.
[0107] As an optional embodiment, the precision spraying management module selects grid points from all target grid points in the target forest area whose heat spot retention severity exceeds a preset threshold, thus obtaining multiple high-risk heat spot grid points; calculates the spatial matching degree between the heat spot retention information of all high-risk heat spot grid points and the global thermal distribution information; calculates the average value of the local heat spot residual probability of all high-risk heat spot grid points to obtain a residual probability parameter; when the residual probability parameter is greater than a preset safe residual threshold and the spatial matching degree is lower than a preset matching threshold, the precision spraying strategy for the target grid point is determined to be an independent high-frequency pulse spraying strategy; when the residual probability parameter is greater than a preset safe residual threshold and the spatial matching degree is not lower than a preset matching threshold, the precision spraying strategy for the target grid point is determined to be a global collaborative continuous spraying strategy; when the residual probability parameter is not greater than a preset safe residual threshold, the precision spraying strategy for the target grid point is determined to be a spray blocking and shielding strategy.
[0108] The calculation logic for the degree threshold and physical application performance of high-risk hot spot grid points, spatial matching degree and residual probability parameters, quantitative definition of various control thresholds, and specific hardware actions and execution parameters corresponding to three direct control spray strategies are defined. By constructing a direct rule control matrix based on multi-dimensional threshold comparison, the real-time performance and interpretability of the system's spray decision are greatly improved, providing clear execution instructions for the hardware control gateway, promoting the solution to the problem of mismatch between macro-unified control and micro-local needs, and reducing the risk of mis-spraying of canopy hot spots.
[0109] As an optional embodiment, the precision spray management module calculates the deviation ratio of the local heating rate of each target grid point from the local heating rate of all target grid points for each target grid point; identifies grid points whose local heating rate is within the same preset range as the target grid point among all other target grid points, obtaining multiple associated grid points; calculates the weighted summation average of the local hot spot retention probabilities of the target grid point and all associated grid points to obtain a probability average parameter; wherein, the calculation weight of the local hot spot retention probability of the associated grid points is less than the calculation weight of the target grid point, and the calculation weight of the local hot spot retention probability of the associated grid points is inversely proportional to the spatial wind direction distance between the associated grid point and the target grid point; calculates the product of the probability average parameter and the deviation ratio weight to obtain the retention severity of the grid hot spot retention information of the target grid point; the deviation ratio weight is directly proportional to the deviation ratio. As can be seen, the above optional embodiments limit the specific statistical calculation method of the deviation ratio, the quantitative range of the preset interval, the numerical allocation boundary of the master-slave calculation weight, and the type of function mapping. In particular, it innovatively introduces the spatial wind direction distance calculation rule that integrates the vector characteristics of the micro-wind field, thereby fully considering the dynamic law of heat drift and superposition of the canopy hot spot along a specific direction under weak wind conditions. This allows the target grid point to perceive in advance the risk of increased thermal convection caused by the approach of the upwind hot spot, greatly improving the spatial coherence and physical fitting ability of the severity assessment.
[0110] As an optional embodiment, the precision spray management module combines the local heating rates of all high-risk hotspot grid points into a first thermal set; combines the heat source driving factors corresponding to all grid point locations in the global thermal distribution information into a second thermal set; combines the local hotspot residue probabilities of all high-risk hotspot grid points and the corresponding target grid point locations into a first probability set; combines all grid point locations in the global thermal distribution information and the corresponding macroscopic thermal stress probabilities into a second probability set; calculates a first similarity based on the first thermal set and the second thermal set; calculates a second similarity based on the first probability set and the second probability set; and uses the product of the first similarity and the second similarity as the spatial matching degree between the grid hotspot retention information of all high-risk hotspot grid points and the global thermal distribution information.
[0111] As can be seen, the above optional embodiments define the underlying data format of each set, the specific measurement algorithms for the first and second similarities, and the cybernetics significance behind the similarity product fusion. This allows for the orthogonal fusion of semantic similarity in the causal dimension and spatial distribution similarity in the probabilistic dimension, based on accurate calculation of spatial differences. This design effectively improves the robustness and fault tolerance of the matching evaluation, ensuring that the system can accurately identify heterogeneous target points where macroscopic and microscopic realities are disconnected, even when facing extremely complex forest stand structures and microclimate disturbances. This provides irrefutable quantitative decision-making basis for ultimately blocking extensive management and triggering independent high-frequency spraying.
[0112] As an optional embodiment, the precision spray management module blocks the unified spray command for the target forest area and generates an independent high-frequency spray command for the target grid when the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid is greater than the safe residual threshold; when the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid is not greater than the safe residual threshold, it blocks the spray trigger signal for the target grid.
[0113] As can be seen, the above optional embodiments define the physical storage and update methods for various thresholds, the hardware and software interception methods for blocking global commands, the message generation logic for independent high-frequency commands, and the system-level locking mechanism for shielding trigger signals. This establishes a robust, secure underlying rewriting and blocking mechanism at the execution end of the precise strategy. This ensures that the system has the authority to bypass the sluggish macro-level unified control to target and extinguish hidden hot spots; simultaneously, it allows for the forced implementation of physical silencing to completely cut off the water mist supply, absolutely eliminating the problem of increased damage caused by repeated wetting of non-hotspot areas at the execution level, thus closing the loop to prevent the risk of mis-spraying of canopy hot spots.
[0114] Example 3 Please see Figure 3 , Figure 3 The diagram shown is a structural diagram of another forest stress automatic identification and precision management system disclosed in an embodiment of the present invention. Figure 3 The described automatic identification and precision management system for forest stress is applied in a data processing system / data processing equipment / data processing server (wherein, the server includes a local processing server or a cloud processing server). For example... Figure 3 As shown, this automatic identification and precision management system for forest stress may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the automatic identification and precise management method for forest stress described in Embodiment 1.
[0115] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of a method for automatic identification and precise management of forest stress described in Embodiment 1.
[0116] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of a method for automatic identification and precise management of forest stress described in Embodiment 1.
[0117] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0118] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.
[0119] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.
[0120] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0121] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0122] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0123] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0124] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0125] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.
[0126] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.
[0127] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0128] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
Claims
1. A method for automatic identification and precise management of forest stress, characterized in that, The method includes the following steps: Acquire forest environment monitoring data of the target forest area, as well as grid canopy sensing data of any target grid point in the target forest area; Forest area environmental monitoring data is input into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area; Based on the thermal accumulation identification model corresponding to the target grid point, the corresponding grid point thermal spot retention information is predicted according to the grid point canopy sensing data. Based on the spatial difference matching calculation between global thermal distribution information and grid hot spot retention information, the precise spraying strategy for the target grid point is determined. The precise spraying strategy for target grid points is determined by matching the spatial differences between global thermal distribution information and grid point hot spot retention information. This includes: selecting grid points from all target grid points in the target forest area whose hot spot retention severity is greater than a preset threshold, thus obtaining several high-risk hot spot grid points. Calculate the spatial matching degree between grid hot spot retention information and global thermal distribution information for all high-risk hot spot grid points; The average local hotspot residual probability of all high-risk hotspot grid points is calculated to obtain the residual probability parameter; When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is lower than the preset matching threshold, the precision spraying strategy for the target grid point is determined to be an independent high-frequency pulse spraying strategy. When the residual probability parameter is greater than the preset safe residual threshold and the spatial matching degree is not lower than the preset matching threshold, the precise spraying strategy for the target grid point is determined to be the global collaborative continuous spraying strategy. When the residual probability parameter is not greater than the preset safe residual threshold, the precise spraying strategy for the target grid point is determined to be the spray blocking shielding strategy. The severity of hot spot retention information at grid points is calculated through the following steps: For each target grid point, the deviation ratio of the local heating rate of the target grid point from the local heating rate of all target grid points is calculated. Among all other target grid points, grid points whose local heating rate is in the same preset range as the target grid point are identified, resulting in multiple associated grid points; The weighted summation average of the local hot spot residual probability of the target grid point and all associated grid points is calculated to obtain the probability average parameter. Among them, the calculation weight of the local hot spot residual probability of the associated grid point is less than the calculation weight of the target grid point, and the calculation weight of the local hot spot residual probability of the associated grid point is inversely proportional to the spatial wind direction distance between the associated grid point and the target grid point. The severity of hot spot retention information at the target grid point is obtained by multiplying the probability average parameter and the deviation ratio weight; the deviation ratio weight is proportional to the deviation ratio.
2. The method for automatic identification and precise management of forest stress according to claim 1, characterized in that, Acquire forest environmental monitoring data of the target forest area, as well as grid canopy sensing data of any target grid point in the target forest area, including: acquiring forest environmental monitoring data through forest meteorological stations set up in the target forest area. The forest environmental monitoring data is used to characterize the overall macro-meteorological characteristics of the target forest area. The grid canopy sensing data is acquired by miniature sensors placed within the target grid points. The grid canopy sensing data is used to characterize the local microclimate features within the target grid points. The target grid point is the smallest spatial unit that contains the tree canopy, exposed patches of forest windows, or closed areas of the understory, obtained by dividing the target forest area into two-dimensional planar grids or three-dimensional spatial grids.
3. The method for automatic identification and precise management of forest stress according to claim 1, characterized in that, The forest area environmental monitoring data is input into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area. This includes: inputting all forest area environmental monitoring data, the corresponding monitoring locations, and the three-dimensional canopy topography structure model of the target forest area into the trained forest area thermal field evolution model to obtain the global thermal distribution information of the target forest area. Global thermal distribution information includes the macroscopic thermal stress probability and heat source driving factors corresponding to each grid point in the three-dimensional canopy topography structure model; The forest thermal field evolution model was trained using a training dataset that included multiple training environment monitoring data and corresponding monitoring location labels, three-dimensional structural model location labels, and thermal distribution labels.
4. The method for automatic identification and precise management of forest stress according to claim 1, characterized in that, Based on the thermal accumulation identification model corresponding to the target grid point, the corresponding grid point hot spot retention information is predicted according to the grid point canopy sensing data, including: determining the grid point parameters corresponding to the target grid point; Based on grid parameters and a similarity algorithm, the target heat accumulation identification model is determined from multiple candidate identification models. The grid canopy sensing data is input into the target heat accumulation identification model to obtain the grid hot spot retention information corresponding to the target grid. The grid hot spot retention information includes the local hot spot retention probability and local heating rate of the target grid.
5. The method for automatic identification and precise management of forest stress according to claim 1, characterized in that, Calculate the spatial matching degree between grid hotspot retention information and global thermal distribution information for all high-risk hotspot grid points, including: The local heating rates of all high-risk hot spot grid points are combined into a first thermodynamic set; The heat source driving factors corresponding to all grid points in the global thermal distribution information are combined into a second thermal set; The local hot spot residual probability of all high-risk hot spot grid points and the corresponding target grid point location are combined into a first probability set; The locations of all grid points in the global thermal distribution information and their corresponding macroscopic thermal stress probabilities are combined into a second probability set. The first similarity is calculated based on the first and second thermodynamic sets; The second similarity is calculated based on the first probability set and the second probability set; The product of the first and second similarities is used as the spatial matching degree between the grid hot spot retention information and the global thermal distribution information of all high-risk hot spot grid points.
6. The method for automatic identification and precise management of forest stress according to claim 1, characterized in that, After determining the precise spraying strategy for the target grid points, the following steps are also included: when the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is greater than the safe residual threshold, the unified spraying command for the target forest area is blocked, and an independent high-frequency spraying command for the target grid point is generated. When the spatial matching degree is lower than the preset matching threshold and the residual probability parameter of the target grid point is not greater than the safe residual threshold, the spray trigger signal for the target grid point is blocked.
7. An automatic identification and precise management system for forest stress, characterized in that, The system includes: The environmental data acquisition module is used to acquire forest environmental monitoring data of the target forest area, as well as grid canopy sensing data of any target grid point in the target forest area; The global thermal monitoring module is used to input forest area environmental monitoring data into the forest area thermal field evolution model to predict the global thermal distribution information of the target forest area; The grid hot spot monitoring module is used to predict the corresponding grid hot spot retention information based on the heat accumulation identification model corresponding to the target grid point and the grid canopy sensing data. The precision spray management module is used to determine the precision spray strategy for the target grid point by matching and calculating the spatial difference between global thermal distribution information and grid point hot spot retention information. Each module is used to execute the method for automatic identification and precise management of forest stress as described in any one of claims 1-6.
8. An automatic identification and precise management system for forest stress, characterized in that, The system includes: Memory containing executable program code; A processor coupled to memory; The processor calls the executable program code stored in the memory to execute the method for automatic identification and precise management of forest stress as described in any one of claims 1-6.
Citation Information
Patent Citations
Photovoltaic module hot spot diagnosis and self-adaptive cleaning method and system
CN120658208A
Energy consumption management system and method of data center
CN120872126A
