A forest health environment comprehensive monitoring method and system

Abstract

The present application relates to the technical field of intelligent regulation and control of health care environment, and in particular to a forest health care environment comprehensive monitoring method and system. The present application first acquires various data of a target forest area; further acquires air mass deposition consumption probability of each grid; further calculates air flow one-way connectivity weight in a two-dimensional graph network topology structure graph, and generates an air flow transfer probability matrix; further executes probability iterative deduction with constant source supply taking the grid where the meteorological master station is located as a divergence source node, outputs the final distribution probability of air mass of each grid, and acquires the retardation attenuation conversion distance of each grid; finally, based on the retardation attenuation conversion distance and the true value of the to-be-measured variable of the meteorological master station, the estimated concentration of the environmental index of each grid is calculated, the estimated distortion caused by terrain obstruction and water vapor interception is eliminated, and the fresh air control system of the semi-closed health care facility is accurately controlled.

Description

Technical Field

[0001] This invention relates to the field of intelligent regulation and control technology for health and wellness environments, specifically to a comprehensive monitoring method and system for forest health and wellness environments. Background Technology

[0002] Forest-based health and wellness bases require continuous monitoring of the concentration distribution of environmental indicators such as negative oxygen ions and fine particulate matter in the forest understory. Limited by high construction costs, park management typically deploys high-precision meteorological main stations only in a few open areas, while using lower-cost conventional temperature and humidity auxiliary monitoring stations in larger forest understory areas. To output an environmental data distribution map covering the entire forest area to the management platform, existing monitoring systems generally use conventional spatial interpolation algorithms based on Euclidean geometry to directly extract the straight-line distance between the grid to be measured and the meteorological main station for numerical estimation.

[0003] However, the terrain within real forest areas is complex and the canopy is dense, inevitably obstructing airflow as it diffuses outwards due to the physical barriers of ridges and tree canopies. Simultaneously, in the relatively saturated, low-lying environment, beneficial factors such as negative oxygen ions readily combine with condensed water droplets and undergo sedimentation and attenuation. Existing linear distance interpolation methods ignore the on-site attenuation caused by terrain obstruction and localized high humidity, failing to accurately reflect the true attenuation cost of airflow diffusion in complex terrain and high humidity environments. This leads to discrepancies between estimated environmental concentrations and actual conditions, consequently preventing the timely and precise activation and control of the fresh air dehumidification system within semi-enclosed health and wellness facilities, thus affecting the proactive maintenance capacity of the health and wellness environment. Summary of the Invention

[0004] To address the technical problem in existing technologies where distorted environmental concentration estimations due to neglecting topographic obstruction and water vapor consumption affect the regulation of fresh air systems in semi-enclosed health and wellness facilities, the present invention aims to provide a comprehensive monitoring method and system for forest health and wellness environments. The specific technical solution adopted is as follows: A comprehensive monitoring method for forest health and wellness environments, the method comprising: Obtain the relative depression depth and canopy shading ratio of each grid in the target forest area, and obtain the ambient temperature, relative humidity and dew point temperature difference of each grid at each time step within the current analysis window; Based on the change in dew point temperature difference of each grid, combined with the relative sinking depth and the canopy shading ratio, the air mass deposition and consumption probability of each grid is obtained; in the two-dimensional graph network topology diagram, based on the downslope difference between adjacent grids, the spatial angle between the geometric geographical orientation vector and the macro wind direction vector between grids, and the canopy shading ratio, the airflow unidirectional connectivity weight is calculated, and an airflow transfer probability matrix is ​​generated. Using the grid where the meteorological main station is located as the divergence source node, based on the air mass deposition and consumption probability and the airflow transfer probability matrix, a probability iterative deduction with constant source replenishment is performed to output the final distribution probability of air masses in each grid, and combined with the relative sinking depth, the stagnation attenuation conversion distance of each grid is obtained. Based on the attenuation reduction distance and the true value of the measured variable at the meteorological station, the concentration of environmental indicators in each grid is estimated, and the fresh air control system of the semi-enclosed health and wellness facility is controlled.

[0005] Furthermore, the method for obtaining the dew point temperature difference includes: For each grid: Substitute the ambient temperature and relative humidity at each time step into the preset Magnus empirical formula to obtain the dew point temperature at each time step; based on the ambient temperature and dew point temperature at each time step, obtain the dew point temperature difference at each time step.

[0006] Furthermore, the method for obtaining the probability of air mass deposition includes: For each grid cell: obtain the time derivative of the dew point temperature difference within the current analysis window; set the temperature difference reduction rate based on the time derivative; By combining the temperature difference reduction rate, the relative subsidence depth, and the canopy shading ratio, a comprehensive baseline for settlement consumption is obtained. After normalizing the overall deposition consumption base with the largest global deposition consumption base, the deposition consumption probability of the corresponding grid is obtained by combining it with a preset attenuation safety limit coefficient.

[0007] Furthermore, the method for obtaining the unidirectional airflow connectivity weight includes: For any two grids that are directly adjacent in space, assume that one grid is the starting grid and the other grid is the ending grid; The difference between the absolute elevation of the starting grid and the elevation of the ending grid is taken as the downslope drop; the cosine value of the spatial angle between the geometric geographic orientation vector from the starting grid to the ending grid and the macro wind direction vector is obtained, and the cosine value, the downslope drop and the canopy shading ratio of the ending grid are fused to obtain the unidirectional airflow connectivity weight from the starting grid to the ending grid.

[0008] Furthermore, the method for obtaining the airflow transfer probability matrix includes: Each grid is taken as a starting grid, and the sum of the unidirectional connectivity weights of the airflow diverging from the starting grid to all directly adjacent grids is obtained as a safety sum term; The airflow unidirectional connectivity weights are normalized using the safety summation term to obtain the airflow transfer probability from the starting grid to each directly adjacent grid; the airflow transfer probabilities between all grids in the forest area are combined to generate the airflow transfer probability matrix.

[0009] Furthermore, the execution method of the probabilistic iterative deduction includes: In the initial round, the initial probability of the divergent source node is set to a preset first value, and the initial probability of the other ordinary grids is set to a preset second value. Each grid is treated as a grid to be analyzed. In the current iteration, the set of neighboring source grids pointing to the grid to be analyzed is obtained. Based on the distribution probability of the neighboring source grids in the previous iteration and the corresponding airflow transfer probability, the theoretical inflow total is calculated. Based on the air mass deposition and consumption probability of the grid to be analyzed, the theoretical inflow total is attenuated and deducted, and a preset source flux constant is added to the diverging source node to obtain the current calculation probability of the grid to be analyzed. A distribution probability vector is constructed based on the probability calculated for each round. When the difference between two adjacent distribution probability vectors is less than a preset tolerance threshold, or when a preset iteration limit is reached, the iteration stops, and the distribution probability corresponding to the termination of the iteration is taken as the final distribution probability of the air mass in each grid.

[0010] Furthermore, the method for obtaining the impedance attenuation calculated distance includes: For each grid: after adding the final distribution probability of the air mass to the preset anti-zero smoothing factor, calculate the natural logarithm and take the absolute value to obtain the probability logarithm penalty term; By combining the relative sinking depth with a preset sinking penalty coefficient, a sinking resistance penalty term is obtained; by combining the probability logarithmic penalty term, the sinking resistance penalty term, and a preset base distance constant, the stagnation attenuation calculated distance is obtained.

[0011] Furthermore, the method for estimating the concentration of the environmental indicator includes: Substitute the calculated resistance attenuation distance into a preset attenuation power function to obtain the comprehensive resistance cost; The ratio of the true value of the variable to be measured to the comprehensive resistance cost of each grid is calculated and used as the estimated concentration of the environmental indicator for each grid.

[0012] Furthermore, the method for controlling the fresh air control system of the semi-enclosed health and wellness facility includes: For each preset interveneable grid, when the estimated concentration of the environmental indicator is less than the preset lower limit threshold and the calculated distance of the blocking attenuation is greater than the preset cost blocking threshold, the communication address of the pre-registered associated semi-enclosed health and wellness facility within the grid is extracted. A control command containing a forced power-on action flag is generated and sent to the fresh air control system corresponding to the communication address via the communication network to control the closure of the fresh air and dehumidification circuits.

[0013] The present invention also proposes a comprehensive monitoring system for forest health and wellness environment, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the steps of the comprehensive monitoring method for forest health and wellness environment described above.

[0014] The present invention has the following beneficial effects: This invention first acquires various data of the target forest area to provide a foundation for subsequent data analysis. It then obtains the air mass deposition and consumption probability of each grid, characterizing the proportion of natural airflow carrying environmental factors that is intercepted by depressions and absorbed by condensed water vapor when passing through the target grid. Furthermore, it calculates the unidirectional connectivity weight of airflow in a two-dimensional graph network topology and generates an airflow transfer probability matrix, accurately reconstructing the non-uniform meteorological diffusion phenomenon caused by complex forest areas. Next, it performs probabilistic iterative deduction with constant source replenishment using the grid where the meteorological master station is located as the divergence source node, outputting the final air mass distribution probability of each grid and obtaining the stagnation and attenuation conversion distance of each grid. This effectively eliminates numerical measurement interference caused by planar straight-line distances and objectively reconstructs the true physical loss law during the infiltration of environmental indicators into complex terrain. Finally, based on the stagnation and attenuation conversion distance and the true value of the measured variables at the meteorological master station, it calculates the estimated concentration of environmental indicators in each grid, controls the fresh air control system of the semi-enclosed health and wellness facility, and effectively ensures the sustainable livability of the forest-based health and wellness environment. This invention combines the dew point temperature difference of each grid, the relative sinking depth, and the canopy shading ratio to generate a matrix of air mass deposition and consumption probability and airflow transfer probability. Through probability iteration, the obstruction attenuation conversion distance is obtained, eliminating the estimation distortion caused by terrain obstruction and water vapor interception, and accurately regulating the facility's fresh air control system. Attached Figure Description

[0015] Figure 1 A flowchart illustrating a comprehensive monitoring method for forest health and wellness environment provided in one embodiment of the present invention; Figure 2 This is a flowchart of a probabilistic iterative deduction execution method provided in one embodiment of the present invention. Detailed Implementation

[0016] The following description, in conjunction with the accompanying drawings, details the specific scheme of the comprehensive monitoring method and system for forest health and wellness environment provided by this invention.

[0017] Please see Figure 1 The diagram illustrates a flowchart of a comprehensive monitoring method for forest health and wellness environments according to an embodiment of the present invention, specifically including: Step S1: Obtain the relative depression depth and canopy shading ratio of each grid in the target forest area, and obtain the ambient temperature, relative humidity and dew point temperature difference of each grid at each time step within the current analysis window.

[0018] In one embodiment of the present invention, the relative depression depth and canopy shading ratio of each grid in the target forest area currently being analyzed are first obtained to provide information on the three-dimensional spatial obstruction and terrain retention attributes of the grid. Considering that the microclimate of the forest area is easily affected by instantaneous environmental noise such as gusts, it is necessary to extract stable and continuous macro-trends of meteorological changes to filter short-term numerical fluctuations. Therefore, the ambient temperature, relative humidity and dew point temperature difference of each grid at each time step within the current analysis window are obtained to provide a basis for subsequent data analysis.

[0019] In a preferred embodiment of the present invention, the system acquires the digital geographic boundary of the target forest area and divides the target forest area into a continuous two-dimensional square grid with fixed side lengths. The side length of the square grid can be set to a value between 10 meters and 50 meters (e.g., 20 meters), and the specific size can be configured according to the total area of ​​the forest area and the required spatial accuracy of the analysis; the total number of grids is recorded. After obtaining the total number of grid cells, the system assigns a unique spatial index number i to each grid cell, where the value of i ranges from 1 to 2. This establishes a benchmark for spatial analysis.

[0020] The system then sets the analysis cycle to 10-30 minutes; this example uses 20 minutes. Each analysis extracts data from the current analysis window. In this example, the analysis window length is set to 60 minutes, with each time step lasting 1 minute. Ambient temperature and relative humidity are also collected once per minute, synchronized with the time steps. This is recorded as the total number of time steps, K, and a time step index t is assigned, where t ranges from [1, K], thus establishing the time series analysis baseline. Negative oxygen ions and fine particulate matter readily combine with liquid water droplets and settle in a condensation state where relative humidity approaches saturation. Therefore, the dew point temperature difference is further obtained: For each grid, the ambient temperature and relative humidity of each time step are substituted into the preset Magnus empirical formula to obtain the dew point temperature of each time step. The dew point temperature difference at each time step is obtained based on the ambient temperature and dew point temperature at each time step.

[0021] As an example, the ambient temperature at each time step is subtracted from the dew point temperature to obtain the dew point temperature difference at each time step.

[0022] It should be noted that for grid division methods in target forest areas with irregular boundaries, existing methods such as minimum bounding rectangle combined with mask clipping can be used, which will not be elaborated further. Considering that not every grid may actually have monitoring sensors, after acquiring the measurement values, a two-dimensional spatial interpolation algorithm (exemplarily, spline interpolation algorithm) is used. Based on the geographical coordinates of auxiliary monitoring stations with measurement values, the system performs two-dimensional spatial numerical amortization extrapolation on the remaining blank grids without deployed sensors throughout the entire forest area.

[0023] In real forest environments, deep depressions and dense foliage create natural barriers that impede airflow. To quantify the impact of this three-dimensional spatial structure on air mass diffusion, the system extracts the relative depression depth and canopy shading ratio of specific grids, representing the degree of terrain depression and vegetation shading, respectively.

[0024] In a preferred embodiment of the present invention, the system receives externally imported static mapping data through a data interface. This static mapping data includes a digital elevation model (DEM) of the forest area reflecting the terrain elevation and a forest compartment distribution map reflecting the vegetation cover.

[0025] The system extracts the absolute elevation of grid i from the digital elevation model based on the geographic coordinate boundaries of the divided grids. Simultaneously, the system queries and extracts the absolute elevations of all surrounding grids that are directly adjacent to grid i in space. In this example, an eight-neighborhood model within a nine-grid layout is used, defining the eight grids centered on grid i as having a direct adjacency relationship. The system calculates the arithmetic mean of the elevation values ​​of these surrounding grids, subtracts the absolute elevation of grid i from this arithmetic mean to obtain the elevation difference, and takes the maximum value between this elevation difference and zero as the relative depression depth of grid i. relative depth of depression The larger the value, the more physically the mesh... The deeper the air mass is embedded in the surrounding terrain, the more likely it is to become physically stagnant in that area.

[0026] By forcibly zeroing out the value by taking the maximum value of zero, the system effectively eliminates the numerical interference of mountain peaks or convex terrain on the calculation of low-lying features, ensuring that the extracted parameters are used only to quantify the interception effect of the sinking depression.

[0027] When extracting vegetation cover levels, the system will use a grid. Map the spatial coordinates onto the forest compartment distribution map, query and extract the grid. The proportion of tree canopy obstruction above This proportion data was obtained in advance by extracting the Normalized Difference Vegetation Index (NDVI) from aerial imagery and combining it with a pixel-based binary model; its value strictly falls within the range of 0 to 1. (Tree canopy occlusion proportion) The larger the value, the wider the area of ​​the sky above the target grid is covered by branches and leaves, and the more severely the vertical channels for air mass convection with the outside are blocked.

[0028] It should be noted that mapping the spatial coordinates of the grid to the digital elevation model (DEM) and extracting the average value of the elevation pixels within the grid coverage area is used as the absolute elevation of the grid. NDVI combined with the pixel bisection model to calculate vegetation cover is a very mature existing technology in the field of remote sensing, and will not be elaborated here.

[0029] Step S2: Based on the change in dew point temperature difference of each grid, combined with the relative sinking depth and the canopy shading ratio, obtain the air mass deposition and consumption probability of each grid; in the two-dimensional graph network topology diagram, based on the downslope difference between adjacent grids, the spatial angle between the geometric geographical orientation vector and the macro wind direction vector between grids, and the canopy shading ratio, calculate the unidirectional airflow connectivity weight and generate the airflow transfer probability matrix.

[0030] Within the current analysis window, considering the changes in the dew point temperature difference of the grid, it reflects the dynamic falling trend of environmental water vapor approaching the condensation state and precipitating liquid water droplets. The relative sinking depth and the canopy shading ratio supplement the physical barrier and air mass retention properties of the grid in three-dimensional space. Therefore, the probability of air mass deposition and consumption of each grid is obtained in this way, which characterizes the proportion of natural airflow carrying environmental factors such as negative oxygen ions that is intercepted by depressions and swallowed by condensed water vapor when passing through the target grid. This allows for the accurate setting of node absorption parameters that reflect local attenuation characteristics in the subsequent graph network model. To simulate the physical diversion state of airflow spreading outward in complex forest terrain, a two-dimensional graph network topology is constructed to represent the spatial connectivity paths between discrete grids within the forest area. In the topology, the downslope difference between adjacent grids reflects the acceleration gain effect of gravity on the downward airflow, the spatial angle between the geometric geographic orientation vector and the macroscopic wind direction vector between grids reflects the driving strength of the downwind force on airflow transport, and the canopy shading ratio reflects the frictional retention resistance caused by dense branches and leaves to the horizontal penetration of airflow. By combining these factors, the unidirectional connectivity weight of airflow is calculated, characterizing the difficulty of airflow overcoming terrain and vegetation obstacles in different directions. Furthermore, an airflow transfer probability matrix is ​​generated, accurately reproducing the non-uniform meteorological diffusion phenomenon caused by complex forest areas.

[0031] As a preferred approach, considering that particles in the ambient air mass are easily attached to condensed water droplets and settle, and that the depressions and dense foliage will hinder the escape of particles, for each grid: a straight line is fitted with the time step as the x-axis and the corresponding dew point temperature difference value as the y-axis to obtain the time derivative of the dew point temperature difference within the current analysis window (since normalization will be performed later, the choice of units here has no impact). Then, the rate of temperature difference reduction is set based on the time derivative; Specifically, if the time derivative corresponding to the grid is less than zero, it indicates that the difference between the ambient temperature and the dew point temperature is decreasing and the air is approaching condensation saturation. In this case, the absolute value of the time derivative is extracted as the rate of decrease in temperature difference. If the time derivative is not less than zero, it indicates that the difference between the ambient temperature and the dew point temperature is widening or remaining constant, and water vapor does not tend to accelerate condensation. In this case, the rate of temperature difference reduction is set to zero. The greater the rate of temperature difference reduction, the easier it is for ambient water vapor to condense and precipitate into liquid water droplets, which in turn makes it easier to absorb negative oxygen ions and fine particulate matter in the environment.

[0032] Further integrate the rate of temperature difference reduction, relative subsidence depth, and canopy shading ratio to obtain a comprehensive baseline for settlement consumption; As an example, to avoid direct fusion of physical quantities with different dimensions, normalization is first performed within each data dimension: Under the current analysis window, the global maximum values ​​of the temperature difference reduction rate and relative sinking depth of all grids are extracted respectively. The temperature difference reduction rate and relative sinking depth of each grid are divided by the global maximum value in their respective dimensions to perform maximum value normalization. The normalized temperature difference reduction rate, the normalized relative sinking depth, and the canopy shading ratio are multiplied together, and the result of the multiplication is used as the comprehensive base of settlement consumption for the corresponding grid.

[0033] Through multiplication, the system can quantify the combined amplification effect of topographic depressions, canopy shading, and high humidity condensation, ensuring that the overall base of grid settling consumption only shows a significant nonlinear increase when multiple blocking conditions are met simultaneously. This objectively restores the physical characteristics of the strong interception of air masses by specific deteriorating environments in complex forest areas.

[0034] If the global maximum value of a certain data dimension is 0, then the normalization result of all grids is set to 0. Simultaneously, to prevent individual data points from being 0 during multiplication (e.g., a grid with no canopy obstruction), which could cause the overall settlement loss base of that grid to directly become zero and thus mask the objective existence of other physical obstruction factors such as depressions or high humidity, a preset feature offset constant is added to each data point before multiplication. The feature offset constant is typically two or more orders of magnitude smaller than the normalization upper limit (i.e., the value 1) (e.g., a value of 0.01), and its impact on the macroscopic numerical distribution gradient of the overall settlement loss base is relatively small.

[0035] Finally, after normalizing the overall sedimentation consumption base with the largest global sedimentation consumption base, the sedimentation consumption probability of the corresponding grid is obtained by combining it with the preset attenuation safety limit coefficient.

[0036] The preset attenuation safety limit coefficient is set to 0.9. The comprehensive settling consumption base of each grid is divided by the global maximum comprehensive settling consumption base for normalization. To prevent the extreme case where the air mass settling consumption probability equals 1 after normalization (which would lead to an environmental retention coefficient in subsequent iterations, thus cutting off local diffusion paths), the product of the normalization result and the preset attenuation safety limit coefficient is used as the air mass settling consumption probability for the corresponding grid. The value of the preset attenuation safety limit coefficient is set according to the overall terrain closure degree of the forest area: for forest areas with open terrain and unobstructed airflow, it can be set to 0.7~0.8; for forest areas with complex terrain and widespread local obstruction, it can be set to 0.85~0.95.

[0037] Similarly, when the global maximum sedimentation consumption baseline is 0, the air mass sedimentation consumption probability is set to zero.

[0038] Preferably, in one embodiment of the present invention, for any two grids that are directly adjacent in space, it is assumed that one grid is the starting grid i and the other grid is the ending grid j. The relationship between the starting and ending points of the two grids can be interchanged, and the adjacent grid pairs are traversed and analyzed one by one. The difference between the absolute elevation of the starting grid and the elevation of the ending grid is taken as the downslope drop. A positive value indicates a downhill slope, and a negative value indicates an uphill slope, to represent the gravity-induced gain. The system obtains the geometric geographic azimuth vector pointing from grid i to grid j (the vector pointing from the center point of grid i to the center point of grid j), and obtains the macroscopic wind direction vector of the meteorological station at the latest time step. Then, considering that the smaller the spatial angle, the larger the cosine value, the system obtains the cosine value of the spatial angle between the geometric geographic azimuth vector from the starting grid to the ending grid and the macroscopic wind direction vector. , indicating the tailwind drive gain; Further integrate the cosine value, downhill drop, and canopy shading ratio of the endpoint grid. Obtain the unidirectional connectivity weight of the airflow from the starting grid to the ending grid.

[0039] As an example, the system pre-configures the drop gain constant. (The value in this example is) Wind direction gain constant (The value in this example is) ) and resistance penalty constant (The value in this example is) ).

[0040] From the grid Flow to grid unidirectional airflow connectivity weight The calculation formulas include: In the formula, This represents an exponential function with the natural constant e as the base.

[0041] Among them, the drop gain term The weighting gain increases with the downhill drop, used to enhance the downward flow of air; wind direction gain term. The term increases as the angle between the grid's geometric connectivity direction and the downwind direction decreases, reflecting the promoting effect of the downwind on flow; the canopy penalty term... The greater the occlusion ratio of the endpoint grid, the stronger the negative offset, which is used to reduce the weight of airflow towards dense canopy areas. Finally, the mapping and fusion are performed using an exponential function, which can effectively amplify the nonlinear promoting effect of various physical gain conditions on airflow penetration preference, while ensuring that the output connectivity weight is always greater than zero, providing a basis for subsequent transition probability calculations.

[0042] It should be noted that, considering that when there is no wind or a light breeze (calm wind) in nature, the airflow does not have a clear macroscopic wind direction vector, the system sets a calm wind judgment threshold (0.2 meters per second in this example). When the meteorological station detects that the average wind speed over the most recent time steps (such as the most recent 5 time steps) is less than the calm wind judgment threshold, the system directly determines that the current state is calm, skips the angle calculation, and directly assigns the wind direction gain constant to 0, thus shielding the wind from its additive effect on airflow diffusion.

[0043] It should be noted that the above-mentioned calm wind judgment threshold, drop gain constant, wind direction gain constant and drag penalty constant are weight adjustment factors for balancing the proportion of the influence of various environmental drag factors. Their specific values ​​are usually pre-calibrated based on the typical terrain roughness of the target forest area and the fitting analysis of historical wind measurement data and the diffusion trajectory of real environmental factors. In other embodiments of the present invention, the implementer can calibrate them according to the actual scenario.

[0044] Preferably, in one embodiment of the present invention, considering the conservation of the total physical flow rate when the airflow diverges and spreads to the surrounding pores, the relative connectivity weights must be converted into strict mathematical transition probabilities in order to perform subsequent pure spatial steady-state topology deduction. Therefore, each grid is taken as the starting grid, and the sum of the unidirectional connectivity weights of the airflow diverging from the starting grid to all directly adjacent grids is obtained as the safety sum term. By normalizing the unidirectional connectivity weights of each airflow using a safety summation term, the airflow transfer probability from the starting grid to each directly adjacent grid (ending grid) is obtained. Specifically, the unidirectional connectivity weights of each airflow are used as the numerator, the total safety term as the denominator, and the ratio of the fractions as the airflow transfer probability of each endpoint grid, characterizing the actual flow distribution ratio of airflow along multiple bifurcation spatial paths under complex forest topography. Among these, due to... It was through The value is derived from the mapping, so it must not be 0.

[0045] Finally, all airflow transfer probabilities between grids within the forest area are combined to generate an airflow transfer probability matrix.

[0046] Specifically, each grid within the forest area is then used as the row index and column index, with the number corresponding to the starting grid as the row index and the number corresponding to the directly adjacent ending grid as the column index. The obtained airflow transfer probability from the starting grid to the ending grid is filled into the corresponding row and column intersection position (for grids that are not directly adjacent in space, the corresponding intersection position element is set to zero), and the main diagonal elements are set to zero, generating an airflow transfer probability matrix with directional uniqueness.

[0047] Step S3: Using the grid where the main meteorological station is located as the divergence source node, based on the probability matrix of air mass deposition and consumption and airflow transfer, perform probability iterative deduction with constant source replenishment, and combine the relative sinking depth to obtain the stagnation attenuation conversion distance of each grid.

[0048] When airflow diffuses within forest areas, it not only continues to spread into surrounding pores but also experiences in-situ consumption as it passes through low-lying, high-humidity grids. Considering that the meteorological master station serves as the benchmark for accurate environmental indicator observation data, it needs to be used as a reference anchor point for spatial attenuation extrapolation. Therefore, using the grid where the meteorological master station is located as the divergence source node, a probabilistic iterative extrapolation with constant source replenishment is performed based on the probability matrix of air mass deposition and airflow transfer. This outputs the final distribution probability of air masses in each grid, accurately quantifying the retention ratio of micro-airflows carrying environmental factors after experiencing multi-path diversion and water vapor interception in depressions. Then, combined with the relative sinking depth, the obstruction attenuation conversion distance of each grid is obtained. This unifies the retention status of environmental factors and the three-dimensional terrain obstruction attributes into a comprehensive spatial attenuation cost, effectively eliminating numerical measurement interference caused by planar straight-line distances. It objectively restores the real physical loss law during the infiltration of environmental indicators into the complex terrain, providing a basis for subsequent accurate extrapolation of environmental indicators and control systems.

[0049] Preferably, in one embodiment of the present invention, please refer to Figure 2 The diagram illustrates a flowchart of a probabilistic iterative deduction execution method provided by an embodiment of the present invention, specifically including: Step S301: In the initial round, the initial probability of the divergent source node is set to a preset first value, and the initial probability of the other ordinary grids is set to a preset second value.

[0050] First, select the meteorological master station closest to the currently operable grid, and mark the grid containing the meteorological master station as the divergence source node; then, the system performs initialization operations: setting the iteration round symbol as... Let k=0 to generate the initial probability vector. The first preset value is 1, and the second preset value is 0. The initial probability of the divergence source node is set to 1, which is intended to serve as the absolute reference source for the outward diffusion of environmental indicators. The other ordinary grids are set to 0, which is intended to simulate the initial blank state of the remaining space before diffusion begins, where there is no contribution from external materials. This ensures that the probability values ​​derived from each subsequent grid strictly and uniquely originate from the divergence source node.

[0051] After initialization, set k=1 and enter the iterative deduction stage for the dynamic diffusion process of each grid.

[0052] Step S302: Each grid is taken as the grid to be analyzed. In the current iteration, the set of adjacent source grids pointing to the grid to be analyzed is obtained. Based on the distribution probability of the adjacent source grids in the previous iteration and the corresponding airflow transfer probability, the theoretical inflow total is calculated. Based on the air mass deposition consumption probability of the grid to be analyzed, the theoretical inflow total is attenuated and deducted, and a preset source flux constant is added to the divergent source nodes to obtain the calculation probability of the grid to be analyzed in the current iteration.

[0053] In the kth round, each grid is taken as the grid to be analyzed (such as grid i). Based on the spatial location of the grid to be analyzed, the system obtains the grids directly adjacent to it as the set of neighboring source grids. The neighboring source grids mean that in the airflow transfer probability matrix, the grid to be analyzed is the end grid and the neighboring grids are the starting grids. The airflow transfer probability is from the neighboring grids to the grid to be analyzed. As an example, the system extracts the air mass deposition and consumption probability of grid i. Calculate 1- As an environmental retention coefficient This represents the proportion of airflow that is not trapped by depressions or absorbed by water vapor condensation after passing through the grid; for each grid n in the adjacent source grid set, the system extracts its distribution probability from the previous round. And the probability of airflow transfer from grid n to grid i. ,Will and The product of these is used as the inflow for a single path; Then, the inflow of single paths of all grids in the adjacent source grid set is summed to obtain the theoretical total inflow of the grid to be analyzed in this round; the product of the theoretical total inflow and the environmental retention coefficient is used as the residual distribution probability after environmental absorption deduction. It should be noted that, in order to characterize the persistent concentration benchmark at the meteorological master station, the system supplements the calculation probability of the grid corresponding to the divergent source node in the current round with a preset source flux constant. That is, after calculating the product of the theoretical total inflow and the environmental retention coefficient, the preset source flux constant is added, and the result serves as the final calculation probability of the divergent source node in the current round. In this example, the preset source flux constant, as an algorithm hyperparameter to maintain iterative stability, is manually set by the system operation and maintenance personnel based on the system's early debugging experience, and is set to 0.15.

[0054] Step S303: Construct a distribution probability vector based on the calculated probabilities of all current rounds in each round. When the difference between two adjacent distribution probability vectors is less than a preset tolerance threshold, or when the preset iteration limit is reached, stop the iteration and use the distribution probability corresponding to the termination of the iteration as the final distribution probability of the air mass in each grid.

[0055] As an example, the probability distribution vector for each round is constructed by calculating the probabilities of all current rounds in a uniform grid index order. In order to prevent the probability values ​​of local nodes from exceeding the upper limit due to iteration with a preset source injection constant, after each iteration, the system uses the maximum current round calculation probability as the denominator and the current round calculation probability of each grid as the numerator, and normalizes them to construct a distribution probability vector.

[0056] After constructing the vector, the probability corresponding to the divergent source node is reset to the preset first value to obtain the distribution probability vector of the current iteration. The reset operation is used to ensure that the divergent source node always maintains the highest potential energy state in the whole iteration cycle, and avoids the source node numerical privilege being diluted or reversed due to global normalization, thereby ensuring that the physical gradient field of the air mass diffuses unidirectionally from the meteorological station to the surrounding grid is not destroyed.

[0057] When the probability distribution vectors of adjacent rounds are small, it indicates that the numerical alternation of diffusion and absorption processes has become stable. The system determines that the algorithm has reached dynamic convergence equilibrium and terminates the iteration. At the same time, a preset iteration upper limit is set to avoid the system getting stuck in a loop. The probability distribution of the latest iteration round at the time of termination is taken as the final probability distribution of air masses in each grid.

[0058] In this example, the Euclidean distance between the probability distribution vectors of two adjacent rounds is calculated, and a preset tolerance threshold is set. The value is 0.001. The smaller the Euclidean distance, the smaller the difference between the two vectors. When the Euclidean distance is less than... When the iteration ends, the iteration is terminated; the preset iteration limit is set to 1000 times.

[0059] It should be noted that the meteorological master station is a benchmark monitoring point equipped with high-precision environmental sensors and capable of outputting the true values ​​of the variables to be measured in real time (such as the concentration of negative oxygen ions or fine particulate matter). The interventionist grid is the target measurement grid with semi-enclosed health and wellness facilities and equipped with a controllable fresh air control system. If there are multiple meteorological master stations and multiple interventionist grids, each interventionist grid is analyzed one by one. During the analysis, the grid containing the meteorological master station closest to the interventionist grid can be selected as the divergence source node, and the data of the corresponding meteorological master station can be selected for analysis.

[0060] In other embodiments of the present invention, the implementer may flexibly adjust the preset source injection constant, preset iteration upper limit and tolerance threshold according to the total number of spatial grids in the target forest area and the required timeliness of the measurement response, so as to achieve a balance between algorithm convergence accuracy and computing power consumption.

[0061] Furthermore, obtaining the reduced distance of the stagnation attenuation is a preferred method: For each grid: Considering that the final distribution probability of the air mass is between 0 and 1, the smaller the value, the more difficult it is for the airflow to reach. The absolute value of the natural logarithm function in the interval between 0 and 1 will increase non-linearly and sharply as the input value decreases. This can map a very small retention probability to a very large drag penalty. At the same time, in order to prevent the logarithmic operation from becoming undefined or collapsing when the distribution probability approaches zero, a preset zero-prevention smoothing factor is introduced. Therefore, after adding the final distribution probability of the air mass to the preset zero-prevention smoothing factor, the natural logarithm is calculated and the absolute value is taken to obtain the probability logarithmic penalty term, which accurately quantifies the non-linear distance penalty caused by the severe obstruction during the diffusion of the air mass. Considering that the relative depression depth represents the degree of spatial concavity and obstruction of the grid into the surrounding terrain, and has a length dimension, it cannot be directly added to the dimensionless probability logarithmic penalty term. Therefore, the relative depression depth and the preset depression penalty coefficient are combined to obtain the depression resistance penalty term, which eliminates the dimensional conflict caused by direct calculation and quantifies the spatial isolation cost caused by the low-lying terrain. Finally, by fusing the probability logarithmic penalty term, the sinking resistance penalty term, and the preset base distance constant, the stagnation attenuation reduced distance is obtained.

[0062] As an example: Preset anti-zero smoothing factor The sinking penalty coefficient is set to 0.5. Its physical meaning represents the equivalent increase in dimensionless resistance penalty weight for every 1 meter of downward depression in the terrain. This value is manually preset by the implementers based on the equivalent comparison experience of the concentration decay in the depression and the distance decay in the flat area in the forest area.

[0063] The product of the relative sinking depth and the preset sinking penalty coefficient is used as the sinking resistance penalty term; the probability logarithmic penalty term, the sinking resistance penalty term and the base distance constant are summed to obtain the stagnation attenuation reduced distance.

[0064] The preset base distance constant is set to 1, which serves as a basic deviation to prevent the final distance from being too small. It is fused by linearly summing the logarithmic penalty representing the multipath probability decay, the terrain resistance representing the terrain barrier, and the base constant representing the basic physical deviation, to reflect the comprehensive equivalent spatial distance cost paid by the airflow for diffusion and infiltration in complex forest habitats.

[0065] Step S4: Based on the attenuation reduction distance and the true value of the measured variable from the meteorological station, estimate the concentration of environmental indicators in each grid and control the fresh air control system of the semi-enclosed health and wellness facility.

[0066] Considering that conventional planar distance measurement ignores terrain obstruction, leading to an overestimation of environmental factor concentrations, and that the real data measured by the meteorological station needs to undergo complex spatial loss before penetrating to the local area, the concentration of environmental indicators in each grid is estimated based on the obstruction attenuation conversion distance and the true value of the variables to be measured at the meteorological station, thus quantifying the local real microclimate level after physical obstruction and water vapor consumption. Finally, the fresh air control system of the semi-enclosed health and wellness facility effectively ensures the continued livability of the under-forest health and wellness environment.

[0067] Preferably, in one embodiment of the present invention, considering that the diffusion and dissipation of environmental factors in the space medium usually follow a nonlinear geometric decay law, the concentration dilution caused by the hindrance condition is not a simple linear relationship. Based on this, the obstruction and attenuation conversion distance is substituted into the preset attenuation power function to obtain the comprehensive resistance cost, thus restoring the true scarcity characteristics of natural breezes being unable to deliver high-concentration factors to the bottom of the depression due to obstruction. As an example, the preset decaying power function is: p is the preset attenuation power exponent, which is pre-calibrated based on the historical measured concentration attenuation fitting curve of the target forest area and is positively correlated with the average canopy density of the forest area (for example, a value of 2.0). Then, the ratio of the true value of the variable to be measured to the comprehensive resistance cost of each grid is calculated, and the concentration is estimated as an environmental indicator for each grid.

[0068] In this example, the true value of the variable to be measured refers to the negative oxygen ion concentration value measured in real time by the meteorological station. In other embodiments of the present invention, the true value of the variable to be measured may also be other indicator parameters reflecting the environmental status of the forest area, such as the air quality index, and the circuit control rules shall be adjusted accordingly.

[0069] Furthermore, for each preset interveneable grid: when the estimated concentration of environmental indicators is less than the preset lower limit threshold and the blocked attenuation conversion distance is greater than the preset cost blocking threshold, it indicates that the grid not only lacks current environmental health care factors, but also that the connection path for natural airflow to transport and spread to the area has been severely blocked, losing the ability to self-recover through natural breezes, and has entered a typical microclimate stagnation and deterioration state. At this time, the communication addresses of the pre-registered and associated semi-enclosed health care facilities in the grid are extracted. A control command containing a forced power-on action flag is generated and sent to the fresh air control system corresponding to the communication address via the communication network. This controls the closure of the fresh air and dehumidification circuits, and uses manual mechanical power to forcibly accelerate the convection and replacement of stagnant indoor air and the dissipation of high humidity moisture.

[0070] In order to avoid energy waste caused by continuous operation of equipment, when the system determines in subsequent time steps that the external natural supply capacity of the area where the grid is located has recovered, that is, when the estimated concentration of environmental indicators rises to a level greater than or equal to the lower limit threshold of the concentration, and the resistance attenuation conversion distance falls to a level less than or equal to the cost blocking threshold, it indicates that the natural environment has regained the conditions for health and wellness. At this time, a control command containing a power outage stop action flag is generated and issued to control the disconnection of the fresh air and dehumidification circuits, so that the facility can return to a natural ventilation state.

[0071] It should be noted that the preset concentration lower limit threshold is statically set based on the benchmark value (such as 1000 cells / cm³) that is beneficial to the human body in forestry or health standards; the preset cost blocking threshold is set as the arithmetic mean of the blocking and attenuation distance of all grids in the entire forest area under historical windless weather conditions.

[0072] An embodiment of the present invention also provides a comprehensive monitoring system for forest health and wellness environment. The system includes a memory, a processor, and a computer program. The memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can implement the comprehensive monitoring method for forest health and wellness environment described in steps S1-S4.

[0073] In summary, to address the technical problem in existing technologies where distorted environmental concentration estimations due to neglecting topographic obstruction and water vapor consumption affect the regulation of fresh air systems in semi-enclosed health and wellness facilities, this invention aims to provide a comprehensive monitoring method and system for forest health and wellness environments. This invention first acquires various data from the target forest area; further, it acquires the air mass deposition and consumption probability of each grid; further, it calculates the unidirectional connectivity weight of airflow in a two-dimensional graph network topology diagram and generates an airflow transfer probability matrix; further, it performs probabilistic iterative deduction with constant source replenishment using the grid where the meteorological master station is located as the divergence source node, outputting the final air mass distribution probability of each grid and acquiring the obstruction attenuation conversion distance of each grid; finally, based on the obstruction attenuation conversion distance and the true values ​​of the measured variables at the meteorological master station, it estimates the concentration of environmental indicators in each grid and controls the fresh air control system of the semi-enclosed health and wellness facility. This invention combines the dew point temperature difference of each grid, the relative sinking depth, and the canopy shading ratio to generate a matrix of air mass deposition and consumption probability and airflow transfer probability. Through probability iteration, the obstruction attenuation conversion distance is obtained, eliminating the estimation distortion caused by terrain obstruction and water vapor interception, and accurately regulating the facility's fresh air control system.

Claims

1. A comprehensive monitoring method for forest health and wellness environments, characterized in that, The method includes: Obtain the relative depression depth and canopy shading ratio of each grid in the target forest area, and obtain the ambient temperature, relative humidity and dew point temperature difference of each grid at each time step within the current analysis window; Based on the change in dew point temperature difference of each grid, combined with the relative sinking depth and the canopy shading ratio, the air mass deposition and consumption probability of each grid is obtained; in the two-dimensional graph network topology diagram, based on the downslope difference between adjacent grids, the spatial angle between the geometric geographical orientation vector and the macro wind direction vector between grids, and the canopy shading ratio, the airflow unidirectional connectivity weight is calculated, and an airflow transfer probability matrix is ​​generated. Using the grid where the meteorological main station is located as the divergence source node, based on the air mass deposition and consumption probability and the airflow transfer probability matrix, a probability iterative deduction with constant source replenishment is performed to output the final distribution probability of air masses in each grid, and combined with the relative sinking depth, the stagnation attenuation conversion distance of each grid is obtained. Based on the attenuation reduction distance and the true value of the measured variable at the meteorological station, the concentration of environmental indicators in each grid is estimated, and the fresh air control system of the semi-enclosed health and wellness facility is controlled.

2. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The method for obtaining the dew point temperature difference includes: For each grid: Substitute the ambient temperature and relative humidity at each time step into the preset Magnus empirical formula to obtain the dew point temperature at each time step; based on the ambient temperature and dew point temperature at each time step, obtain the dew point temperature difference at each time step.

3. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The method for obtaining the probability of air mass deposition includes: For each grid cell: obtain the time derivative of the dew point temperature difference within the current analysis window; set the temperature difference reduction rate based on the time derivative; By combining the temperature difference reduction rate, the relative subsidence depth, and the canopy shading ratio, a comprehensive baseline for settlement consumption is obtained. After normalizing the overall deposition consumption base with the largest global deposition consumption base, the deposition consumption probability of the corresponding grid is obtained by combining it with a preset attenuation safety limit coefficient.

4. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The method for obtaining the unidirectional airflow connectivity weight includes: For any two grids that are directly adjacent in space, assume that one grid is the starting grid and the other grid is the ending grid; The difference between the absolute elevation of the starting grid and the elevation of the ending grid is taken as the downslope drop; the cosine value of the spatial angle between the geometric geographic orientation vector from the starting grid to the ending grid and the macro wind direction vector is obtained, and the cosine value, the downslope drop and the canopy shading ratio of the ending grid are fused to obtain the unidirectional airflow connectivity weight from the starting grid to the ending grid.

5. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The method for obtaining the airflow transfer probability matrix includes: Each grid is taken as a starting grid, and the sum of the unidirectional connectivity weights of the airflow diverging from the starting grid to all directly adjacent grids is obtained as a safety sum term; The airflow unidirectional connectivity weights are normalized using the safety summation term to obtain the airflow transfer probability from the starting grid to each directly adjacent grid; the airflow transfer probabilities between all grids in the forest area are combined to generate the airflow transfer probability matrix.

6. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The execution method of the probabilistic iterative deduction includes: In the initial round, the initial probability of the divergent source node is set to a preset first value, and the initial probability of the other ordinary grids is set to a preset second value. Each grid is treated as a grid to be analyzed. In the current iteration, the set of neighboring source grids pointing to the grid to be analyzed is obtained. Based on the distribution probability of the neighboring source grids in the previous iteration and the corresponding airflow transfer probability, the theoretical inflow total is calculated. Based on the air mass deposition and consumption probability of the grid to be analyzed, the theoretical inflow total is attenuated and deducted, and a preset source flux constant is added to the diverging source node to obtain the current calculation probability of the grid to be analyzed. A distribution probability vector is constructed based on the probability of all current rounds in each round. When the difference between the distribution probability vectors of two adjacent rounds is less than a preset tolerance threshold, or when the preset iteration limit is reached, the iteration stops, and the distribution probability corresponding to the termination of the iteration is taken as the final distribution probability of the air mass in each grid.

7. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The method for obtaining the attenuation reduction distance includes: For each grid: after adding the final distribution probability of the air mass to the preset anti-zero smoothing factor, calculate the natural logarithm and take the absolute value to obtain the probability logarithm penalty term; By combining the relative sinking depth with a preset sinking penalty coefficient, a sinking resistance penalty term is obtained; by combining the probability logarithmic penalty term, the sinking resistance penalty term, and a preset base distance constant, the stagnation attenuation calculated distance is obtained.

8. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The methods for estimating the concentration of the environmental indicators include: Substitute the measured resistance attenuation distance into a preset attenuation power function to obtain the comprehensive resistance cost; The ratio of the true value of the variable to be measured to the comprehensive resistance cost of each grid is calculated and used as the estimated concentration of the environmental indicator for each grid.

9. The method for comprehensive monitoring of forest health and wellness environment according to claim 1, characterized in that, The method for controlling the fresh air control system of the semi-enclosed health and wellness facility includes: For each preset interveneable grid, when the estimated concentration of the environmental indicator is less than the preset lower limit threshold and the calculated distance of the blocking attenuation is greater than the preset cost blocking threshold, the communication address of the pre-registered associated semi-enclosed health and wellness facility within the grid is extracted. A control command containing a forced power-on action flag is generated and sent to the fresh air control system corresponding to the communication address via the communication network to control the closure of the fresh air and dehumidification circuits.

10. A comprehensive monitoring system for forest health and wellness environment, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the comprehensive monitoring method for forest health and wellness environment as described in any one of claims 1 to 9.

Images