CN113033035B - Dynamic simulation method, system and device for pollutant diffusion area - Google Patents

Abstract

The invention discloses a dynamic simulation method, system and device for a pollutant diffusion area, including selecting a habitat to be estimated, obtaining satellite data of the habitat area, and constructing an initial sample area set based on the satellite data; Historical pollutant information constructs a reachable set of pollutants over time. Since the velocity set at each point in the area can establish a functional relationship with the position information of the point, using the obtained convex combination coefficient, the velocity set at the representative point can be used to predict the velocity at all points in the area through convex combination Set; add the time factor to estimate the dynamic spread range of the forest fire reachable set in the habitat area over time. The introduction of differential inclusion in the present invention is to better predict the spread range of fire in different directions over time. At the same time, the present invention introduces the idea of finite element, which solves the problem of too much calculation of velocity sets at all points in the region in the actual prediction calculation work.

Description

Method, system and device for dynamic simulation of pollutant diffusion area

technical field

The invention relates to the technical field of computational mathematics, in particular to a dynamic simulation method, system and device for a pollutant diffusion area.

Background technique

On a global scale, natural disasters occur frequently, causing major damage to the living environment of human beings and natural ecosystems. Among them, the annual forest fires bring devastating disasters to humans and other creatures. Large-scale disasters such as forest fires Fires are difficult to extinguish manually, and containment requires the effective establishment of barriers to protect human and animal habitats. Therefore, how to construct the fire isolation zone with the minimum speed has become the top priority of the current research related to forest fire control. In order to protect the habitats of humans and wild animals from natural disasters, it is necessary to construct a fire isolation zone as quickly as possible, and the premise of constructing the isolation zone is to predict the speed and direction of fire spread as accurately as possible.

The existing fire spread control strategies show that many algorithms are carried out under the condition that the habitat satisfies the convex set property and the default fire spread speed remains constant in all directions, but in reality, many habitats are non-linear. convex, then these masking strategies are not generally feasible.

Contents of the invention

In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is to provide a method, system and device for dynamic simulation of the pollutant diffusion area to solve the deficiencies of the prior art.

In order to achieve the above object, the present invention provides a dynamic simulation method for pollutant diffusion area, including:

Step 1. Select the habitat to be estimated, obtain the satellite data of the habitat area, and construct the initial sample area set based on the satellite data;

Step 2. Construct a forest fire reachable set based on the historical fire information of the habitat;

Step 3. Based on the finite element idea, select effective discrete representative points in the habitat area to triangulate the entire area, and divide the area into multiple small triangular areas with convex set properties. These discrete representative points are the vertices on these small triangular areas, because each small triangular area is a convex set, so the three vertices on the small triangular area and all the discrete points in the small triangle can be used to simultaneously establish the convex combination formula, Find the convex combination coefficient.

Step 4. Using the convex combination coefficient obtained in step 3, use the velocity sets at the three vertices on the small triangular area to predict the velocity sets at all points in the small triangular area; thus the velocities at the finite representative discrete points can be used The set predicts the set of velocities at all points in the entire region. So far, the differential inclusion spread discrete model has been established. The differential inclusion modulus model estimates velocities in all directions at various locations throughout the habitat area;

Step 5. Adding the time factor to estimate the dynamic spread range of the forest fire reachable set in the habitat area over time.

Further, the satellite data of the habitat area at least includes habitat longitude and latitude data, surface temperature data, surface temperature difference data between day and night, surface sulfide data, surface vegetation data, and surface infrared remote sensing image data.

Further, the step 2 constructs a forest fire reachable set based on the historical fire information of this habitat, specifically: the sulfide concentration of this habitat is high, and the surface temperature is high and the temperature difference between day and night is less than a certain threshold, and finally calculate the The fire area of the habitat in a certain period of time is taken as a reachable set.

Further, in the step 3, the whole habitat area is triangulated with a limited number of discrete representative points, and then the speed sets at all points in the whole area are predicted using the speed sets at the limited number of representative points. Specifically:

Initialize the habitat area, select n discrete points in the habitat area, the point set is x=(x ₁ ,x ₂ ,…,x _n ), here, the velocity set at point x _k is defined as F(x _k ) , 1≤k≤n; then select m discrete representative points x _i , 1≤i≤m, m≤n. Triangulate the habitat area.

Further, the step 4 uses the velocity sets at m discrete representative points to predict the velocity set at each point in the entire habitat area, and establishes a differential inclusion spread discrete model. In the differential inclusion model, the velocity of each location point in each direction in the entire habitat area is estimated, specifically: select the most effective m discrete representative points x _i in the habitat area, 1≤i≤m, use these m discrete The representative point triangulates the entire area to obtain p small triangular areas d with convex set properties, and the set of small triangular areas is

d=(d ₁ ,d ₂ ,d ₃ ,...,d _p ),

Use the vertices on these small triangular areas and all discrete points in the small triangular area to coordinate convex combination formulas, x _i is the point in the qth small triangular area, x _q1 , x _q2 , x _q3 are the small triangles The three vertices on the area, the specific formula is as follows:

Calculate the convex combination coefficients α ₁ , α ₂ , α ₃ . Since the discrete points in the small triangle can be represented by a convex combination of three vertices on the small triangle area, by analogy, the velocity set at the discrete point in the small triangle can also be expressed by the velocity at the three vertices on the small triangle area set for convex combination prediction representation. The formula is:

Further, the step 5 adds the time factor to estimate the range of the dynamic spread of the forest fire reachable set in the habitat area over time, specifically: find the velocity set of each fire point on the forest fire initial reachable boundary, and introduce the time Factor, calculate the trajectory of each fire point in different directions on the boundary of the initial set of forest fires, find the trajectory of all fire points after time t, and then combine these motion trajectories to obtain the forest fire after time t The new extent of the spread is the new reachable set.

The present invention also provides a dynamic simulation system for pollutant diffusion area, including:

The area set acquisition module is used to select the habitat to be estimated, obtain the satellite data of the habitat area, and construct the initial sample area set based on the satellite data;

The reachable set acquisition module is used to construct the forest fire reachable set based on the historical fire information of the habitat;

The regional division module, based on the finite element idea, selects a limited number of discrete representative points to triangulate the entire habitat area, divides the entire area into multiple small triangular areas, and each small triangular area is a convex set. These discrete representative points These are the vertices on these small triangular areas.

The fire speed prediction module uses the three vertices on the small triangular area and the simultaneous convex combination formula of all discrete points in the small triangular area to obtain the convex combination coefficient, because the velocity set can be expressed as a function related to the discrete points, thus By analogy, the obtained convex combination coefficient can be used to predict the velocity sets at all discrete points in the small triangular area by convex combination with the velocity sets at the three vertices in the small triangular area. The above steps have established a differential inclusion spread discrete model.

The fire spread estimation module, by adding the time factor, estimates the range of the forest fire reachable set dynamic spread over time in the habitat area.

An apparatus, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, and implementing the steps of the above method when the processor executes the computer program.

The beneficial effects of the present invention are:

The present invention introduces a mathematical model of differential inclusion to simulate the forest fire spread model, and the purpose of introducing differential inclusion is to better predict the range of fire spread in different directions over time.

At the same time, the present invention introduces the idea of finite element again, and divides the fire spread area into a plurality of small triangular areas. Since the small triangular areas are convex sets, the three vertices on the small triangular area and the three vertices in the small triangular area can be connected simultaneously. The point coordinate convex combination formula of all points is used to obtain the convex combination coefficient. Then, using the convex combination coefficient, the velocity sets at all points in the area of the small triangle are predicted by the velocity sets at the three vertices of the small triangle. The vertices on all these small triangular areas form a limited number of discrete representative points, and use the speed sets at these limited number of discrete points to predict the speed sets at all points in the entire area. Use a limited number of points to make predictions, and the prediction effect is not good.

In addition, the method of dividing the area is to use the triangulation method to solve the problem of non-convex areas. The small triangular areas divided by the triangulation are all convex, and the three vertices of the small triangles can be used for convex combination. Predict all point sets inside the small triangle.

The idea, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings, so as to fully understand the purpose, features and effects of the present invention.

Description of drawings

Fig. 1 is a schematic diagram of selecting n discrete points in the habitat area of the present invention.

Fig. 2 is a schematic diagram of triangulation of a limited number of discrete points in the habitat area according to the present invention.

Fig. 3 is a schematic diagram of the interior and exterior of the triangle after triangulation of the fire point of the present invention.

Fig. 4 is an algorithm flow chart of the present invention.

Fig. 5 is a data flow diagram of the present invention.

Fig. 6 is an application architecture diagram of the present invention.

detailed description

The focus of the invention is to establish a discrete model of differential inclusion based on the finite element idea and to design a numerical simulation method for the dynamic change of the fire area based on Minkowskiaddition. Specifically, the selected and predicted habitat area is firstly divided into a limited number of small areas, and the method of dividing the area is to use the triangulation method, the purpose is to solve the problem of whether the area is non-convex. The small triangle area is convex, so the convex combination of the three vertices of the small triangle can be used to predict all point sets in the small triangle. Then based on the differential inclusion, the spreading speed of each position point in the area in a direction is established. Finally, Minkowski addition is used to simulate the model of forest fire reachable set dynamic spread. For the convenience of description, the present invention selects forest fires as a specific example of pollutant diffusion; based on the finite element idea, a limited number of the most representative discrete points in the region are selected to triangulate the region, and each small triangle after triangulation The area is a convex set, and the point coordinate convex combination formula of the representative point and all points in the area is established to obtain the convex combination coefficient. Since the velocity set at each point in the area can establish a functional relationship with the position information of the point, using the obtained convex combination coefficient, the velocity set at the representative point can be used to predict the velocity at all points in the area through convex combination Set; add the time factor to estimate the dynamic spread range of the forest fire reachable set in the habitat area over time. The introduction of differential inclusion in the present invention is to better predict the spread range of fire in different directions over time. At the same time, the present invention introduces the idea of finite element, which solves the problem of excessive calculation amount of velocity sets at all points in the region in the actual prediction calculation work.

Concrete technical scheme of the present invention is as follows:

1. Explanation of mathematical terms

Differential inclusion: A general differential equation is defined as follows:

是完全由x决定的，则它的变化模型可由如下微分方程表示：Suppose x is a set of n position points x=(x ₁ ,x ₂ ,x ₃ ,…,x _n ), then the rate of change (speed) of x

is completely determined by x, then its variation model can be expressed by the following differential equation:

Adding the initial condition x(t ₀ )=x ₀ , we can find out the different states of point x changing with time, and establish a prediction model of point x.

其中u(·)是控制方程，U是一系列控制方程的集合，点x的变化发展不仅依赖于x本身，也依赖于控制因子，u＝(u₁,u₂,…,u_m)。通过调整控制方程，可以让模型朝着我们预期的方向发展。公式(3)写成微分包含的形式如下：The control system theory introduces control factors on this basis, so that the change track of point x can be controlled manually, and the control model becomes

Where u(·) is the control equation, U is a set of a series of control equations, the change and development of point x not only depends on x itself, but also depends on the control factors, u=(u ₁ ,u ₂ ,…,u _m ). By tweaking the governing equations, we can steer the model in the direction we want. Formula (3) is written in the form of differential inclusion as follows:

Where F(x) is the set of all possible rates of change (speeds) of x, written as follows:

是在点x处受约束u限制下的变化速度，而这一速度向量只是所有可容许速度向量之一。总而言之，微分包含的解是一组集合的映射，利用这一点，本发明将微分包含引入火灾蔓延模型，模拟火点在同一时刻不同方向上按一定速度蔓延的轨迹。here

is the velocity of change limited by constraint u at point x, and this velocity vector is only one of all admissible velocity vectors. In a word, the solution of the differential inclusion is a set of mappings. Taking advantage of this, the present invention introduces the differential inclusion into the fire spread model to simulate the trajectory of the fire point spreading at a certain speed in different directions at the same time.

Finite element method: The core idea of finite element method is "numerical approximation" and "discretization". The continuous complex area is discretized into several simple basic units, and the solutions on the entire continuous complex area are approximated by combining the solutions on the finite number of discrete simple areas. The present invention is based on the idea of finite elements, triangulates the prediction area, divides the entire area into a plurality of small triangular areas with convex set properties, and uses the velocity sets at a small number of discrete representative points to predict all Velocity set at the point.

Triangulation: Triangulation must meet two principles. One is that the small triangles obtained by division do not overlap each other; the other is that no new points are generated after the area is divided. One of the important properties of triangulation is that it is adjacent in the area. The two small triangles have at most one common edge and at least one common vertex, and each small triangle divided is a convex set.

Minkowski addition: A method for summing two sets.

2. Preparatory stage

2.1 Generating the region set Ω of the habitat

Select the habitat to be predicted, obtain the surface satellite remote sensing data of the area from the NASA official website, and extract the longitude and latitude position information of the habitat. Rasterize the location information of the habitat to generate a region set Ω in two-dimensional space.

2.2 Generate forest fire reachable set R

Obtain habitat-related satellite data such as surface temperature data, surface diurnal temperature difference data, surface sulfide data, surface vegetation data, and surface infrared remote sensing image data from my country's National Meteorological Administration and domestic Fengyun series satellites. For the extracted regional meteorological data, data mining and data filtering are carried out to obtain the fire point data in the region. The main basis for judging that a certain location is a fire point is: the concentration of sulfide in this place is high, the surface temperature is high, and the temperature difference between day and night is less than a certain Threshold, and finally calculate the fire area of the habitat in a certain period of time as the reachable set R in the project of the present invention.

3. Pretreatment of Habitat Areas

3.1 Initialize the habitat area

点x在各个方向上有无数个速度向量构成它对应的速度集F(x)。区域Ω中有无数个位置点x，为了优化算法本发明选择区域Ω中有代表的离散点对应的速度集来近似求解整个区域的速度集，本发明在栖息地区域内选择n个离散点，点集为x＝(x₁,x₂,x₃,…,x_n),在此，定义点x_k处的速度集为F(x_k),1≤k≤n，在计算机程序中运行结果如图1所示。Obtain the set of regions for the selected habitat according to step 2. In order to predict the diffusion trajectory of the fire point, the present invention assumes that each position point x has its own velocity set. According to the definition of differential inclusion:

A point x has an infinite number of velocity vectors in each direction to form its corresponding velocity set F(x). There are countless position points x in the area Ω. In order to optimize the algorithm, the present invention selects the velocity set corresponding to the representative discrete point in the area Ω to approximate the velocity set of the entire area. The present invention selects n discrete points in the habitat area, point The set is x=(x ₁ ,x ₂ ,x ₃ ,…,x _n ), here, the velocity set at point x _k is defined as F(x _k ), 1≤k≤n, the result of running in the computer program As shown in Figure 1.

3.2 Triangulating habitat areas

Select m discrete representative points x _i in the habitat area, 1≤i≤m, m≤n/10. Use these m discrete points to triangulate the habitat area, and get p small triangular areas d, the set of small triangular areas is

d=(d ₁ ,d ₂ ,d ₃ ,...,d _p ),

The result of running the computer program is shown in Figure 2.

After triangulation, the small triangles in the area do not overlap each other, and no new points are added, and adjacent small triangles directly have at most one common edge. Since each small triangle is convex, with the nature of a convex set, a limited number of discrete location points in the habitat area can be used to predict all points in the entire habitat area, thereby simplifying the algorithm and speeding up the calculation.

3.3 Find the convex combination coefficient

The main realization of the present invention is to simulate the dynamic spread model of forest fire based on some mathematical principles. To simulate the dynamic process of fire spread, it is necessary to obtain the speed set of the fire point in all directions. Therefore, the strategy that the present invention takes is to carry out convex combination with the initially selected limited position points in the region to predict the velocity set at the fire point where the position can change dynamically, and the simultaneous equations solve the convex combination coefficient, and then can use the solution to get The convex combination coefficients of predict the velocity set at the fire point within the region. The specific mathematical operation process is as follows:

After the habitat is triangulated, it is necessary to judge which small triangle the fire point X belongs to. For example, to judge whether X is in the small triangle ΔABC, you can use the equal area judgment method, as shown in the figure below, if the point X is outside the small triangle ΔABC , then the sum of the area of the triangle formed by point X and points A, B, and C is greater than the area of the triangle formed by points A, B, and C, as shown in Figure 3.

Establish the corresponding inequality, such as formula (6)

S(ΔABC)<S(ΔABX)+S(ΔACX)+S(ΔBCX) (6)

If point X is inside the small triangle ΔABC, then the sum of the area of the triangle formed by point X and points A, B, and C is equal to the area of the triangle formed by points A, B, and C, and the corresponding equation is established, as follows As shown in (7):

S(ΔABC)=S(ΔABX)+S(ΔACX)+S(ΔBCX) (7)

If the equation is established, it is determined that the small triangle where the fire point X is located is ΔABC, and the convex combination of the three vertices A, B, and C of the small triangle is used to represent the fire point X, and because the sum of the coefficients of the convex combination is 1, the simultaneous equations can be established Group to find convex combination coefficients.

The coordinates of vertices A, B, and C of the small triangle are (x _a , y _a ), (x _b , y _b ), (x _c , y _c ), and the coordinates of point X are (x, y). Simultaneously the following equations:

Construct an augmented matrix, such as formula (9)

Find the unique solution of the augmented matrix, and obtain the convex combination coefficients α ₁ , α ₂ , α ₃

3.4 Convex combination predicts velocity sets of other fire points in the region

In step 3.3, it has been solved that the fire point X is in the triangle ΔABC, and the convex combination coefficient of the point X represented by the convex combination of the three vertices A, B, and C of the small triangle has been obtained. The velocity set F(X) corresponding to the fire point X is a set of functions related to X, so the velocity set corresponding to the point X can also use the velocity sets F(A), F corresponding to the three vertices A, B, and C of the small triangle (B), F(C) carry out convex combination prediction, and the predicted value is:

By analogy, the velocity sets at other fire points in the habitat can also be predicted by convex combination with the velocity sets at the three vertices of a small triangle in the habitat. After obtaining the velocity set at each fire point plus the time factor, the fire spread trajectory can be calculated.

4. Based on Minkowski addition, simulate the dynamic change trajectory of the disaster-affected area

The ultimate purpose of the present invention is to simulate the dynamic spread model of the affected area. The discrete model of the velocity set in the entire area of the habitat is established by the above three steps. A process of dynamic change over time.

4.1 Calculate the velocity set at the fire point in the region based on convex combination

由此估计栖息地区域内每个火点处的速度集。先求出这些火点分别属于栖息地区域Ω内的哪个小三角形，然后用该火点隶属的小三角形的顶点处的速度集进行凸组合估算出该火点的速度集，有了森林火灾可达集内各个火点的速度集，就可以计算这些火点经过了时间t之后所到达的新区域了，即求出新的森林火灾可达集，从而建立起该栖息地的火灾动态蔓延模型。This method can be used to calculate the velocity set convex combination in formula (8):

From this the set of velocities at each fire point within the habitat area is estimated. First find out which small triangle these fire points belong to in the habitat area Ω, and then use the velocity set at the apex of the small triangle to which the fire point belongs to perform a convex combination to estimate the speed set of the fire point. The velocity set of each fire point in the reach set can calculate the new area that these fire points reach after the time t, that is, find a new reachable set of forest fires, so as to establish a fire dynamic spread model in this habitat .

The specific implementation is as follows:

(1) The initial set of forest fires. The data in the set is the boundary latitude and longitude information of a certain fire range in the habitat at time 0 (set the observation start time as time 0). To calculate the diffusion trajectory of the boundary of the disaster area, it is necessary to calculate the velocity sets of all points on the boundary, but there are infinitely many position points and infinitely many velocity sets on the boundary. The method that the present invention takes here is to represent the velocity set of each point on the line segment with the velocity set at the endpoint of each line segment on the forest fire initial set boundary, because the same line segment belongs to a certain small triangle in the habitat. All points are linearly related, so only the fire points at the endpoints can be selected as representative points. The specific implementation is as follows: bring the end points of the line segment on the boundary of the forest fire reachable set into the velocity set discrete model established in step 1, step 2, and step 3, and use the velocity set corresponding to the finite number of discrete points on the area Ω to perform convex combination. Represents the set of velocities at corresponding points between the two endpoints of a line segment on the boundary of the disaster area.

(2) Because the line segments on the boundary of the disaster-affected area are not necessarily included in a certain small triangle after the area Ω triangulation. There is one of the following situations: the two endpoints of a line segment on the boundary of the disaster area belong to different small triangles in the area Ω, and at this time, the convex combination of three vertices of the same small triangle cannot be used to represent the two endpoints on the boundary of the disaster area. All velocities of a line segment. In this case, it is necessary to divide the line segment, and only intercept the part completely belonging to the same small triangle as a new line segment, and use the three vertices of the small triangle to perform convex combination to predict the velocity set at the two ends of the new line segment. By analogy, the velocity set of each fire point on the boundary of the forest fire area is obtained.

4.2 Calculate the dynamic change process of the disaster-affected area over time

The above steps have already calculated the velocity set of each fire point on the initial reachable R(0) boundary of the forest fire, introduced the time factor kt, and used Minkowskiadditoin to calculate the motion area of each edge on the polygon boundary of the initial set of forest fires, and calculated all edges After the motion area after time t, these motion areas are combined to obtain the new range of forest fire spreading after time t, that is, the new reachable set R(t).

The present invention also provides a device, including a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the computer program, the steps of the above method are realized.

The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make many modifications and changes according to the concept of the present invention without creative efforts. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art shall be within the scope of protection defined by the claims.

Claims (8)

1. A method for dynamic simulation of a pollutant diffusion region, characterized in that it comprises: Step 1. Select the habitat to be estimated, obtain the satellite data of the habitat area, and construct the initial sample area set based on the satellite data; Step 2. Construct the forest fire initial reachable set R(0) based on the historical fire information of the habitat; Step 3. Based on the finite element idea, select a limited number of discrete representative points to triangulate the habitat area, and divide a number of small triangular areas with convex set properties; establish three vertices on the small triangular area and the inner The point coordinate convex combination formula of all points is used to obtain the convex combination coefficient; Step 4. Using the convex combination coefficient obtained in step 3, use the velocity sets at the three vertices on the small triangle area to predict the velocity sets at all points in the small triangle area; thus use the velocity sets at the finite discrete representative points Predict the velocity set at all points in the entire area; so far, the differential inclusion spread discrete model has been established; Step 5. Adding the time factor to estimate the dynamic spread range of the forest fire reachable set in the habitat area over time. 2. A method for dynamically simulating a pollutant diffusion area as claimed in claim 1, wherein the satellite data of the habitat area at least includes habitat latitude and longitude data, surface temperature data, surface temperature difference between day and night data, surface sulfide data, surface vegetation data, and surface infrared remote sensing image data. 3. A kind of method for dynamic simulation of pollutant diffusion area as claimed in claim 1, characterized in that, said step 2 constructs a forest fire reachable set based on the historical fire information of the habitat, specifically: the sulfidation of the habitat If the concentration of pollutants is high, the surface temperature is high and the temperature difference between day and night is less than a certain threshold, the fire area of the habitat in a certain period of time is finally calculated as the reachable set. 4. A kind of dynamic simulation method of pollutant diffusion area as claimed in claim 1, it is characterized in that, described step 3 is based on the finite element idea, triangulates the whole habitat area with a limited number of discrete representative points, specifically : Initialize the habitat area, select n discrete points in the habitat area, the point set is x=(x ₁ ,x ₂ ,x ₃ ,...,x _n ), here, the velocity set at point x _k is defined as F(x _k ), 1≤k≤n; then select m discrete representative points x _i , 1≤i≤m, m≤n, use these m discrete representative points to triangulate the habitat area, and get p small triangular area d, the set of small triangular areas is d=(d ₁ , d ₂ , d ₃ , . . . , d _p ). 5. a kind of pollutant diffusion area dynamic simulation method as claimed in claim 1, is characterized in that, described step 4 utilizes the velocity set at finite number of discrete representative points to predict the velocity set at each point in the whole habitat area, A differential inclusion spread discrete model is established; the differential inclusion model estimates the velocity of each location point in each direction in the entire habitat area, specifically: select the most effective m discrete representative points x _i in the habitat area, 1≤ i≤m, m≤n/10, use these m discrete representative points to triangulate the entire area, and obtain p small triangular areas d with convex set properties, the set of small triangular areas is d=(d ₁ , d ₂ , d ₃ ,..., d _p ), use the vertices on these small triangular areas and all the discrete points in the small triangular area to use the simultaneous point coordinate convex combination formula, x _i is the point in the qth small triangular area , x _q1 , x _q2 , x _q3 are the three vertices on the small triangle area, the specific formula is as follows: α ₁ *x _q1 +α ₂ *x _q2 +α ₃ *x _q3 = x _i (1) Where: x _i ∈ x, x _i in d _q , 1≤i≤n, 1≤q≤p; Calculate the convex combination coefficients α ₁ , α ₂ , α ₃ ; since the discrete points in the small triangle are represented by the convex combination of three vertices on the small triangle area, by analogy, the velocity set at the discrete points in the small triangle can also be Convex combined predictive representation with the velocity sets at the three vertices on the small triangular area; the formula is:

Among them, f(x _q1 ), f(x _q2 ), f(x _q3 ) represent the known velocity sets at the three vertices of x _q1 , x _q2 , and x _q3 respectively;

Represents the approximate velocity set at the point x _i obtained by reconstructing the velocity sets f(x _q1 ), f(x _q2 ), and f(x _q3 ) at the three vertices. 6. A method for dynamically simulating a pollutant diffusion area as claimed in claim 1, wherein said step 5 adds a time factor to estimate the range of the forest fire reachable set dynamically spreading over time in the habitat area, specifically: : Find the velocity set of each fire point on the initial reachable boundary of the forest fire, introduce the time factor, calculate the trajectory of each fire point in different directions on the initial set boundary of the forest fire, and find out all the fire points after time t After the motion trajectories, these motion trajectories are summed to obtain the new range that the forest fire spreads after the elapse of time t, that is, the new reachable set. 7. A pollutant diffusion area dynamic simulation system, characterized in that it comprises: The area set acquisition module is used to select the habitat to be estimated, obtain the satellite data of the habitat area, and construct the initial sample area set based on the satellite data; The reachable set acquisition module is used to construct the forest fire reachable set based on the historical fire information of the habitat; The regional division module, based on the finite element idea, selects a limited number of discrete representative points to triangulate the entire habitat area, divides the entire area into multiple small triangular areas, and each small triangular area is a convex set. These discrete representative points are the vertices on these small triangular areas; The fire speed prediction module uses the three vertices on the small triangular area and all the points in the small triangular area to combine the convex combination formula to obtain the convex combination coefficient, because the velocity set can be expressed as a function related to discrete points, and so on , using the obtained convex combination coefficient, use the velocity sets at the three vertices on the small triangular area to perform convex combination to predict the velocity sets at all discrete points in the small triangular area; establish a differential inclusion spread discrete model to predict the fire point at Velocity sets in all directions; The fire spread estimation module, by adding the time factor, uses the Ekowski addition method to estimate the dynamic spread range of the forest fire reachable set in the habitat area over time. 8. A dynamic simulation device for a pollutant diffusion area, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program When realizing the steps of the method as described in any one of claims 1 to 6.

Images